Dr. Hilty is associate professor of clinical psychiatry and director of the Mood Disorders and Health Services Research Program at the University of California, Davis in Sacramento.

Dr. Nesbitt is professor of family and community medicine and associate dean of the Department of Regional Outreach and Telehealth at the University of California, Davis.

Ms. Marks is research assistant in the Department of Psychiatry at the University of California, Davis.

Dr. Callahan is professor of family and community medicine and associate director of the Center for Health Services for Research in Primary Care at the University of California, Davis.

Acknowledgments: The authors report no financial, academic, or other support of this work. 



Telepsychiatry offers enormous opportunities for clinical care, education, research, and administration in the field of medicine. This article reviews the telepsychiatric literature—specifically videoconferencing—to evaluate effects of telepsychiatry on the doctor-patient relationship. A review was conducted of the MEDLINE, PsycINFO, Embase, Science Citation Index, Social Sciences Citation Index, and Telemedicine Information Exchange databases (July 1965–June 2001). Preliminary studies report no major impediments to the development of the doctor-patient relationship in terms of communication and satisfaction. Evaluation is currently limited by uncontrolled trials that measure ill-defined terms or concepts. Many personal, professional, technical, psychological, and social factors influence how telepsychiatry is experienced in the doctor-patient relationship. More prospective, randomized, quantitative, and qualitative research is needed.



Telepsychiatry offers enormous opportunities for clinical care, education, research, and administration in the field of medicine. One significant advantage of telepsychiatry has been improvement of access to psychiatric care in urban,1 suburban,2 and rural3,4 areas, often by providing academic specialists to areas with provider shortages.5 Videoconferencing is live, interactive, audio/video communication or television. Typical equipment for videoconferencing includes Pentium computers with 128–512 megabytes of random access memory (RAM), cameras with local and remote pan-tilt-zoom control, color monitors, and a coder-decoder (CODEC) for converting the audio and visual information into the binary code for transmission. Dial-up integrated service digital network (ISDN) or T1 lines are rented, with transmission at speeds of 128–768 kilobits per second (KBS).

The assessment of telepsychiatry’s impact on the doctor-patient relationship is complicated by the many types of patients, settings, and practice styles for which it is employed. Patient types vary by mental disorder, age, culture, and setting. Sites of service include primary care and mental health clinics, medical and psychiatric emergency rooms, nursing homes, shelters, hospices, schools, forensic facilities, the battlefront, public health, and academic centers.6 One-time evaluation by consultation-liaison or private psychiatric practice, ongoing evaluation, and psychiatric management services have been provided.

This article discusses the effect of telepsychiatry on the doctor-patient relationship, including data, concepts, and theories about how communication, satisfaction, and other factors affect development of the relationship. Problems with reports in the literature will be discussed to suggest improvements in quantitative and qualitative research.


A comprehensive review of the telepsychiatric literature was conducted in the MEDLINE, PsycINFO, Embase, Science Citation Index, Social Sciences Citation Index, and Telemedicine Information Exchange databases, from July 1965–June 2001, using the keywords telepsychiatry, telemedicine, videoconferencing, doctor, physician, patient, relationship, communication, verbal, nonverbal, and satisfaction. Articles were selected for review if they mentioned such aspects in the title or abstract. From the articles selected, salient referenced articles were also obtained and reviewed.



A host of factors affect perception of the telemedicine visit and communication by participants. Disclosure is affected by the presence of others in the room, belief of being videotaped, and stigma. In addition, if participants have never used telemedicine before, they may feel anxious, distracted due to the equipment, and self-conscious when seeing themselves on the screen.

A critical variable in communication is telemedicine’s ability to simulate real-time experiences in terms of image and interaction. The speed of transmission has a profound affect on audio and video quality. Most services transmit 128–768 KBS. Terrestrial transmission at 128 KBS provides a good picture with a signal delay of 0.3 seconds, whereas 768 KBS may have almost no delay. Satellite transmission involves a delay of 0.5–1.0 seconds, as seen on worldwide broadcasts. Equipment problems (poor audio or video, lack of camera control, disconnection) affect communication, but are rarely reported. The most important issue is having technology adequate for the clinical task at hand and putting alternative plans in place if a limitation exists (eg, a primary care physician evaluates a tremor which cannot be seen).

Signal delay is caused by time to process (digitalize) and transmit the signal.7 Conversation is the main mode of communication in therapy and is essential for the building of rapport. With slight delays in signal (eg, 0.3 seconds), a turn-taking conversation occurs rather than the free-flowing conversation indicative of a high-rapport interaction.8 In addition, collisions take place as parties speak at the same time and perceive the other as interrupting. No differences with the development of rapport were found in a small cohort comparing signal delays of 0, 0.3, and 1.0 seconds, as measured on self-report questionnaires.7

Another concept that bears on communication is presence, as recently reviewed by Turner.9 Kim and colleagues10 defined presence as “… the fact or condition of being at the specified or understood place.”  They postulate that physical and virtual environments affect presence: In a physical environment, informational cues are incorporated without conscious awareness (eg, a patient is seen walking in a reticent way). Participants need to be aware that the virtual environment created by telemedicine is not the same as a regular physical environment as cues are missing.10 Currently, it is assumed that the videoconferencing provides “enough” of the physical environment for good decision-making, but differences may occur between the environments. Cukor and colleagues11 reported that telepsychiatry facilitates a “social presence” that permits participants to share a virtual space, get to know one another, and discuss complex issues, even when low-cost systems are used.

A few studies evaluated the effect of telemedicine on nonverbal communication, which fundamentally establishes mutual connections and understanding. Examples are eye contact, gestures, posture, fidgeting, nods, grins, smiles, frowns, and lip-reading.12 Decreased ability to detect nonverbal cues has been reported during videoconferencing of patient interviews.13 This has been previously described as the “cuelessness” phenomenon.14 A task-oriented focus with a depersonalized content may occur.15 On a spectrum of detecting cues, telepsychiatry may be the cross between the telephone and in-person communication.16

Ball and colleagues17 compared communication behaviors between six physicians and six patients using in-person, telephone, hands-free telephone, and low-cost telepsychiatry (KBS not specified) modes. Higher levels (75% of the time) of mutual gaze were recorded for the visual modes (in-person and telepsychiatry), which was higher than usual interpersonal interactions (~50%).18 Self-report questionnaires by patients revealed lower levels of anxiety in the visual modes, but some sense of having been misunderstood. Physicians reported increased anxiety, but a better understanding of patients with the visual modes.

The nature of what is exchanged also varies depending on the mode used. Information exchange takes place primarily on an audio channel rather than a video channel.11,19,20 Participants respond in a “conservative” or “stilted” way when audio delay occurs with videoconferencing,12 resulting in more interruptions of the interview than with video disruption.21 For the same conversation, in-person takes less time than telephone, which in turn, takes less time than videoconferencing.19,20

Some worry that telemedicine may adversely affect the development of a positive therapeutic alliance because of potential limitations: They suggest that a preexistent relationship is necessary to minimize the potential negative effects of telepsychiatry. In a study using the California Psychotherapy Alliance Scale, 41 patients and psychiatrists were randomly assigned to a diagnostic interview by either in-person or telepsychiatry (KBS not specified).22 Manchanda and colleagues23 used the Working Alliance Inventory scale to assess therapeutic alliance during videoconferencing at 128 KBS. Neither significantly interfered with the development of an alliance.

Overall, telepsychiatry appears to have advantages and disadvantages with regard to communication (Table 1).


Satisfaction With Telepsychiatry

Standardized assessment of satisfaction is a key determinant of whether telepsychiatry will be integrated into medical practice. A review of the telehealth studies on satisfaction revealed many limitations: few randomized trials, small samples, use of short (one or two items) quantitative questionnaires, lack of standardization in terminology, and confounding variables.24 Another systematic review of 32 satisfaction telemedicine studies, including nine of telepsychiatry, cited variable designs and unspecified criteria for patient selection.25

When asked what it might be like to have a telemedicine visit, some patients expected a less satisfactory physician-patient interaction than in a traditional physician-patient encounter.26 Other patients preferred telepsychiatry for their care. McLaren and colleagues13 reported that increased interpersonal distance by telepsychiatry enhanced communication. Impressions of patients for whom telepsychiatry is indicated, contraindicated, or the treatment of choice, are listed in Table 2. Further research is required on this subject.


Expectations for telepsychiatry service come into play for both patients and telepsychiatrists. For example, a patient may be expecting psychotherapy, while the primary care provider may prefer medication management. Both, one, or neither of these expectations may be accommodated by the telepsychiatrist if a consultation-liaison service model is being used to “train” the primary care physician.2 Telepsychiatry may not fit into a physician’s idea of practice, despite an understanding of advantages to patients and systems.27 Finally, patients and psychiatrists may experience a “break” in the doctor-patient relationship due to telepsychiatry,27 as experienced by the fields of internal medicine, pathology, and surgery.28

Patient Satisfaction Studies

Patient satisfaction studies are summarized in Table 3.29-43 Key predictors of satisfaction with telepsychiatry have not been clearly delineated, though transmission speed is an important variable. Current predictors include type of care provided,16 frame speed (eg, 30 frames/second is television quality),44 demographic factors (eg, age, gender, or ethnicity),2 state- and trait-dependent factors (eg, acute depression versus depression in remission),2 cost, reduced time to travel,33,40 reduced waiting time, and satisfaction with and availability of local services. Dimensions like technical quality of care, financing, physical environment, continuity, humanity, efficacy, competence, empathy, trust, cooperation, safety, and autonomy are assessed less frequently assessed.24


Clinicians have wondered if satisfaction differs for patients seen for mental health versus other specialty problems. A prospective, open study by Callahan and colleagues compared patient satisfaction for psychiatry (N=31) versus other specialty services via telemedicine (N=51).33 There was no significant difference on patient rating between the groups in terms of their ability to speak freely when using telemedicine, their preference for using telemedicine on subsequent visits, and their experience of the telemedicine physician. A larger study (N=221) reported similar results from a prison population.45

Several open studies have assessed patient satisfaction with telepsychiatry in adult, child, adolescent, and geriatric patients.32,46 Generally, satisfaction is high and problems that occur do not appear to have a detrimental effect on the relationship.

A randomized trial of children and adolescents reported similar findings.15 Geriatric satisfaction was similar to that of younger adults.42,44 Table 3 lists these and other studies which have informally evaluated satisfaction with small samples and/or in combination with a study on reliability, outcomes, or cost.6,35,39,47,48

Many studies have compared telepsychiatry to in-person care, but relatively few have assessed satisfaction in a detailed, prospective fashion. Dongier and colleagues35 found no difference in patient satisfaction between telemedicine (N=50) and usual care (N=35) for adults and children in a comparison with in-person care. In a longitudinal study, Hilty and colleagues2 offered primary care patients the opportunity to select their preference (telepsychiatry or in-person) for evaluation and follow-up care (if applicable). The groups were controlled for the clinic attended, length of waiting time, presence of insurance, demographic information, and diagnoses. More patients chose in-person care for initial (71%) and follow-up (65%) appointments. The appointment adherence rate was similar between the groups. No studies have collected information on patients who refused to participate or who dropped out without completing standard questionnaires—a key deficit in the data.

Studies of Provider Satisfaction

It is unclear whether variables that have been shown or proposed to affect patient satisfaction are relevant to physician satisfaction, and additional variables not mentioned here are likely to affect satisfaction. Primary care satisfaction has been evaluated in a number of ways and provider satisfaction studies are summarized in Table 4. Hilty and colleagues48 reported that primary care physician satisfaction, as measured by self-report questionnaires, was high initially with psychiatric consultation-liaison service and improved compared to baseline after two or more consultations over a 1-year period; this was statistically higher for rural than suburban/urban physicians. This finding was attributed to limited access to psychiatric care for patients in rural areas, who preferred consultation rather than management. On the other hand, some providers may prefer to refer patients to a psychiatrist rather than manage the case themselves.


Kopel and colleagues39 reported rural clinician (n=101) and psychiatrist satisfaction (n=136). When including “good” or “excellent” on a 4-point Likert scale, the physicians rated the quality highly and reported overall satisfaction of 89% to 100%; over 97% felt consultation was as good as in-person care. Finally, primary care provider satisfaction may also depend on the development of a relationship with the consulting psychiatrist through a meeting, previous phone calls, requests for teaching, and referrals of “VIP” patients outside regular channels.

Psychiatrist satisfaction has only rarely been formally evaluated and shows mixed results. Concerns surfaced about technical problems (eg, unclear picture, video freeze), decreased ease with the process, decreased ability to express oneself, and poorer quality of the interpersonal relationship.6,15,35

In a study of 200 telepsychiatric consultations, telepsychiatrists rated their overall satisfaction with telepsychiatry at 6.6 and the quality of the audiotransmission and videotransmission at 6.8 on a scale of 1 (poor) to 8 (excellent).6 Similar overall results, including improvement over time,49 were reported by others. Kopel and colleagues39 reported child psychiatrist satisfaction (N=136) on a 4-point Likert scale, with a mean satisfaction of 2.1. Finally, findings were similar for a study by Elford and colleagues36 which included adolescents. Other rudimentary issues likely to impact satisfaction are ability to use one’s regular screening questionnaires, obtaining adequate information from others, easy access to other providers (if any), and collaboration or cooperation with telecoordinators (since they may not be directly hired by the psychiatrist).

Dermatologists, otolaryngologists, pathologists, and psychiatrists reported the following positive aspects: less traveling, more time for other work, less need to travel in poor weather, new opportunities, and an increased sense of professional security.50


Telepsychiatry offers enormous opportunities for clinical care, education, research, and administration in the field of medicine. Each new technology offers advantages and disadvantages to what is currently offered. Based on preliminary studies, relationship building appears possible via telepsychiatry, which offers clear advantages compared to telephone consultation or no care at all. Still, telepsychiatry has disadvantages compared to in-person care (eg, limiting nonverbal communication), although these drawback are not fully understood. Telepsychiatry offers an opportunity to evaluate new options for care and new models for providing care, and to reflect on nontelepsychiatric practice (since reaction to telepsychiatry is partly due to how the individual physician typically practices). It is also an opportunity for the field to collaborate with other fields that intersect with medical practice regarding telepsychiatry.

It is clear that more rigorous research is indicated for telepsychiatry. Quantitative measurement of satisfaction and its effects on communication and the clinical relationship also need to be more specific to a particular variable. Researchers need to better define terms and concepts, use standardized methods, and incorporate randomized designs. Telepsychiatric satisfaction may change over time such that baseline and follow-up assessments are indicated. Qualitative measurement will help identify hypotheses and theories that can be further assessed. PP


1.    Straker N, Mostyn P, Marshall C. The use of two-way TV in bringing mental health services to the inner city. Am J Psychiatry. 1976;133:1202-1205.
2.    Hilty DM, Servis ME, Nesbitt TS, Hales RE. The use of telemedicine to provide consultation-liaison service to the primary care setting. Psychiatric Ann. 1999;29:421-427.
3.    Brown FW. Rural telepsychiatry. Psychiatr Serv. 1998;49:963-964.
4.    Preston J, Brown FW, Hartley B. Using telemedicine to improve health care in distant areas. Hosp Community Psychiatry. 1992;43:25-32.
5.    Nesbitt TS, Hilty DM, Kuenneth T, et al. Development of a telemedicine program: a review of 1,000 videoconferencing consultations. Western J Med. 2000;173:169-174.
6.    Hilty DM, Nesbitt TS, Kuenneth TA, et al. Development of a successful telepsychiatry program: a look at the first 200 consultations at UC Davis. Psychiatr Serv. 2002. In press.
7.    Manning TR, Goetz ET, Street RL. Signal delay effects on rapport in telepsychiatry. Cyber Psychology & Behavior. 2000;3:119-127.
8.    Tickle-Degnen L, Rosenthal R. The nature of rapport and its nonverbal correlates. Psychological Inquiry 1990;1:285-293.
9.    Turner JW. Telepsychiatry as a case study of presence: do you know what you are missing? www.ascusc.org/jcmc/vol6/issue4/turner.html. Accessed August 2002.
10.    Kim T, Biocca F. Telepresence via television: two dimensions of telepresence may have different connections to memory and persuasion. Available at:www.ascusc.org/jcmc/vol3/issue2/kim.html. Accessed August 2002.
11.    Cukor P, Baer L, Willis BS, et al. Use of videophones and low-cost standard telephone lines to provide a social presence in telepsychiatry. Telemed J. 1998;4:313-321.
12.    Fussell SR, Benimoff NI. Social and cognitive processes in interpersonal communication: implications for advanced telecommunications technologies. Hum Factors. 1995;37:228-250.
13.    McLaren P, Ball CJ, Summerfield AB, Watson JP, Lipsedge M. An evaluation of the use of interactive television in an acute psychiatric service. J Telemed Telecare. 1995;1:79-85.
14.    Rutter DR. Looking and Seeing: the Role of Visual Communication in Social Interaction. New York City: Wiley, Chichester; 1984.
15.    Elford R, White H, Bowering R, et al. A randomized, controlled trial of child psychiatric assessments conducted using videoconferencing. J Telemed Telecare. 2000;6:73-82.
16.    Hilty DM, Luo JS, Morache C, et al. Telepsychiatry: an overview for psychiatrists. CNS Drugs. 2002;16(8):527-548.
17.    Ball CJ, McLaren PM, Summerfield AB, Lipsedge MS, Watson JP. A comparison of communication modes in adult psychiatry. J Telemed Telecare. 1995;1:22-26.
18.    Argyle M. Bodily Communication. London, England:?Methuen; 1975.
19.    Ochsman RB, Chapanis A. The effects of 10 communication modes on the behavior of teams during co-operative problem-solving. International Journal of Man-Machine Studies. 1974;6:579-619.
20.    O’Malley C, Langston S, Anderson A, et al. Comparison of face-to-face and video-mediated interaction. Interacting with Computers. 1996;8:177-192.
21.    McLaren P, Mohammedali A, Riley A, Gaughran F. Integrating interactive television-based psychiatric consultation into an urban community mental health service. J Telemed Telecare. 1999;5:100-102.
22.    Stevens A, Doidge N, Goldbloom D, Voore P, Farewell J. Pilot study of televideo psychiatric assessments in an underserviced community. Am J Psychiatry. 1999;156:783-785.
23.    Manchanda M, McLaren P. Cognitive behaviour therapy via interactive video. J Telemed Telecare. 1998;4:53-55.
24.    Williams TL, May CR, Esmail A. Limitations of patient satisfaction studies in telehealth care: a systematic review of the literature. Telemedicine Journal and e-Health. 2001;7:293-316.
25.    Mair F WP. Systematic review of studies of patient satisfaction with telemedicine. British Medical J. 2000;320:1517-1520.
26.    Brick JE, Bachshur RL, Brick JF, et al. Public knowledge, perception, and expressed choice of telemedicine in West Virginia. Telemed J. 1997;3:159-172.
27.    May C, Gask L, Atkinson T, Ellis N, Mair F, Esmail A. Resisting and promoting new technologies in clinical practice: the case of telepsychiatry. Soc Sci Med. 2001;52:1889-1901.
28.    Jewson N. The disappearance of the sick man from medical cosmology 1770-1870. Sociology. 1976;10:225-244.
29.    Baer L, Cukor P, Jenike MA, Leahy L, O’Laughlen J, Coyle JT. Pilot studies of telemedicine for patients with obsessive-compulsive disorder. Am J Psychiatry. 1995;152:1383-1385.
30.    Baigent MF, Lloyd CJ, Kavanaough SJ. Telepsychiatry: “tele” yes, but what about the psychiatry? J Telemed Telecare. 1997;3:3-5.
31.    Blackmon LA, Kaak HO, Ranseen J. Consumer satisfaction with telemedicine child psychiatry consultation in Kentucky. Psychiatr Serv. 1997;48:1464-1466.
32.    Bratton RL, Cody C. Telemedicine applications in primary care: a geriatric patient pilot project. Mayo Clin Proc. 2000;75:365-368.
33.    Callahan EJ, Hilty DM, Nesbitt TS. Patient satisfaction with telemedicine consultation in primary care: a comparison of ratings of medical and mental health applications. Telemed J. 1998;4:363-369.
34.    Chae YM, Park HJ, Cho JG, Hong GD, Cheon KA. The reliability and acceptability of telemedicine for patients with schizophrenia in Korea. J Telemed Telecare. 2000;6:83-90.
35.    Dongier M, Tempier R, Lalinec-Michaud M, Meunier D. Telepsychiatry: psychiatric consultation through two-way television. A controlled study. Can J Psychiatry. 1986;31:32-34.
36.    Elford DR, White H, St John K, Maddigan B, Ghandi M, Bowering R. A prospective satisfaction study and cost analysis of a pilot child telepsychiatry service in Newfoundland. J Telemed Telecare. 2001;7:73-81.
37.    Graham MA. Telepsychiatry in Appalachia. American Behavioral Scientist. 1996;39:602-615.
38.    Johnston D, Jones BN, 3rd. Telepsychiatry consultations to a rural nursing facility: a 2-year experience. J Geriatr Psychiatry Neurol. 2001;14:727-725.
39.    Kopel H, Nunn K, Dossetor D. Evaluating satisfaction with a child and adolescent psychological telemedicine outreach service. J Telemed Telecare. 2001;7:35-40.
40.    McCloskey-Armstrong T. Rural psychiatric collaborative care via telemedicine. San Diego, Calif: American Psychiatric Association; 1997:106.
41.    Mielonen ML, Ohinmaa A, Moring J, Isohanni M. The use of videoconferencing for telepsychiatry in Finland. J Telemed Telecare. 1998;4:125-131.
42.    Ruskin PE. Efficacy of telepsychiatry in treatment of depression. Abstract presented at: 18th Annual Meeting of the Veterans Administration Health Service Research and Development Service; March 22-24, 2000; Washington, DC.
43.    Trott P, Blignault TP. Cost evaluation of a telepsychiatry service in northern Queensland. J Telemed Telecare. 1998;4:66-68.
44.    Jones BN, Ruskin PE. Telemedicine and geriatric psychiatry: directions for future research and policy. J Geriatr Psychiatry Neurol. 2001;14:59-62.
    Mekhijan H, Turner JW, Gailun M, et al. Patient satisfaction with telemedicine in a prison environment. J Telemed Telecare. 1999;5:55-61.
46.    Montani C, Billaud N, Tyrrell J, et al. Psychological impact of a remote psychometric consultation with hospitalized elderly people. J Telemed Telecare. 1997;3:140-145.
47.    Clarke PH. A referrer and patient evaluation of a telepsychiatry consultation- liaison service in South Australia. J Telemed Telecare. 1997;3:12-14.
48.    Hilty DM, Nesbitt TS, Kuenneth TA. Development of a successful telepsychiatry program: a look at the first 200 consultations at UC-Davis. Poster presented at: 47th Annual Meeting of the Academy of Psychosomatic Medicine; November 2000; Palm Springs, Calif.
49.    Gelber H. The experience in Victoria with telepsychiatry for the child and adolescent mental health service. J Telemed Telecare. 2001;7:32-34.
50.    Aas IH. Changes in the job situation due to telemedicine. J Telemed Telecare. 2002;8:41-47.

Acknowledgments:The author reports no financial, academic, or other support of this work. 



What are the potential uses for personal digital assistants or handheld computers in the practice of medicine? Pagers and cellular phones are mainstays in the practice of medicine to access and communicate information, but an electronic version of a day planner appears not to be a valuable tool to most physicians. With the increasing hardware capabilities of these handheld devices, more software has been developed for clinical purposes such as drug references, prescription writing, medical record keeping, and charge capture. After reviewing the different medical applications available for the handheld device, most physicians will find many compelling reasons to adopt this new technology.



The practice of medicine has increasingly become more dependent on computer technology. Computer terminals provide access to laboratory results, and personal computers facilitate note writing. The facsimile machine eliminates dependence on couriers and mail for document exchange. Pagers were the first devices that allowed physicians to roam from the office yet still remain available for urgent contact. The cellular phone extends this availability by eliminating the time required to locate a phone. In this progression to increasing dependence on technology, personal digital assistants (PDAs) have quickly become an essential tool in the practice of medicine. A recent article discussed how the use of PDAs may improve patient safety and decrease medication errors.1 The Agency for Healthcare Research and Quality has funded four research projects involving PDAs to prevent medical errors.2 This article focuses on the different ways that a PDA can assist primary care physicians in the care of their patients.

Background and Basics

PDAs first appeared in the 1990s primarily as electronic versions of their paper-based cousins the day planner or personal information manager. Functions of the PDA included a calendar, an address book, a to-do list, and a memo area for notes. All of these features worked similarly to paper-based organizers, but the screen size was small and text entry was cumbersome. Backup of information was quite limited because the cables and software for desktop computer connection often cost more than the device itself. An obvious advantage of the PDA is that the entries are in machine text, which is legible, but the early PDAs were not exchangeable with other devices. Today’s PDAs have improved capabilities such as larger screens and improved input, but what sets these devices apart from paper organizers is their capability to run additional programs or medical applications.

There are many other advantages to using PDAs. A key element involves their form factor—their size varies from a pack of cards to a thick checkbook. For the busy physician moving from the office to the hospital or even between examination rooms, this small form factor provides great portability and function. Almost all PDAs provide some linkage to information on desktop computers. This capability allows synchronization of data on the PDA and on the desktop computer to keep information current. In addition, this synchronization process serves as a crucial backup feature when data are lost.

One common concern is that a “computer illiterate” physician will not be able to use a PDA. In actuality, using a PDA is much simpler than operating a desktop computer. The user interface is quite friendly and intuitive, and no typing skills are really necessary. All functions can be accessed with a push of a button or a tap on the screen. In fact, taps on the screen with the stylus are the primary way that the different functions of the software programs are accessed. To learn how to install programs and maximize productivity, many physicians attend classes that are offered by local medical societies, read books on PDA use (Table 1), or seek information on the Internet (Table 2).


Choosing a Personal Digital Assistant

Choosing a PDA is really a simple matter despite the somewhat daunting variety of models and manufacturers. It is important to note that because technology improves and changes rather quickly. Newer models that may offer additional features or desirable capabilities appear almost every quarter. Waiting for prices to drop is one strategy, but in doing so you will not benefit from the many capabilities that these devices offer, as described in this article. A good plan of action is to consider keeping your PDA for at least a year, and then upgrade only when new features are compelling, such as more memory or wireless capability.


Which PDA to purchase should be determined first on the basis of its operating system (OS). Currently, four main kinds are available: Palm OS–based PDAs, which are at present the most popular worldwide; EPOC OS–based PDAs, which are more popular in Europe; Pocket PC–based PDAs, which are gaining increasing market share; and Linux OS–based PDAs, which have just begun to enter the market. Generally, all PDAs have similar basic capabilities, but what sets them apart are available software, hardware features, and accessories.

Palm OS (www.palmos.com) PDAs are the most popular because they are quite quick and portable. They are generally smaller than their counterparts, noted for an “instant-on” capability, intuitive user interface, and good battery life. The available general and medical software for the Palm OS is the largest of the four operating systems. In addition, many hardware accessories are available such as external memory, modems, keyboards, Bluetooth, and wireless local area network (LAN) adapters. Manufacturers to consider include Palm, IBM, Sony, HandEra, Acer, and Handspring (Table 3). Newer devices based on Palm OS 5 due in the latter half of 2002 will be faster, multitasking, and will have built-in security and Web-browsing capability.


Pocket PC–based PDAs (www.microsoft.com/mobile/pocketpc/default.asp) are considered the powerhouses of the PDA market with more memory and a faster central processing unit. These devices are noted for enterprise-level capability but are notorious for a poor battery life of ~1 day. A significant advantage of devices based on the Pocket PC OS is that exchange of documents with Microsoft Office on the desktop computer is easier. Although the number of software developers for the Pocket PC is significantly smaller than for the Palm OS, this trend is slowly beginning to change. Manufacturers include Toshiba, Compaq, Hewlett-Packard, Casio, Urthere, Audiovox, and NEC.

EPOC OS and Linux OS are both quite capable PDAs and, from a technological standpoint, are worth consideration. However, both of these PDAs lack a significant software developer base. Psion (www.psion.com), one of the primary manufacturers of EPOC OS PDAs, has withdrawn from the consumer market in order to focus on enterprise and vertical markets. Linux OS PDAs by Agenda(www.agendacomputing.de/agenda-e/index-e.htm), Samsung (www.yopy.com), and Sharp (www.myzaurus.com) are quite feature rich but should only be used by those comfortable with the Linux OS. Although the Opensource software community is writing and porting (converting) more software for the Linux OS PDAs, physicians using these devices should be prepared for limited selection and some confusion regarding compatibility of software owing to different installation packages.

General Use

The basic capabilities of PDAs work well for most physicians and can be well adapted for medical purposes. These programs are called read-only memory (ROM)–based programs, because they are permanently located in the PDA’s memory. By comparison, any additional software program installed will be located in random access memory (RAM). This type of memory is not permanent, and information residing in RAM can be erased as needed. The more RAM that a PDA has the more programs, as well as information, can be stored.

An advantage to the calendar program is the ability to set repeating appointments which prevents double booking. For example, patients who are extremely needy and call for appointments frequently, can schedule for regular visits. The address book program is quite versatile in storing not only patient demographic information, but also other important information such as preferred pharmacy or medical record number. The to-do program provides an obvious mechanism to keep track of activities such as returned calls and prescription refills. It has the capacity to record the date of the action completion, which may help with medical record documentation. The memo or note program in the PDA is very useful for entering short notes such as drug information or a brief synopsis of symptoms for diseases.

In the Palm OS PDAs, only short notes can be entered because of the 4-kilobyte file-size limit, which represents ~<1 full page. More extensive text editing is available on Palm OS PDAs with software such as QuickWord (www.cesinc.com) or Wordsmith (www.bluenomad.com). These programs offer conversions between Microsoft Word and the Palm PDA versions of the document. On Pocket PC devices, there is no size limit because Pocket Word is a standard application with linked conversion to documents on the desktop. Documents can be printed with the use of additional infrared or wireless printing software specific to the PDA.

E-mail is an increasingly more popular medium for physicians to communicate with their patients. However, the small screen and relative inability to work with attachments limit E-mail on the PDA. E-mail can be handled either with direct PDA Internet connection such as a modem or through synchronization with an Internet-connected computer. Drawbacks to electronic communication with patients include security and timeliness of response. A policy regarding E-mail communication is advised with clear expectations regarding content, response time, copies to the paper chart, and mechanisms for emergencies.3 For an extensive list of Palm OS E-mail software, visit Handheld Computing magazine’s Web site (www.pdabuzz.com/about) or check PalmGear (www.palmgear.com) and search the links for E-mail software. Although the Pocket PC includes Pocket Outlook for E-mail, PocketGear (www.pocketgear.com) has additional related software that adds more features.

Medical Software

Drug Reference

Drug information at the point of care is one of the most valuable uses of a PDA (Table 4). Programs that supply such information cannot replace resources such as the Physician’s Desk Reference4(www.pdr.net/homepage_template.jsp), but they do provide concise content that is extremely portable. Reviews of programs such as ePocrates qRx (www.epocrates.com) have reached publications such as the Journal of the American Medical Association, elevating their status on par with medical texts.5 A significant advantage is that many of these programs link to other programs that will check for drug-drug interactions or have this capability built in. This capability is important to physicians who prescribe antidepressant medications, in particular owing to the cytochrome P450 enzyme inhibition of selective serotonin reuptake inhibitors. Several years ago, such drug interaction software was not available, and lists of drug information tables stored in memos were the only way to determine potential problems. This mechanism was rather cumbersome and time-consuming, especially for patients who were taking more than four medications. With the use of programs such as MultiCheck, which is available as part of ePocrates qRx, drug interactions are calculated with a few screen taps to select the medications in question and then cross-checked automatically. Additionally, medications can be selectively added or deleted from the list and cross-checked again.


Prescription Writing

Prescription writing is a mainstay of medical practice, and PDAs serve an extremely useful function in this area. Software is available to generate legible prescriptions, which will decrease potential transcription errors at the pharmacy (Table 5). For example, a recent notice from AstraZeneca was sent to physicians warning about reported errors between prescriptions filled for quetiapine (Seroquel) and nefazodone (Serzone), due to the similar trade names. Many programs print the prescription on specialized paper or with specialized printers. In addition, some programs such as Iscribe 5000 (www.iscribe.com) allow the physician to send the prescription electronically to pharmacies via fax from the desktop computer or by wireless via company servers, as in the case of Ephysician (www.ephysician.com). Another advantage of handheld prescription writing is that the software can check for availability of the medication on the health plan formulary. By doing so on the handheld device before printing the prescription, the physician will be alerted when a treatment authorization request must be done. Patients find this capability extremely helpful because it eliminates the need to call their health plan or the physician’s office.


Reference Texts

Psychiatrists rely on reference texts such as the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR)6 or the American Psychiatric Press Textbook of Psychiatry.7 Only a few reference texts for psychiatry are now available in electronic form for review on the PDA (Table 6). There are numerous general and specialty medical texts such as the Merck Manual,8 5 Minute Clinical Consult Series,9 Harrison’s Principles of Internal Medicine Handbook,10 and DeGowin & DeGowin’s Diagnostic Examination11 that are available in Pocket PC and Palm OS formats from Handheldmed(www.handheldmed.com) and PDAMD (www.pdamd.com/vertical/home.xml). The primary drawback to reading these reference materials is the small screen and memory availability in certain handheld devices.


Diagnosis Assistants

The diagnosis of mental disorders is based on matching symptoms to criteria listed in the DSM-IV-TR. For the sake of portability, many people have modified an electronic version of reference texts for the desktop and converted the information for the PDA. DSM-IV-TR diagnosis codes and disease criteria for the PDA are available on the Internet in a variety of software program formats. Good resources for this software on the Palm OS are PocketPsych (www.pocketpsych.com/Resources/dowloads.htm) and Memoware (www.memoware.com). Medical Piloteer (www.medicalpiloteer.com) sells a program called PsychDx that has the DSM-IV-TR criteria summarized and presented in a structured program versus reading a rather static document file. For specific assistance in diagnosis of different mental disorders, the available software is rather limited. Medical Piloteer offers DepressQ and ManiaQ, which are checklist-based programs to assist in the diagnosis of mood disorders for the Palm OS. On the Pocket PC side, there are no specific mental disorder diagnosis programs, but DiagnosisPro from Medtech (www.medicalamazon.com) offers differential diagnoses based on different sign and symptom entries.

Treatment Guides

Available psychiatric treatment guides include the American Psychiatric Association Treatment Guidelines, a comprehensive treatment resource available from Handheldmed. However, it does not lend itself for quick navigation. For primary care physicians who need quick and easy-to-navigate information on the treatment of mental disorders, Compendica (www.compendica.com) has an excellent reference product, but only for the Palm OS. In addition, Medical Piloteer’s PsychRx has summarized treatment recommendations.

Medical Calculators

Medical calculators are available to assist the physician with determining parameters such as body mass index, absolute neutrophil count, and corrected blood volume, among many others (Table 7). Medcalc (http://medcalc.med-ia.net/ desc.html) is a well-known calculator that is available free of charge, but only in the Palm OS. James Suliburk of Handheldmed has written a summary of reviews for many medical calculator programs for the Palm OS (www.handheldmed.com/ newsmore.php?NID=262 &DETAIL=). For the Pocket PC, MedicalPocketPC provides an overview of the available calculator software (www.medicalpocketpc.com/software/calculator.shtml). In psychiatry, the primary “calculation” is the Mini-Mental State Exam. There are a variety of software programs to assist in carrying out and scoring the examination (Table 8); however, they currently only exist for the Palm OS.


Patient Tracking

Patient tracking can be done on the PDA using either the to-do feature or specific software (Table 9). Although many programs exist, even those for mental health purposes, such as the Virtual-Briefcase, organization of patient care information is a matter of preference. Some of the programs listed in Table 9 have integrated features such as charge capture and coding. Programs such as HandDBase (www.handdbase.com) allow users to create specific databases to fit their needs or purchase databases designed for mental health practitioners. Patient Keeper, Patient Tracker, and Ward Watch are specialized programs to store patient information. These are ideal for primary care physicians.



Billing and Coding

Charge capture is part of the practice of medicine, and PDAs can help with the billing and coding. Numerous programs are available (Table 10), such as PocketBilling and Pocket Patient Billing. These programs allow physicians to document evaluation and management (E&M) charges, Current Procedural Terminology codes, procedures, and patient visits. A very useful program is STAT E&M Coder (www.statcoder.com), which has evaluation and management algorithms to determine the proper E&M code for the office visit. Zapmed (www.zapmed.com) and e-MDs (www.e-mds.com) have similar products as well. Many of these companies have products that link to each other as well as to programs on the desktop computer.



Because of the PDA’s portability, security measures must be taken into account, especially in light of sensitive mental health information. The Health Information Portability and Accountability Act (HIPAA) of 1996 was developed to improve the efficiency and effectiveness of electronic information, but many physicians know it as the legislation that mandates protection for health information beginning in 2003. The HIPAA specifies that all providers who conduct electronic billing directly or via clearinghouses must implement security in these transactions to maintain the privacy of an individual’s medical record. Because of the possibility of theft, information in electronic form on the PDA is quite vulnerable. Although the HIPAA does not specify the security mechanism to be used, the best mechanism for security on the PDA is to use encryption software.12 Such software programs (Table 11) require the correct password to access the device as well as to decrypt the data for viewing. Some of the programs will delete all information on the device after a user-defined number of trials, which may be indicative of improper access.



The numerous medical capabilities of PDAs described are applicable to the individual physician or a particular service in the hospital. Additional hardware, such as a keyboard or extra memory, expand the capabilities of these devices even further (Table 12). For example, PDAs in consultation psychiatry13 at the University of California, Davis Medical Center in Sacramento,  have been involved in document editing with software and portable keyboards, accessing drug references, checking DSM-IV diagnostic criteria, determining proper DSM-IV diagnosis coding, and providing electronic sign-out to weekend staff.14 PDAs have been used to decrease medication error rate and have been beneficial in meeting regulatory requirements of the Joint Commission on Accreditation of Healthcare Organizations survey.15 As these devices gain more hardware capabilities such as expansion cards and faster central processing units, new software will provide more assistance to the provision medical care in ways such as remote information access via wireless networks. Newer devices such as the OQO ultra–personal computer (www.oqo.com) and the Tiqit eightythree (www.tiqit.com) will further push the boundary of desktop computing devices in a handheld form factor. Eventually, the desktop computer will be a historic memory in the practice of medicine. PP



1.    Rothschild JM, Lee TH, Bae T, Bates TW. Clinician use of a palmtop drug reference guide. JAMA. 2002;9:223-229.
2.    Agency for Healthcare Research and Quality. Available at: www.ahrq.org/qual/newgrants/ it.htm. Accessed July 28, 2002.
3.    Kane B, Sands DZ. Guidelines for the clinical use of electronic mail with patients. Available at: www.agendacomputing.de/agenda-e/index-e.htm. Accessed: August 2002.
4.    Physician’s Desk Reference. 56th ed. Montvale, NJ: Medical Economics; 2002.
5.    Hogan R. New media: therapeutics—ePocrates qRx. JAMA. 2001;286:229-230.
6.    Diagnostic and Statistical Manual of Mental Disorders. 4th ed., text rev. Washington, DC: American Psychiatric Association; 2000.
7.    Hales RE, Yudofsky SC, eds. The American Psychiatric Press Textbook of Psychiatry. 3rd ed. Washington, DC: American Psychiatric Association; 1999.
8.    Beers MH, Berkow R, eds. The Merck Manual of Diagnosis and Therapy. 17th ed. Rahway, NJ: Merck & Co, Inc; 1999.
9.    Dambro MR, ed. Griffith’s. 5 Minute Clinical Consult-A Reference for Clinicians. 9th ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2001.
10.   Braunwald E, Fauci AS, Kasper DK, eds. Harrison’s Principles of Internal Medicine. 15th ed. New York, NY: McGraw-Hill Professional; 2001.
11.   Degowin RL, Brown DD, eds. DeGowin’s Diagnostic Examination. 7th ed. New York, NY: McGraw-Hill Professional; 1999.
12.   Brown M. Mobile solutions: keep it in your pocket. PC Magazine. 2002;27:77-78.
13.   Luo J, Hales RE, Servis M, Gill M. Use of personal digital assistants in consultation psychiatry. Psychiatr Serv. 2002;53:271-279.
14.   Luo J, Hales RE, Hilty D, Brennan C. Electronic sign-out using a personal digital assistant.  Psychiatr Serv. 2001;52:173-174.
15.    Grasso BC, Genest R. Use of a personal digital assistant in reducing medication error rates. Psychiatr Serv. 2001;52:883-886.


Print Friendly 

Steven E. Hyler, MD, FAPA and Dinu P. Gangure, MD

Primary Psychiatry. 2002;9(9): 24-28

Dr. Hyler is senior attending psychiatrist at St. Luke’s-Roosevelt Hospital Center, clinical professor of psychiatry in the Department of Psychiatry at Columbia University, and staff member of the New York State Psychiatric Institute, all in New York City.

Dr. Gangure is resident in psychiatry at St. Luke’s-Roosevelt Hospital Center and Chair of the Residents’ Committee of the American Psychiatric Association, New York County District Branch.

Acknowledgments: The authors report no financial, academic, or other support of this work.



What are the current technological advances in telepsychiatry and how can clinicians begin to use this new technology in their practice? This article begin with a historical overview and reviews the language of technology and communications used in telepsychiatry. The four state-of-the-art telepsychiatry systems are presented: budget personal computer low, medium, and high bandwith systems, and the satellite broad bandwidth system with signal encryption. The needs of special populations discussed include?mobile cameras for children, higher image resolution for geriatric populations, increased safety of technological devices in prisons, special camera positioning in psychoanalysis, and adjunctive add-ons for testing.



Transport of the mails, transport of the human voice, transport of flickering pictures—in this century, as in others, our highest accomplishments still have the single aim of bringing men together.

—Antoine de Saint-Exupery1

Did Sigmund Freud ever use the telephone in his clinical work with patients? Since the telephone was invented in 1876, it certainly would have been available for use by the upper middle class in Vienna by the early 20th century. This question was recently posed to Drs. Robert Glick and Otto Kernberg of the Columbia Psychoanalytic Institute in New York City, and their response was that they were not aware of any consistent writing on the issue. An Internet search conducted by the authors revealed a single excerpt2 in which Freud mentions the telephone:

I forbade a patient to speak on the telephone to his lady-love, with whom he himself was willing to break off all relations, as each conversation only renewed the struggling against it. He was to write her his final decision, although there were some difficulties in the way of delivering the letter to her. He visited me at one o’clock to tell me that he had found a way of avoiding these difficulties, and among other things he asked me whether he might refer to me in my professional capacity.

At two o’clock, while he was engaged in composing the letter of refusal, he interrupted himself suddenly and said to his mother, “Well, I have forgotten to ask the Professor whether I may use his name in the letter.” He hurried to the telephone, got the connection, and asked the question, “May I speak to the Professor after his dinner?” In answer he got an astonished, “Adolf, have you gone crazy?” The answering voice was the very voice which at my command he had listened to for the last time. He had simply “made a mistake,” and in place of the physician’ number had called up that of his beloved.

Fast forward to the early 21st century. New telecommunications technologies promise to profoundly change the spatial and temporal relationship between health professional and patient. At the beginning of the 21st century, there are no surprises in videoconferencing and the transmission of sound and vision through analogue or digital relays.3 This technology seems to have become an unremarkable component of the normal social experience, and its ubiquity has become almost unquestionable even though it is the harbinger of a “revolution” in communications on an epoch-making scale.4

The United Kingdom National Health Service has commented thus: “Opportunities in the field of telemedicine will be seized to remove distance from health care, to improve quality of that care, and to help deliver new and integrated services.”5 Conceivably, any channel of telecommunication can be used as an adjunctive support for psychiatric service. These include telephone, fax, mail, E-mail, recorded videotapes, the Internet, closed-circuit television systems, and interactive videoconferencing.

The purpose of this article is to review recent technological advances in telecommunications as applied to psychiatry. We begin with a historical overview, review the language of technology and communications used in telepsychiatry, present state-of-the-art telepsychiatry hardware and systems, discuss the special needs of certain populations, and finish with some prognostication about the future.

Historical Overview

The year 1844 marks the invention of the electrical telegraph and coincidental founding of the American Psychiatric Association. (See Table 1 for a psychiatry and telecommunications timeline). The invention of the electrical telegraph made it possible to reliably separate doctor and patient in time and space for the purposes of particular kinds of communicative acts.6,7

The word telephone comes from the Greek roots tele (far) and phone (sound).8 On March 10, 1876, Alexander Graham Bell spoke the first words ever transmitted by telephone: “Mr. Watson, come here, I want you.” These words were actually a call for medical help; Bell had just overturned the wet battery powering his transmitter and spilled sulfuric acid on his clothes.8 Around the same time, Charcot began teaching at the Salpetriere in Paris.

By the mid-1950s television was becoming the focus of attention in living rooms across the world, and Dr. Cecil Wittson of the Nebraska Psychiatric Institute initiated the use of telecommunications in psychiatry.9,10 Via a 2-way closed-circuit television setup, he introduced both staff education and treatment sessions at a distance. Some 20 years later, the first experimental telepsychiatry session via satellite transmission took place from Salt Lake City.11

The term “telepsychiatry” first appeared in the professional literature in 1973 in an article by Dwyer,12 referring to “psychiatric consultation via interactive television.” A recent MEDLINE search (March 2002) by the authors revealed some 380 published studies from 1956–2002 related to telepsychiatry.

The Language of Telepsychiatry

For the purposes of this article we will use the term telepsychiatry rather than “videoconferencing,” “video teleconferencing,” or “teleconferencing” to refer to telepsychiatric equipment as well as the process. A typical telepsychiatry setup involves a “host” site that can connect to one or more “remote” locations. Each site generally has a capturing device that consists of a video camera and a microphone.

More expensive video cameras yield better quality pictures. Factors to consider include lens quality, whether the camera is fixed or can pan around a room, and whether the camera can zoom in and out. The clarity of the image and motion handling are primarily a function of the bandwidth and algorithm used to compress the image prior to transmission to the other site. At the receiving end, the viewing device consists of the monitor, speakers, and a computer powerful enough to process and, optionally, memorize the information received. The monitoring device might consist of the personal computer display in less expensive configurations, while a high-end television monitor is used in more expensive systems.

The channel of communication between sites is the coder-decoder (CODEC), which is the heart of the system. It transforms the analog signal (ie, the picture that is picked up by the video camera and the audio signals picked up by the microphone) to digital signals and compresses them for transmission to the remote site. At the receiving end (remote site) another CODEC transforms the digital signal back to analog, allowing it to be viewed on the video monitor with the audio coming through the speakers. At each end, the CODECs are generally managed by a computerized device similar to those found in most personal computers.

There are a variety of avenues for data transmission between sites. These include geosynchronous satellite transmission (capable of the largest presently available bandwidth), fiber optic lines (the largest bandwidth of terrestrial technologies), Integrated Services Digital Network (ISDN), and the Plain-Old Telephone System (POTS).

There are several compression algorithms responsible for the quality of the signal that is ultimately received. There are industry standards for video and sound compression and for internetwork compatibility with other teleconferencing systems. Bandwidth refers to the amount of data that can be transmitted electronically in a unit of time. POTS/analog lines operate at 56–64 kilobits per second (KBS), which is enough bandwidth to handle smooth voice communication only (motion and image distortions appear when video signals are transmitted in this low bandwidth). ISDN lines operating at 128 KBS are currently the most commonly used. Several years ago, the use of high bandwidth (T1-384 KBS, half of T1-768 KBS, as well as 1.54 megabits per second [MBS]) started to gain prominence, the thought behind it being that such high bandwidths were essential for adequate resolution to assure clinical accuracy. However, lately it has been observed that the quality of information transmission can be preserved in low bandwidth with an improvement in the compression algorithm. The emerging technologies of tomorrow include low-earth orbit satellites, digital wireless technology, cable television, and digital subscriber line.13

The equipment and transmission systems described above form the “hard” technology. Bloomfield and Vurdubakis14 have observed that “hard” technology is not the only kind in play in telepsychiatry. Equally important is the “soft” technology that is constituted by the body of knowledge and practice surrounding, structuring, framing, and enacting physician teleinteractions with their patients. This “soft” technology is formed around the intricately constructed set of interaction techniques and communication skills employed by clinicians as they try to assess or manage the patient in the medical interview.15 Although the “hard” technology is unquestionably important, it is not the most complex aspect of telemental health program management. Because the “hard” technologies involved in telehealth are now robust, the connection between patients and technology, not the technology itself, is usually the major management challenge.16

The framing and presenting of the patient and the practitioner in telemental health is of particular importance and is now referred to by many in the field as “telemedicine etiquette.” For instance, the lighting and sound properties of the teleconsulting room can make all the difference between an experience that the patient enjoys and wants to repeat, and one that is an instant aversion to therapy.16

Telepsychiatry Hardware

Technical specifications of several commercially available videoconference systems in various price ranges are listed in Table 2. Price range is current as of mid-2002, and cost continues to decrease as quality increases. For those with serious interest in setting up a telepsychiatry system, it is recommended that they obtain price quotes from several vendors before deciding on a system.


One of the most frequent problems that occur when setting up a system concerns the coordination of the hardware with the line connections of the telecommunications companies, such as the local telephone and long-distance service providers. As there are no worldwide standards that insure “plug and play” compatibility, it is imperative that buyers obtain from vendors a guarantee of compatibility between the connected sites and other outside systems. Establishing connections to systems that are not directly compatible may be difficult and expensive, requiring the establishment of a temporary “bridge” connection between the two sites.

As with the hardware, the prices of communications lines are also tumbling. A rough estimate of a typical monthly fee is $120 for a 384 KBS connection, with additional fees of ~18 cents/minute.


Despite the technological triumph of this description, there are currently only a few telepsychiatry systems in play, most of which operate on a small scale. Telepsychiatry, the conduct of psychiatric practices mediated by telecommunications systems, offers the possibility of the routine separation of clinician and client in space, and possibly in time. In doing so, it undercuts a central convention of medical work: the involvement of both parties in a physical co-presence.3

Conventionally, clinician and patient encounter each other in a specified place that has a well-understood symbolic identity. Physical colocation and the sense that the clinician is “with” the patient have enormous cultural significance.17 Over the course of 2 millennia, the actual presence of the doctor has been regarded as necessary for the proper conduct of clinical work.18 Today, many debate on the greater importance of nonverbal interactions such as handshakes, olfaction, real visualization, and the ability to extend courtesies such as the use of a handkerchief rather than personal presence.19

A standard telepsychiatry setup as described in this article may be adequate for general practice. Special populations, however, require technological particularities. For example, mobile cameras in specially designed rooms for children,20 higher image resolution for application to geriatric populations,21 increased safety measures (eg, increased protection of technological devices) in prisons,22 and special positioning of the camera and monitor for use in psychoanalysis.23 Adjunctive add-ons might include writing tablets or white boards to facilitate testing of children or performing complete Mini-Mental State Examinations.

Rapid technological “advances” in the medical field seem to be consistently met with contradictory impulses. On one hand, there are demands and expectations for even more effective medical treatments and interventions, while on the other there is growing mistrust of the complex of professional and commercial interests that underpins treatment, and of the potential iatrogenic form that clinical practices might take.24,25 As with any technology, telepsychiatry users should anticipate having to deal with equipment failure. It is essential that there be adequate personnel at the remote site in case of an equipment failure and/or clinical emergency.

A concern in the use of telepsychiatry is the level of signal delay, as even the most expensive systems do not always function perfectly. Manning and colleagues26 conducted a study of signal delay effect (0 ms, 300 ms, and 1,000 ms) on the rapport between patient and clinician compared with in-person sessions (N=48). The level of client-perceived rapport was measured using a self-report instrument that was administered following the session. The results did not provide evidence of an effect on rapport from the delay.

It is difficult to predict whether a limited number of dominant formats will be available for telepsychiatry or whether multiple technologies will be used in combination. Due to continuous technological advances in the telecommunications field, which in turn generate a dynamic of costs both domestically and globally, it is uncertain what the prevailing mode of telepsychiatry technology will be.27 A clinic of the future might have a wing with several telepsychiatry offices used to connect with providers at distance. Once the technology reaches the average patient’s home, one could receive routine telepsychiatry follow-ups, with the recommendation to make periodic in-person psychiatric check-ups—an equivalent to the periodic physical examination.

Commercially run teleconference sites currently charge ~$200/hour per site to provide video teleconference services, mainly to large corporations. These sites could see an increased demand for their services by providing patient access to the psychiatrist. Instead of traveling great distances to a psychiatrist’s office, the patient would go to a centrally located videoconference service center and connect with a psychiatrist who could be located in a city several hundreds miles away. Rigorous assessment of telepsychiatry technology has only recently begun. Key questions that remain to be answered include: Given current technology, in which situations is telepsychiatry cost-effective? What is an appropriate control condition against which to assess clinical and cost outcome data when in-person psychiatry is not available in a geographic area? Which patients will benefit from telepsychiatry in properly controlled, randomized trials?28 What is missing and what is added by using telepsychiatry? Or, as Malagodi and colleagues13 ask, “Will telepsychiatry be a passing fad or the wave of the future?”


New telecommunications technologies promise to profoundly change the spatial and temporal relationship between health professionals and their patients. Conceivably, any channel of telecommunication can be used as an adjunctive support for psychiatric service. These include telephone, fax, mail, E-mail, recorded videotapes, the Internet, closed-circuit television systems, and interactive videoconferencing.

A typical telepsychiatry setup involves a “host” site that can connect to one or more “remote” locations. Each site generally has a capturing device that consists of a videocamera and a microphone. At the receiving end, the viewing device consists of the monitor, speakers, and a computer powerful enough to process and, optionally, memorize the information received. Because the “hard” technologies involved in telehealth are now robust, the major management challenge is not the technology itself but how we can connect people with that technology. Furthermore, rapid technological “advances” in the medical field seem to be consistently met with contradictory impulses. On one hand, there are demands and expectations for even more effective medical treatments and interventions, while on the other there is growing mistrust of the complex of professional and commercial interests that underpins treatment, and of the potential iatrogenic form that clinical practices might take.

Telepsychiatry removes the very presence of another human being, and with it disappears an entire range of levels of interpersonal interactions, including pherohormonal and olfaction senses, symbolic gestures like shaking hands or offering a tissue, unconscious fantasies of physically interacting with the psychiatrist in the very space of the office, and the mystical belief that a human presence can induce feelings and states that no technology could replace.

Telepsychiatry offers assistance to people who would not visit an in-person psychiatrist, a reduction of environmental stimuli that may facilitate an even greater introspection in the inside world of a patient, a chance to have a consultation with the most skilled psychiatrists available in major university centers, potential cost savings, and convenience. Is applying telecommunication technologies to psychiatry a greater benefit or a greater risk? Future research is needed to scientifically answer that question. PP


1.    de Saint-Exupery A. Wind, Sand, and Stars. New York, NY: Harcourt Brace Jovanovich; 1968:69.
2.     Freud S. Psychopathology of Everyday Life. The Complete Psychological Works of Sigmund Freud. Vol 4. London, England: Hogard Press; 1969:222.
3.     May C, Gask L, Atkinson T, Ellis N, Mair F, Esmail A. Resisting and promoting new technologies in clinical practice: the case of telepsychiatry. Soc Sci Med. 2001;52:1889-1901.
4.     Robins K, Webster F. Times of the Technoculture: From the Information Society to the Virtual Life. London, England: Routledge; 1999.
5.    National Health Service Executive. Information for Health: An Information Strategy for the Modern NHS,1998-2001. London, England: National Health Service Executive; 1998.
6.    Reiser SJ. Medicine and the Reign of Technology. Cambridge, England: Cambridge University Press; 1978.
7.    Yoxen E. Seeing with sound: a study of the development of medical images. In: Bijker WE, Hughes T, Pinch T, eds. The Social Construction of Technological Systems. Cambridge, Mass: MIT Press; 1987:281-306.
8.    Grumet GW. Telephone therapy: a review and case report. Am J Orthopsychiatry. 1979;49:574-584.
9.    Wittson CL, Dutton R. A new tool in psychiatric education. Ment Hosp. 1956;7:11-14.
10.    Wittson CL, Affleck DC, Johnson V. Two-way television in group therapy. Ment Hosp. 1961;2:22-23.
11.    Giannetti RA, Johnson JH, Williams TA. Using satellite transmission for computerized assessments of patients in remote facilities. Hosp Community Psychiatry. 1977;28:427.
12.    Dwyer TF. Telepsychiatry: psychiatric consultation by interactive television. Am J Psychiatry. 1973;130:865-869.
13.    Malagodi M, Smith S. Prospective role for telemedicine as a communication tool for rural rehabilitation practice. Work. 1999;12:245-259.
14.    Bloomfield B, Vurdubakis T. Boundary disputes: negotiating the boundary between the technical and the social in the development of IT systems. Information Technology and People. 1994;7:9-24.
15.    Goldberg D, Benjamin S, Creed F. Psychiatry in Medical Practice. 2nd ed. London, England: Routledge; 1994.
16.    Darkins A. Program management of telemental health care services. J Geriatr Psychiatry Neurol. 2001;14:80-87.
17.    Good BJ. Medicine, Rationality and Experience. Cambridge, England: Cambridge University Press; 1994.
18.    Hunter KM. Doctors’ Stories: The Narrative Structure of Medical Knowledge. Princeton, NJ: Princeton University Press; 1991.
19.     Simpson J, Doze S, Urness D, Hailey D, Jacobs P. Evaluation of a routine telepsychiatry service. J Telemed Telecare. 2001;7:90-98.
20.    Ermer D. Experience with a rural telepsychiatry clinic for children and adolescents. Psychiatr Serv. 1999;50:260-261.
21.    Jones BN 3rd, Ruskin PE. Telemedicine and geriatric psychiatry: directions for future research and policy. J Geriatr Psychiatry Neurol. 2001;14:59-62.
22    Zaylor C, Whitten P, Kingsley C. Telemedicine services to a county jail. J Telemed Telecare. 2000;6(suppl 1):93-95.
23.    Kaplan EH. Telepsychotherapy. Psychotherapy by telephone, videotelephone, and computer videoconferencing. J Psychother Pract Res. 1997;6:227-237.
24.    Lupton D. Medicine as Culture. London, England: Sage; 1994.
25.    Lupton D. Foucault and the medicalization critique. In: Petersen A, Bunton R, eds. Foucault, Health and Medicine. London, England: Routledge; 1997:94-112.
26.    Manning TR, Goetz ET, Street RL. Signal delay effects on rapport in telepsychiatry. Cyberpsychol Behav. 2000;3:119-127.
27.    Frueh BC, Deitsch SE, Santos AB, et al. Procedural and methodological issues in telepsychiatry research and program development. Psychiatr Serv. 2000;51:1522-1527.
28.    Baer L, Elford DR, Cukor P. Telepsychiatry at forty: what have we learned? Harv Rev Psychiatry. 1997;5:7-17.

Mr. Kennedy is editor and program director of Medscape Psychiatry and associate in psychiatry in the Department of Psychiatry and Behavioral Health at Albert Einstein College of Medicine in New York City.

Acknowledgments: The author reports no financial, academic, or other support of this work. 



Physicians have become more active Internet users over the past few years. Approximately 95% have online access and an estimated 62% connect to the Internet every day. Online activities that interest physicians include participation in continuing medical education, searching for drug information, E-mail communication with colleagues, and searching professional literature databases.



The Internet was devised in the late 1960s as a “network of networks” originally to be used by the United States military and government as a means of uninterrupted communication across the country. The brilliance of this network is that it established a standard way of communicating where there was no true hierarchy. Each computer on the network had the responsibility of ensuring that communication continued to the rest of the computers on the network.

The government handed the network over to academic and research institutions to use once they no longer needed it. Scientists around the world began sharing ideas, research, and clinical findings through this early network. E-mail was, and still is, the most popular use of the Internet but other forms of communication evolved including file transfer, telnet, and, eventually, the World Wide Web.

With the advent of the World Wide Web, an interactive, easy-to-use “front end” was placed on the Internet with graphics, multimedia, and the concept of “hyperlinks.” Being able to “jump” to pages on the same computer or on another computer anywhere in the world made the Internet a seemless web of connections that only required the click of a mouse.

Early and Late Adopters

Survey data initially generalized that physicians were late adopters to technology and the Internet,1 although this has changed in recent years. Some medical specialties were early adopters of computerization and use of the Internet. These were driven not only by some technically sophisticated specialists but also by those who had a vision of better medical information through the use of computers.

Computers have provided a dramatic change in medical education in the last decade, and medical schools adapted to it early on. This encouraged senior faculty, who were feeling intimidated to catch up, to become computer literate. As the “digital age” medical students graduated, the landscape of medicine transformed to incorporate our “digital senses” and the concept of extended/global communication.

How Physicians Use the Internet

According to Fulcrum Analytics survey2 of 1,200 practicing United States physicians in 2001, 95% of physicians who have online access have been on the Internet in the last 12 months (90% in 2000) and 74% have Internet access in their office. The survey divided physicians into three groups: general online users, which comprised 95% of the total of all practicing physicians; daily Internet users, which comprised 62% of practicing physicians; and professional users—those who described the Internet as essential to their practice—which comprised 21% of the physician population.

According to the researchers, these statistics represented a 5% to 9% increase since the year 2000. Other measures showed that 60% of physicians use dial-up connections, 32% use broadband, 4% have some other connection method, and 4% do not know what kind of connection they have. Of the total amount of physician Internet users, 60% access the Internet from home, 31% access it form work, 1% access it utilizing public computers, and 1% have other means.

The online activities rated as most popular include searching literature databases, access to professional society information, drug searches, E-mail communication, patient education information, continuing medical education (CME) courses, and clinical trial information (Figure).


Search Engines

The Internet has been described as the largest library in the world that has no card catalogue. There are literally billions of Web sites on every imaginable topic, but they serve little purpose if no one knows about them. Search engines,  special Web sites that organize and keep track of other Web sites, were created to solve this problem. Some are organized into catalogues, others contain indices, and others have a combination of both. Many are categorized with human assistance while others are totally automated.

Search engines obtain information about the World Wide Web by either direct submission of a form by the Web site owner or through the accumulation of information gathered by “Web crawlers,” “spiders,” or “robots.” These “robots” are automated programs that roam the World Wide Web from one Web site to another. Some search engines take all their information from the “home page” (often merely a table of contents) while others go levels deeper into the real content of a site. New Web sites are added daily as information is culled from the World Wide Web. The more comprehensive search engines refresh their links at regular intervals to make sure that any changes or updates are included.

Since each search engine gathers different information, conducting a search with different search engines can yield inconsistent results. The analogy of a search engine being a virtual Internet card catalogue is somewhat misleading and each of several virtual card catalogues is slightly different from the other. However, all search engines use some variation of Boolean searches so that the searches can become quite specific. This is important when searching such a vast library. Combining words with “and,” “or,” “not,” or other Boolean operators allows searches to be refined. The more information you give a search, the better information you will get in return. Search results evaluate quantity, not necessarily quality. They are often based on algorithms which count the number of times a particular word is mentioned on a Web site’s home page.

It is important to remember that the Internet is a democracy (ie, anyone can post whatever material they choose). One search might turn up someone’s poetry or a scientific research paper with equal importance.

National Library of Medicine: MEDLINE

MEDLINE(www4.ncbi.nlm.nih.gov/entrez/query.fcgi), a site from the National Library of Medicine,?is probably the most widely used site by physicians and medical professionals. One of the most comprehensive collections of published material in the areas of health and medicine, MEDLINE yields specific results by allowing for Boolean operators when searching for published information or studies. The site also features special collections of information compiled by the National Institutes of Health. One of these collections is MEDLINEplus (http://medlineplus.gov), which contains information on over
500 diseases and conditions.

MEDLINE also provides lists of hospitals and physicians, a medical encyclopedia, medical dictionaries, health information in Spanish, extensive information on prescription and nonprescription drugs, health information from the media, and links to thousands of clinical trials. MEDLINEplus is updated daily and PubMed Central (www.pubmedcentral.nih.gov/), is an archive of life-science journal literature managed by the National Center for Biotechnology Information at the National Library of Medicine.

The Cochrane Collection (www.updatesoftware.com/Cochrane/default.HTM) is an online database of evidenced-based medicine. Other large databases of important medical and health-related information are Ovid (www.ovid.com/products/databases/index.cfm), which contains over 90 commercial databases and bibliographic resources in many research areas, and PsycINFO (www.apa.org/psycinfo/), which provides psychological abstracts.

Associations, Guidelines, and Publications

Sites of major associations, such as the American Psychiatric Association (APA; www.psych.org), are usually great sources of news, information, and the latest practice guidelines. Information about APA sponsored meetings, CME activities, and members is also available. Other subspecialty sites, such as the American Psychosomatic Society (www.psychosomatic.org), the Academy of Psychosomatic Medicine (www.apm.org), Geriatric Psychiatry (www.Aagpgpa.org), or Addiction (www.addictionpsych.org) offer special news, conference information, and other valuable information.

One of the best collections of practice guidelines is the National Guideline Clearing House (www.guidelines.gov). This site offers evidence-based clinical practice guidelines and is sponsored by the Agency for Healthcare Research and Quality in partnership with the American Medical Association and the American Association of Health Plans. It offers guidelines in all specialties from many sources. The site is current, searchable, and conveniently organized by topic collections.

Publisher Sites

Other important sites that are of interest are professional publisher sites. Many of these have relationships with the associations or societies that publish their journals. Some offer abstracts while others offer full-text online access to the journals. American Psychiatric Press (www.appi.org), publisher of many psychiatric journals and books, is the publishing arm of the APA. Their journals are available online in abstract format and full text for subscribers.

Medical megasites are large multispecialty sites that offer many different types of information and CME for the generalist or the specialist. Examples of these are Medscape (www.medscape.com) and Docguide (www.docguide.com). These offer news, original articles, summaries from journals, conference reports, and access to numerous journals and publications. Sites such as these are free but require registration.

The National Institute of Mental Health    

The National Institute of Mental Health (NIMH; www.nimh.nih.gov), which is part of the National Institute of Health (NIH; www.nih.gov/), offers information for professionals and consumers.

Information for Patients

Patient-education information is available from a variety of sources online. The various professional organizations/associations often offer information for patients as do the medical megasites. The NIMH has a section for general information about the various mental disorders. Patient advocacy organizations, such as the National Alliance for the Mentally Ill (www.nami.org), also offer information for patients and families.

Clinical Trial Database

The Clinical Trials Database is offered under the auspices of NIH and provides access to clinical trial information (www.clinicaltrials.gov). This is a site for current information about clinical research studies, both for professionals and patients and their families.

Drug Information

There are a number of drug information sites on the Internet. Many are geared toward consumers and only a few are noteworthy. The National Library of Medicine has a drug reference Web site (www.nlm.nih.gov/medlineplus/druginformation.html) and the Food and Drug Administration has an online database (www.fda.gov/cder/drug/default.htm). Also on the Internet is the Physician’s Desk Reference, (www.pdr.net) a traditional source for drug references, and the Clinical Pharmacology 2000 (http://cp.gsm.com/), which is one of the better drug reference Web sites that is by subscription only.

Continuing Medical Education

Participation in CME programs on the Internet is one of the more popular online activities. A large percentage of CME credits are now being given out on the Internet each year. These programs are offered by the associations, societies, universities, and some of the larger medical Web sites. The topics are interesting and styles vary from simple slide shows to high quality clinical updates and sophisticated interactive programs.

Online CME has a distinctive advantage because it is available 24 hours a day, 7 days a week. This is a great advantage for the busy physician as the Internet traverses great distances to bring information to the desktop. Some CME sites charge per credit or per program while others are free. The most comprehensive list of online CME is organized and catalogued by Bernard Sklar, MD (www.cmelist.com/)

Taking CME to the next level with an “always on” connection (as opposed to a dial-up connection) means that credible, medical information can be available instantaneously. It is even possible to utilize the Internet for “just in time” learning at the point of care.


E-mail, the largest activity that takes place on the Internet, has evolved into an important medium of communication becoming a must in professional and personal communication. E-mail has significantly changed the speed of communication. Many physicians use E-mail to contact colleagues about referrals, report the results of a referral, obtain results of tests, or to give or receive a consultation from a specialist. According to a recent Health on the Net survey, 96% of physicians use E-mail.

A collaborative use of E-mail is called a “listserv,” a list of generally private topic-based E-mails where one needs to request permission to join in order to participate. In a listserv, all of the members receive all of the correspondence; thus, dozen of messages may appear in the mailbox if discussions are particularly active.

One of the pressing questions that physicians are currently debating is whether or not to communicate with their patients via E-mail. Before it is widely adopted, issues, such as liability and reimbursement for time, need to be addressed. In the study, 23% of all physicians reported that they interact with their patients via E-mail, up only 4% from last year, according to the Deloitte Research and Fulcrum Analytics survey.2 This is an increase from the approximate 2% who reported communicating with patients via E-mail in 1999.1 Of the doctors who do not currently E-mail their patients, 79% indicated that their preference for face-to-face contact was the primary reason for not interacting with patients online. Of those physicians, 54% say insurance reimbursement is the leading motivation for them to email their patients in the future.2

The Empowered Consumer

One consequence of the great volume of information on the Internet is that consumers now have access to a proliferation of health information. Increasingly, patients are coming to physicians offices with printouts from the health sites on the Internet. This has both positive and negative aspects to it. Patients are becoming increasingly sophisticated in their understanding of illnesses and asking more intelligent questions and feel that they can participate more in the management of their health. However, it can be problematic if the information that they obtain is not from a credible or reliable source.

Patients and families are increasingly asking their doctors for recommended Web sites to understand illnesses and obtain online support. Physicians need to be aware of some of these sites and be prepared to discuss them.

Electronic Medical Record

The electronic medical record, also known as the digital health record, will soon be available on the Internet. Because of the established standards of communication that governs the activity on the Internet, it can easily become a common link for communications between practices and hospitals with appropriate security measures. It can also empower patients/consumers to manage their portion of the digital health record in ways that previously seemed impossible. For example, patients can request appointments online, complete screening assessments before an appointment, or look up laboratory results. Physicians can monitor treatment compliance and obtain updates that may have been difficult or time consuming by telephone. Patients can actively participate in their treatment.

The Future

The future holds some interesting promises for physicians. The promise of wireless communication means access to medical records and information whenever it is needed regardless of time or location. As computers and connections become ubiquitous and virtually everyone is connected to the “network,” physicians will focus less on the machines or technology and begin to cultivate the knowledge and information that we can be accessed. Timesavers like “intelligent assistants” or “robots” are software agents that can traverse the Internet and gather information for us, make appointments, purchase things, or even order dinner.

There are vast amounts of other advantages the Internet can offer the physician as well. Hundreds of daily newspapers can be found online as are sites for sports, weather, travel, financial information, online banking, investing, and shopping. In short, the Internet is no longer a novelty, but a serious tool for communication and commerce. Just like everyone else, when asked what physicians want from the Internet, the real answer is “everything.”


The Internet has become a part our lives in many more ways than we would have imagined a decade ago. Physicians have adopted the technology of the Internet and the World Wide Web and perceive this technology as a tool to enhance productivity, knowledge, and collaboration. The Internet has facilitated communication with colleagues, institutions, and patients. It has created a global repository of high quality information that the medical profession has incorporated into its normal, daily routine. There are certainly challenges ahead in areas such as confidentiality and security but the world has gotten smaller, faster, and smarter through the opportunities that have become available through the Internet.  PP


1.    Health on the Net Foundation. Evolution of Internet use for health purposes – Feb/Mar 2001. Available at: www.hon.ch/Survey/FebMar2001/ survey.html. Accessed August 14, 2002.
2.    Fulcrum Analytics and Deloitte Research. Taking the Pulse v 2.0: Physicians and Emerging Information Technologies. January 2002. Available at: www.cyberdialogue.com/news/ releases/2002/01-29-ful-takingthepulse.html. Accessed August 14, 2002.

Dr. Norton is assistant professor of psychiatry and neurology at the University of Mississippi College of Medicine and associate professor of neurology and psychiatry at the University of Mississippi Medical Center, both in Jackson.

Acknowledgments: The author reports no financial, academic, or other support of this work.



Attention-deficit/hyperactivity disorder (ADHD) has become a national epidemic. There is a great deal of misunderstanding about the diagnosis and treatment of ADHD, including concerns about the stimulant class of medication typically used in treatment of the disorder. Modafinil has been used to treat narcolepsy and has been shown to attenuate the daytime sleep attacks with a minimum of autonomic side effects that can complicate the use of classic stimulation medications. The following case studies discuss the use of modafinil in two patients who suffer from ADHD but were not able to tolerate stimulant medications and had experienced only moderate efficacy while taking antidepressants. The drug’s mechanism of action is unknown. However,  it does not inhibit growth, and it has little effect on the autonomic nervous system. Reports indicate that modafinil may be a useful treatment option for select patients with ADHD.


Introduction: Modafinil Use in ADHD

During the past decade, there has been great interest in attention-deficit/hyperactivity disorder (ADHD), including concerns about the use of stimulant medications for its treatment. Clearly, this concern is overstated but provides incentive for discovery of newer medications with fewer side effects. The following two case studies involve the use of modafinil in patients with ADHD. Both of these patients had experienced adverse side effects while taking stimulants and did not improve while taking bupropion. Modafinil in the treatment of patients with narcolepsy has shown excellent efficacy. It may therefore provide an alternative option for patients who suffer from this common and difficult disorder.

Case 1

The patient, a 21-year-old male with a history of ADHD, had been diagnosed some 10 years earlier. He had been maintained on dextroamphetamine for more than 5 years but had developed tics, and hence, discontinued medication. He was then administered bupropion, which was marginally effective despite a dose of 150 mg/day BID of the sustained-release form. He still had complaints of inattention, hyperactivity, and impulsivity, which made it difficult for him to maintain work and relationship commitments.

The patient had no other active medical problems, did not abuse illicit drugs, and was not taking any other medications. He had a normal general medical and neurological examination. Laboratory profile included urinalysis, urine toxicology screen, electrolytes, renal and hepatic profile, erythrocyte sedimentation rate, antistreptolysin-O, antinuclear antibody, and cranial magnetic resonance imaging, all of which were within normal limits. The patient was placed on modafinil 200 mg/day, which was increased to 400 mg/day 2 weeks later.

The patient noted that his symptoms significantly improved by 4–6 weeks into the medication trial. He was able to attend work and to concentrate on his tasks. He noted that he experienced much less impulsivity, which was also noted by his friends and family. He did not suffer any adverse side effects while taking modafinil and on 4-month follow-up had sustained his clinical response.

Case 2

The patient, a 17-year-old male with a history of ADHD, had been diagnosed at 9 years of age. His physician had initially prescribed a variety of effective stimulant medications, but the patient had developed autonomic symptoms that became quite distressing to him, resulting in medication discontinuation. The patient had also taken venlafaxine and bupropion with moderate but incomplete resolution of symptoms. He had a C average in school despite a much higher level of cognitive ability, and was also noted to be hyperactive in a variety of settings at home and school.

The patient had no history of other medical problems. He denied substance use and had no history of illegal activity. General physical and neurological examinations were within normal limits. Laboratory profile included electrolytes, renal and hepatic functions, urinalysis, erythrocyte sedimentation rate, and urine toxicology screens, all of which were within normal limits.

The patient was gradually withdrawn from bupropion over a period of 4 weeks and concurrently began taking modafinil 200 mg/day. He noted that by week 3 his attention and concentration had improved and he felt less irritable. His family also noted that he had improved significantly. Furthermore, there was no evidence of autonomic side effects. The patient has been maintained on the same does for more than 3 months with continued resolution of symptoms.


Modafinil is effective in the treatment of narcolepsy. It has stimulating properties with fewer autonomic side effects than the classic stimulant medications. The exact mechanism of action of modafinil is not known. It does not bind to dopamine, serotonin, norepinephrine, or γ-aminobutyric acid receptors. It also does not inhibit monoamine oxidase or phosphodiesterase.1,2 Modafinil has a half-life of 15 hours, and steady state is reached after 2–4 days. Peak plasma concentrations are obtained 2–4 hours after a single dose. Liver metabolism is through cytochrome P450 2C9. Some autoinduction is noted at doses above 400 mg/day. Excretion is through the kidneys.3,4

The major side effects associated with modafinil include headache, anxiety, nausea, and diarrhea. Modafinil is a category C agent, but care must be used in women who are pregnant or nursing. Drug interactions have been minimal. The starting dose is 200 mg/day, and can be increased to 400 mg/day in 1–2 weeks. There is little evidence to suggest that the 400-mg dose has significantly increased efficacy over the 200-mg dose, yet a higher dose is sometimes helpful in the individual patient with an incomplete response to the lower dose.2,5

Modafinil has the ability to alert the patient without leading to the same degree of autonomic changes that can be seen with other stimulant medications. The lack of peripheral sympathetic effects may make this drug more desirable in patients who cannot tolerate other stimulants. There has also been at least one case report  indicating that modafinil has mood-elevating properties that may help with associated depressive symptoms in ADHD.6


ADHD is a major clinical problem that has been misunderstood by the general population and that has been negatively impacted by the press. There are patients who may benefit from the more typical stimulating agents but are unable to continue taking them because of side effects. These patients may not have an optimal response to other agents. In a manner similar to that of bipolar illness, because of varied responses to treatment, it is essential to have a spectrum of agents that are able to treat the underlying illness. Modafinil may be an agent that can be added to the treatment armamentarium. Future double-blind, placebo-controlled trials may help to clarify the role of this agent in the treatment of ADHD. PP


1.    Bassetti C. Narcolepsy. Curr Treat Options Neurol. 1999;1:291-298.
2.    Provigil [package insert]. Westchester, Pa: Cephalon; 1999.
3.    Elovic E. Use of provigil for underarousal following TBI. J Head Trauma Rehab. 2000;15:238-240.
4.    Robertson P, DeCory HH, Madan A, Parkinson A. In vitro inhibition and induction of human hepatic cytochrome P450 enzymes by modafinil. Drug Metab Dispos. 2000;28:664-671.
5.    Jasinski DR. An evaluation of the abuse potentiial of modafinil using methylphenidate as a reference. J Psychopharmacol. 2000;14:53-60.
6.    Menza MA, Kaufman KR, Castellanos A. Modafinil augmentation of antidepressant treatment in depression. J Clin Psychiatry. 2000;61:378-381.

Dr. Martinez-Gonzalez is fellow, Dr. Obermeyer is associate scientist, and Dr. Benca is professor, all in the Department of Psychiatry at the University of Wisconsin–Madison.  Reprinted from TEN. 2001;3(9):48-57. 



Why is the recognition and treatment of insomnia important? Insomnia is highly prevalent and associated with significant costs and morbidities, although it is seldom a focus of medical attention. Patients with insomnia have an increased incidence of health problems and reduced quality of life; they also are at greater risk for developing depression. There has been increasing evidence for a causal relationship between insomnia and health. Appropriate diagnosis and treatment of insomnia may lessen its associated morbidities and ultimately reduce healthcare costs.



Insomnia is one of the most common medical symptoms in industrialized countries, with significant associated costs and comorbidities. Individuals with insomnia complain of difficulties falling asleep, staying asleep, and/or the perception of disturbed or nonrestorative sleep. They also may report fatigue and impairment of daytime functioning. Chronic insomnia affects up to 40% of adults over the course of a year, and 10% of adults complain of moderate to severe sleep problems.1-6 Insomnia is more prevalent with aging and in women. Sleep disturbance is also strongly associated with medical, and particularly, psychiatric illnesses, with up to half of adults with insomnia having concomitant psychiatric disorders.


Costs of Insomnia

The costs associated with insomnia are enormous. The direct costs of treatment include visits to physicians, prescription and nonprescription medications, and nursing home care. In 1995, it was estimated that these costs totaled approximately $13.9 billion in the United States7; whereas in France, with a population approximately one fifth the size of US, the 1995 costs for insomnia treatment were slightly over $2 billion.8 These numbers exemplify the significant direct costs of insomnia in industrialized societies. Since most individuals do not discuss their sleep problems with a physician, and only a minority of patients receives treatment, the potential costs of treating insomnia may be even higher.

The indirect costs related to insomnia, though more difficult to quantify, are likely to be far greater in magnitude. Sleep disturbance is correlated with fatigue-related accidents, decreased productivity, poorer health, reduced quality of life, and medical and psychiatric disorders. It is not clear, however, whether these associations are causes or results of insomnia, although evidence is emerging that both types of relationships exist.

In general, people with insomnia use more medical and psychiatric services than good sleepers.5,9,10 Simon and VonKorff5 reported a 10% prevalence of severe insomnia in primary care patients, and they found that insomnia was associated with significantly greater disability from medical disorders and with increased rates of healthcare utilization and depression. In another large survey of the general population, people with insomnia had higher rates of utilization of general medical services.1 A study using data (1980–1984) from the National Institute of Mental Health Epidemiological Catchment Area Study compared morbidity and mortality of insomnia in subjects with or without a history of psychiatric illness during the previous year.9 Both groups of insomniacs showed greater utilization of general medical and psychiatric services, although rates were even higher in subjects with a prior history of psychiatric disorders.

Subjective reports of poorer quality of life have been consistent with incidence of insomnia in various populations, including a National Sleep Foundation/ Gallup poll,6 a survey of US primary care patients in California and Hawaii,11 and a Japanese study.12 Insomnia is also correlated with decreased work productivity, increased absenteeism, higher rate of accidents, and complaints of daytime fatigue.13,14 Deficits in cognitive and psychomotor functioning, including memory, concentration, attention, reasoning, problem-solving, and reaction time, have also been demonstrated in people with insomnia.6,11,13,15

Insomnia and Medical Disorders

In addition to their subjective sense of poorer health quality, insomniacs show increased rates of medical disorders. Mellinger and colleagues2 reported that over half of individuals with serious insomnia had two or more health problems, in comparison to only about one quarter of those with no trouble sleeping. Not only do insomniacs have more medical problems, but medically ill patients also have more sleep complaints. A survey of patients in primary care clinics in Hawaii and California found a total prevalence of insomnia of 69%, with 19% reporting chronic insomnia.11 In a study of medical outpatients, Katz and McHorney16 found that half complained of sleep disturbance, with severe insomnia reported by 16%. Sleep problems were associated with a variety of medical conditions, including cardiopulmonary disease, musculoskeletal conditions, prostate problems, and depression.

Sleep disturbance may also contribute to medical disorders. A recent meta-analysis by Schwartz and colleauges17 demonstrated that trouble falling asleep was associated with coronary events, independent of other risk factors for cardiac disease.17 Several studies of the associations between insomnia and medical illness have demonstrated that sleep disturbance tends to change in relation to health status. In geriatric subjects, resolution of insomnia was significantly associated with improvement in self-perceived health.18 Severity of symptoms in patients with irritable bowel disorder was predicted by the prior night’s sleep, suggesting a potential causal relationship.19 Thus, sleep disturbance may be both a cause and a result of medical disorders, and treatment of insomnia may decrease the need for utilization of other medical services.

Insomnia and Psychiatric Disorders

Insomnia is most strongly associated with psychiatric disorders, particularly anxiety and depression. Psychiatric patients commonly report sleep difficulties, and polysomnographic studies show objective sleep abnormalities in association with all major psychiatric disorders (Table 1).20,21 Sleep complaints are a primary or associated diagnostic criteria for most psychiatric illnesses, which disrupt sleep through a variety of mechanisms, including the increased anxiety and arousal that accompany most acute episodes of illness and the secondary effects of psychotropic medications.


Chronic insomnia is also strongly associated with psychiatric symptomatology and psychiatric disorders. Increased rates of psychological stress and poorer ability to cope with stress have been connected with insomnia in several surveys.6,12 Insomniacs also show more abnormalities on psychological testing. Almost 80% show significant increases on one or more clinical scales on the Minnesota Multiphasic Personality Inventory (MMPI).22 These results may not have been due solely to an increase in psychiatric disorders, since even people whose insomnia was due to identified medical factors showed elevations on the MMPI, suggesting a direct effect of sleep disturbance on psychological symptomatology.

Epidemiologic studies of the general adult population have shown that one third to one half of people with chronic insomnia suffer from primary psychiatric disorders, predominantly anxiety and mood disorders. Mellinger and colleagues2 noted that among adults reporting “a lot” of trouble falling asleep or staying asleep over the past year (17% of the population), almost half of them had high levels of psychiatric distress with symptoms consistent with depression and anxiety disorders.2 In a large-scale survey of almost 8,000 individuals, Ford and Kamerow1 reported that 40% of those with significant insomnia met criteria for psychiatric disorders, whereas only 16% of those without sleep complaints had psychiatric illnesses. Breslau23 found a strong correlation between lifetime prevalences of sleep problems and psychiatric disorders in a study of young adults. Anxiety, depression, and substance abuse were the most common disorders in this population.

The comorbidity between insomnia and psychiatric disorders may be even higher in clinical populations. In one study, over half of insomnia patients presenting to general medical or sleep disorders clinics were reported to have psychiatric symptoms.24 At least three quarters of insomnia patients in sleep or general medical clinics were diagnosed with a psychiatric illness in another study.25 Katz and Horney16 found that, in medical outpatients, major depressive disorder and depressive symptoms were more strongly associated with insomnia than were other chronic illnesses.

Most psychiatric patients complain of sleep disturbance, not only during periods of acute illness but also during periods of remission.26 Virtually all psychiatric patient groups show changes in sleep architecture that are associated with insomnia. These changes include reduced sleep efficiency, prolonged latency to sleep onset, increased time awake during the sleep period, and reduced amounts of total sleep.21 Thus, the increase in subjective complaints of insomnia among psychiatric patients is based on objective changes in sleep, and are not a consequence of a simple reporting bias or misperception of sleep state related to their psychiatric illness.

Mood Disorders

Depression has been studied more extensively than any other psychiatric disorder. Patients with depression frequently report insomnia, although a minority complain of hypersomnia. Hypersomnia is more common in those with bipolar disorder or winter depression,27 often characterized by symptoms of prolonged nocturnal sleep and daytime napping. Depressed people may also complain of intense or distressing dreams and daytime fatigue. Manic patients may have severe insomnia during episodes of acute illness, accompanied by the sense of a decreased need for sleep. Insomnia often precedes mania and has been identified as a precipitant or perpetuating factor. The relationship between sleep loss and mania is one of the most robust associations between sleep disturbance and resulting illness.

Patients with major depression show the most robust and consistent changes in sleep architecture versus other psychiatric disorders.21 These sleep abnormalities are usually grouped into three categories28:

(1) Decreased sleep continuity, consisting of prolonged sleep-onset latency, increased wakefulness during the sleep period, and early morning awakening, resulting in reduced sleep efficiency and total time spent asleep.

(2)  Decreased slow-wave sleep (SWS), including reductions in SWS as percentage of total sleep, total time spent in SWS, and decreased δ-wave activity during sleep versus age-matched normal control subjects.

(3)  Rapid eye movement (REM) sleep abnormalities, including reduced latency to REM sleep onset, increased REM sleep as percentage of total sleep, increased proportion of REM sleep in the first third of the night, and increased total number or density of REMs across the night.

Sleep is generally most disturbed during acute episodes of depression, although certain sleep abnormalities tend to persist during periods of clinical remission. These persistent “trait” abnormalities are reduced REM sleep latency and loss of SWS, which have been considered markers for depression. Other sleep changes, such as REM density and sleep continuity, may reflect the state of depression because they have been reported to fluctuate with the illness.29

While bipolar patients in an episode of acute mania show the same sleep abnormalities as depressives,30,31 patients with dysthymia or subclinical depression seem to be indistinguishable from normal control subjects.

Cause and Effect Between Insomnia and Depression

There is increasing evidence supporting causal relationships between sleep and mood disorders. In prospective studies, complaints of insomnia were found to have a positive predictive value for subsequent development of depression. Ford and Kamerow1 found that subjects with insomnia had a greatly increased risk of developing a new episode of major depression 1 year later, compared to individuals with no insomnia. A prospective study of community-dwelling, elderly over a 2-year period, found that sleep disturbance was the best predictor for development of depression.32 Another study assessed the sleep habits of over 1,000 men while they were medical students and followed them for a median of 34 years. Individuals with insomnia during medical school had an increased risk of depression later in life (relative risk 2.0), as did those reporting difficulty sleeping under stress (relative risk 1.8), versus those with no sleep problems.33 Breslau23 assessed the incidence of depression over a 3.5-year period in young adults with a history of insomnia compared to those with no prior history of insomnia. Even though controlling for the prior history of other depressive symptoms would naturally have reduced the odds ratio, the risk for developing depression with a history of insomnia remained high (odds ratio 2.1). Another study that controlled the independent contributions of various depressive symptoms was performed in older adults.34 In this population, other depressive symptoms were found to have higher predictive values; insomnia ranked third among women, but last among men. These data suggest that sleep disturbance is an important risk factor and possibly a precipitant for depression.

Anxiety Disorders

Anxiety disorders, including generalized anxiety disorder, panic disorder, posttraumatic stress disorder (PTSD), and obsessive-compulsive disorder (OCD), are commonly associated with insomnia. In general, sleep studies performed in patients with any of these disorders report prolonged latency to sleep onset, increased time awake during the sleep period, early morning awakening, reduced sleep efficiency, and decreased total sleep.

Patients with generalized anxiety disorder experience chronic and persistent anxiety and, not surprisingly, most report problems with insomnia. Sleep studies have shown prolonged sleep latency, reduced sleep efficiency, early morning awakening, and reduced total sleep time.35,36

Sleep-related panic attacks affect most patients with panic disorders; one third or more of patients with panic disorder suffer recurrent nocturnal panic attacks.37 Symptomatology of sleep panic attacks is similar to wake panic attacks and they occur more commonly at transitions from stage 2 to SWS.38,39 Nocturnal panic attacks are characterized by waking in a state of intense fear or anxiety associated with palpitations, shortness of breath, choking sensation, chest discomfort, and chills or hot flushes. In contrast to night terrors, which are characterized by incomplete arousal from sleep, in a sleep panic attack, patients are awake and alert immediately after the attack begins. Patients with recurrent sleep panic attacks may become fearful of going to sleep, which can contribute further to their insomnia.

Individuals with PTSD have a history of experiencing a traumatic event and reexperiencing the event in flashbacks, intrusive recollections of the event, or recurrent dreams of the event. This has led to investigation of possible REM sleep abnormalities in PTSD.40 PTSD patients exhibit a higher percentage of REM sleep, fewer arousals from non-REM sleep, and perceptions of poorer sleep quality.41

Patients with OCD may have reduced sleep continuity, increased wake percent during sleep period, and decreased REM latency.42-44 Further, the obsessions and compulsions of OCD patients may directly disturb their sleep.


Schizophrenic patients suffer significant sleep disruption, particularly during acute exacerbations of illness.21,45 They typically report increased nocturnal wakefulness, daytime fatigue and napping, and frightening dreams. Objective studies performed in schizophrenics show reduced REM sleep latency46-48 and decreased amounts of SWS; this loss of SWS appears to be correlated with negative symptoms and abnormalities in the prefrontal cortex.49,50

Eating Disorders

Patients with eating disorders may have a variety of sleep complaints and objective sleep abnormalities.51 Those with anorexia nervosa typically report excess energy and symptoms of insomnia, particularly during periods of weight loss, whereas those with bulimia nervosa may experience hypersomnia following eating binges. Polysomnographic studies have documented sleep abnormalities generally similar to those seen in depression, such as sleep continuity disruption, loss of SWS, and reduced REM sleep latency.52,53

Binge eating in bulimics occurs in the evening or during the night, and some patients may binge-eat during sleep54; they typically get up sometime after sleep onset and consume large amounts of high-calorie foods. Polysomnographic studies have shown SWS parasomnias in about half of patients studied, and many have prior histories of sleepwalking.55,56 Sleep-related eating has also been reported in patients with other sleep disorders (eg, periodic limb movements, narcolepsy, and obstructive sleep apnea), substance abuse (eg, benzodiazepine and alcohol abuse), and other psychiatric disorders (mood and anxiety disorders).55,57


Most alcoholic patients have significant insomnia symptoms. In a study of patients receiving treatment for alcohol dependence, 61% complained of insomnia during the preceding 6 months.58 Chronic alcoholics tend to show loss of SWS and sleep disruption, particularly if they abstain from drinking.59 Sleep during withdrawal from alcohol is characterized by reduced total sleep, even greater prolongation of sleep latency, and a relative loss of SWS. Increased REM density and/or increased REM sleep amount may also be seen. Both subjective sleep disturbance and objective changes in sleep architecture may predict increased risk for relapse of alcoholism.58,60

Development and Aging

Sleep is profoundly affected by age; not surprisingly, the prevalence of particular kinds of sleep problems clearly varies with age. For example, infants and toddlers have more difficulty falling asleep compared with preadolescent school-age children.61,62 Younger children show increased incidence of parasomnias, such as sleepwalking, nightmares, bruxism, and enuresis, whereas adolescents and adults have more problems with insomnia and daytime sleepiness. Even in children, however, there are significant associations between insomnia and behavioral problems and/or psychiatric disorders.21,63-65

Children with various medical disorders (eg, allergies, asthma, upper respiratory infections, otitis media) also have increased rates of sleep problems.64,66 Overall, sleep problems in children are significantly associated with psychopathology, behavioral problems, and other medical disorders similar to associations in adults.21,64,67

In the elderly, sleep is more shallow and disrupted, with reduction of total sleep, decreased sleep efficiency, prolonged sleep latency, increased arousals during sleep, loss of SWS, and increase in daytime napping. In addition, some primary sleep disorders, such as sleep apnea, periodic movements in sleep, and REM sleep behavior disorder, are more prevalent among the elderly.1,68 The circadian rhythm tends to advance in older people, causing early morning awakening.69

Insomnia is common in the elderly and strongly associated with depressed mood and physical disease.68-70 Sleep disturbance is frequently seen in persons with dementia and Parkinson’s disease, which are often associated with old age. In dementia, the degree of sleep disturbance is likely related to the severity of the disease.71 “Sundowning” refers to episodes of confusion and agitation during the sleep period in patients with dementia.72 In addition, dementia has been associated with prolongation of sleep latency, reduced sleep efficiency, and loss of total sleep time in comparison to age-matched normal subjects.21 Individuals with Alzheimer’s disease may also show decrements in SWS and REM sleep, perhaps related to loss of cholinergic neurons.73-75 Insomnia should be considered a significant symptom in the elderly and not simply assumed to be part of normal aging.76

Drug Effects on Sleep

Most psychotropic medications have significant effects on sleep patterns that may either improve or worsen sleep problems (Table 2).77 The newer antidepressants, including the selective serotonin reuptake inhibitors (SSRIs) (eg, fluoxetine, paroxetine, sertraline) and venlafaxine, may induce insomnia, which may lead to exacerbation of sleep difficulties in some patients. The older tricyclic antidepressants tend to cause sedation, which is why they have typically been administered in a single bedtime dose. Most antidepressants, particularly tricyclics and SSRIs, are associated with higher rates of some primary sleep disorders, including restless legs syndrome, periodic leg movements, and REM sleep behavior disorder.78-80 Increased eye movements during non-REM sleep have also been reported in patients taking SSRIs; the clinical significance of this finding is unknown.80


Given the importance of brainstem monoaminergic systems in REM sleep regulation, it is not surprising that most antidepressants have profound effects on sleep architecture, probably related to their effects on increasing monoaminergic transmission (norepinephrine and/ or serotonin) in the central nervous system. Most antidepressants suppress REM sleep, prolong REM latency, and reduce total amounts of REM sleep.81 Monoamine oxidase inhibitors can lead to profound suppression of REM sleep,82,83 but tricyclics and SSRIs also reduce REM sleep amounts significantly. REM sleep rebound consisting of increases in REM sleep amounts, reduced REM sleep latency, and sleep disruption, may occur following discontinuation of REM sleep-suppressing antidepressants. Several effective antidepressants, including nefazodone,84 trimipramine, iprindole, and amineptine,81 appear to have no REM sleep-suppressing properties.

Although all mood stabilizers typically used in bipolar patients have sedative effects, they have few other effects on sleep architecture. Lithium, however, may increase SWS and can have REM sleep-suppressing effects.85

Antipsychotic medications tend to have sedative effects, and objective sleep studies have demonstrated reduced sleep latency and increases in total sleep.86-88

Treatment of Insomnia

Treatment of insomnia depends on accurate assessment of the underlying causes. Medical illnesses, psychiatric disorders, and primary sleep disorders need to be addressed specifically. Most patients with insomnia, however, may also benefit from attention to sleep hygiene, behavioral therapy, and/or symptomatic treatment with hypnotic medications, regardless of the underlying cause.

Treatment of primary insomnia and sleep disturbance related to psychiatric illness should begin with a careful review of sleep-related habits and practices. Insomnia patients frequently engage in activities that interfere with sleep, and the development of good sleep habits is an important first step in behavioral treatment. The principles of sleep hygiene, outlined in Table 3, include establishment of a regular sleep-waking schedule to reinforce the circadian or daily sleep-waking rhythm, avoidance of sleep-disrupting activities, elimination of stimulants, and maintenance of a sleep-conducive environment. Sleep hygiene is particularly important for psychiatric patients, many of whom may lack daily structure due to their illness.


Other more focused behavioral treatments have been developed, particularly for patients with primary insomnia. These may also be effective for many patients with chronic sleep disturbance from other causes. Several different types of behavioral interventions have been found to be effective in chronic insomnia, but there is not yet a clear consensus as to which specific types or schedules of treatments are more effective. Various types of relaxation techniques have also been used in insomnia. These include progressive muscle relaxation, guided imagery, meditation, and biofeedback, all of which may decrease physiologic and/or cognitive arousal and thus promote sleep.

Stimulus control therapy, developed by Bootzin and Nicassio,89 was primarily designed to reestablish a positive association between the bed and falling asleep; most chronic insomniacs actually become aroused in the sleeping environment after repeated experiences of insomnia. With stimulus control, time spent awake in bed is minimized since patients are only allowed to remain in bed if they are drowsy or fall asleep. Patients are instructed to go to bed only when they are sleepy and not to engage in activities other than sleep or sexual activity in bed. If they do not fall asleep in about 10 minutes, or find themselves becoming aroused or anxious, they are to get out of bed, go to another room, and engage in quiet or relaxing activities until they feel sleepy, at which point they may return to bed. If they cannot fall asleep, they are to get out of bed again and repeat the process until they successfully fall asleep within 10 minutes. Napping is not allowed during the day, and patients must have a fixed waking time in the morning.

Sleep restriction therapy,90 like stimulus control, may also work in part by creating more consistent experiences of falling asleep quickly in bed. Sleep restriction attempts to increase homeostatic pressure for sleep by limiting the hours spent in bed. Patients are instructed to restrict the time spent in bed to the actual hours that they sleep, but not less than 4.5 hours; a set wake-up time is established at the beginning of treatment. If sleep efficiency increases to an average of 90% or more for 5 days, the patient increases the time in bed by 15 minutes. Conversely, if sleep efficiency decreases below 85%, time in bed is reduced to the average sleep time from the previous 5 days.

In general, behavioral therapies, alone or in combination with medication, have been shown to be more effective in treating chronic insomnia than medications alone. Many medications with hypnotic properties, however, are frequently used in the treatment of insomnia (Table 4). The most commonly prescribed medications are benzodiazepine-receptor agonists. They tend to be relatively safe and effective in comparison with other available agents, particularly the barbiturates, which they have largely replaced. Benzodiazepines tend to promote sleep onset, increase total time asleep, and suppress SWS. Some of the newer nonbenzodiazepine drugs that act as benzodiazepine-receptor agonists, such as zolpidem and zaleplon, have been reported not to have SWS suppressive effects.91-93


Low doses of antidepressants are commonly prescribed for insomnia.94 Generally tricyclic antidepressants improve sleep latency and sleep continuity in insomnia patients with depression.95 Trazodone, a nontricyclic antidepressant with sedating properties, also increases total sleep time in depressed patients with insomnia.96 The drug has shown efficacy as a hypnotic in patients with chronic insomnia even in the absence of depression.97 Trazodone, along with other newer antidepressants with sleep-inducing properties, such as nefazodone and mirtazapine, is frequently combined with other antidepressants to improve sleep.98

Histamine-receptor type 1 antihistamines are frequently used as hypnotics. Antihistamine constitutes the primary active ingredient in most over-the-counter sleeping aids. Although studies have shown that antihistamines increase sleepiness in healthy normal individuals, no studies have clearly established the dose range over which hypnotic effects in people with insomnia might be found.99 They are also associated with potentially significant side effects, such as daytime sedation, orthostatic hypotension, and other anticholinergic effects. In some cases, they can have paradoxic activating effects.

Melatonin, a natural hormone, is frequently self-administered for insomnia. Unlike the benzodiazepine hypnotics, melatonin is not a tranquilizer in the classic sense or even a particularly sedating drug. Melatonin has sometimes been referred to as a soporific (to make drowsy) rather than a hypnotic per se.100 Little data are available on melatonin’s effects as a hypnotic in people with insomnia. One study, using very large doses of melatonin (75 mg orally) for 14 consecutive days, showed a significant increase in subjective total sleep time and daytime alertness.101 Studies in which wrist motor activity was used as an outcome variable have reported that elderly insomniacs have reduced motor activity at night after taking both low and high doses of melatonin just before bedtime.102-104

This has raised the hope that melatonin may prove to be effective in the treatment of disrupted sleep, particularly in the elderly. However, there are no positive polysomnographic data to document that melatonin improves sleep maintenance insomnia comparable to the demonstrated efficacy of the short-acting benzodiazepines.105 Melatonin may be useful, however, as a chronobiotic to help with resetting or entraining circadian rhythms in shiftworkers or others with sleep schedule disorders.

Treatment of Sleep Problems in Psychiatric Patients

When insomnia is related to a psychiatric disorder, primary treatment should be directed at the disorder itself. In patients with mood disorders, the choice of antidepressant is usually influenced by the sleep complaints; sedating antidepressants are often prescribed for patients with insomnia (eg, tricyclics, trazodone, nefazodone, or mirtazapine), whereas more activating medications (eg, SSRIs or bupropion) are given to patients with fatigue and/or hypersomnia. When an SSRI or other activating antidepressant is used in patients with insomnia, sedating antidepressants and/or hypnotics may be added to improve sleep. Interestingly, antidepressant response does not appear to depend on eliminating insomnia, as indicated by a comparison of fluoxetine and nefazodone; both were effective, although patients treated with fluoxetine continued to show evidence of insomnia whereas nefazodone improved sleep quality.106 Insomnia should be monitored carefully in bipolar patients and treated aggressively, since sleep loss can trigger or exacerbate mania.

Antianxiety drugs often have sedative-hypnotic effects. Benzodiazepines are often given to patients with anxiety disorders in a larger dose at bedtime, for sleep induction and maintenance.

In schizophrenic patients, most antipsychotic drugs have sleep-promoting effects, particularly the low-potency neuroleptics (eg, chlorpromazine) and clozapine, and hypersomnolence is a significant side effect of these medications. Behavioral disorganization is frequently a significant contributing factor to sleep disturbance in schizophrenics, and should be addressed through improved sleep hygiene.

In general, sedative-hypnotic drugs should not be given to patients with histories of substance abuse. Treatment must be based on substance withdrawal and maintenance of abstinence. Patients should be educated in principles of sleep hygiene and comorbid psychiatric disorders should be treated. Sedating antidepressants may be helpful for those who continue to suffer from significant sleep disruption.


Although insomnia is widely prevalent and associated with significant negative health outcomes, it is generally underdiagnosed and undertreated.4 To date, insomnia research and treatment have concentrated on careful description and symptomatic relief, respectively.

Symptomatic relief has been achieved via both behavioral and pharmacologic means. Studies have shown that this symptomatic relief may have broader benefits. Quality of life has been reported to improve following the treatment of insomnia.107,108 On the other hand, at least one recent study has been unable to demonstrate an effect of insomnia treatment on reported health (both physical and emotional), anxiety, depression, optimism, and absenteeism,13 although the study was limited by lack of information regarding resolution of insomnia.

The functional relationship between insomnia and the wide variety of associated medical and psychiatric disorders has been difficult to clarify. Studies have suggested that insomnia may precede or exacerbate medical and psychiatric disorders. Perhaps more to the point, treatment of insomnia may or may not reduce associated sequelae.

It has been difficult to tease out the effects of available treatments from the putative effects of insomnia. Many hypnotics have residual effects that can impair daytime functioning. For example, a recent meta-analysis of the effects of benzodiazepines in the treatment of insomnia found that they increased both sleep duration and daytime drowsiness and light-headedness.91 Other studies have reported increased impairment in cognitive function with benzodiazepine use, particularly with longer-acting compounds. Thus, it is possible that treating insomnia actually reduces the sequelae of insomnia while at the same time replaces them with drug side effects displaying a similar constellation of daytime symptoms.

It is now possible to sidestep that problem. Newer agents, especially those with ultra-short half-lives, appear to have less tendency to impair cognitive and psychomotor performance the morning after use.92 Together with behavioral and environmental intervention, we can now see the results of insomnia treatment in a clearer light.

Appropriate recognition and treatment of insomnia will ultimately depend on a better appreciation of its true costs and morbidities and the demonstration that the costs can be reduced through medical intervention.  PP


1.    Ford DE, Kamerow DB. Epidemiologic study of sleep disturbance and psychiatric disorders: an opportunity for prevention? JAMA. 1989;262:1479-1484.
2.    Mellinger GD, Balter MB, Uhlenhuth EH. Insomnia and its treatment. Prevalence and correlates. Arch Gen Psychiatry. 1985;42:225-232.
3.    Bixler EO, Kales A, Soldatos CR, Kales JD, Healey S. Prevalence of sleep disorders in the Los Angeles metropolitan area. Am J Psychiatry. 1979;136:1257-1262.
4.    Ancoli-Israel S, Roth T. Characteristics of insomnia in the United States: results of the 1991 National Sleep Foundation Survey. I. Sleep. 1999;22(suppl 2):S347-S53.
5.    Simon GE, VonKorff M. Prevalence, burden, and treatment of insomnia in primary care. Am J Psychiatry. 1997;154:1417-1423.
6.    Roth T, Ancoli-Israel S. Daytime consequences and correlates of insomnia in the United States: results of the 1991 National Sleep Foundation Survey. II. Sleep. 1999;22(suppl 2):S354-358.
7.    Walsh JK, Engelhardt CL. The direct economic costs of insomnia in the United States for 1995. Sleep. 1999;22(suppl 2):S386-393.
8.    Leger D, Levy E, Paillard M. The direct costs of insomnia in France. Sleep. 1999;22(suppl 2):S394-401.
9.    Weissman MM, Greenwald S, Nino-Murcia G, Dement WC. The morbidity of insomnia uncomplicated by psychiatric disorders. Gen Hosp Psychiatry. 1997;19:245-250.
10.    Chevalier H, Los F, Boichut D, et al. Evaluation of severe insomnia in the general population: results of a European multinational survey.
J Psychopharmacol. 1999;13:S21-24.
11.    Shochat T, Umphress J, Israel AG, Ancoli-Israel S. Insomnia in primary care patients. Sleep. 1999;22(suppl 2):S359-365.
12.    Kim K, Uchiyama M, Okawa M, Liu X, Ogihara R. An epidemiological study of insomnia among the Japanese general population. Sleep. 2000;23:41-47.
13.    Zammit GK, Weiner J, Damato N, Sillup GP, McMillan CA. Quality of life in people with insomnia. Sleep. 1999;22(suppl 2):379-385.
14.    Balter MB, Uhlenhuth EH. New epidemiologic findings about insomnia and its treatment. J Clin Psychiatry. 1992;53(suppl):34-39; discussion 40-42.
15.    Hauri PJ. Cognitive deficits in insomnia patients. Acta Neurol Belg. 1997;97:113-117.
16.    Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Intern Med. 1998;158:1099-1107.
17.    Schwartz S, McDowell Anderson W, Cole SR, Cornoni-Huntley J, Hays JC, Blazer D. Insomnia and heart disease: a review of epidemiologic studies. J Psychosom Res. 1999;47:313-333.
18.    Foley DJ, Monjan A, Simonsick EM, Wallace RB, Blazer DG. Incidence and remission of insomnia among elderly adults: an epidemiologic study of 6,800 persons over three years. Sleep. 1999;22(suppl 2):S366-372.
19.    Goldsmith G, Levin JS. Effect of sleep quality on symptoms of irritable bowel syndrome. Dig Dis Sci. 1993;38:1809-1814.
20.    Benca RM. Sleep in psychiatric disorders. In: Aldrich MS, ed. Sleep Disorders. Philadelphia, Pa: W.B. Saunders; 1996:739-764.
21.    Benca RM, Obermeyer WH, Thisted RA, Gillin JC. Sleep and psychiatric disorders: a meta-analysis. Arch Gen Psychiatry. 1992;49:651-668.
22.    Kalogjera-Sackellares D, Cartwright RD. Comparison of MMPI profiles in medically and psychologically based insomnias. Psychiatry Res. 1997;70:49-56.
23.    Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry. 1996;39:411-418.
24.    Charon F, Dramaix M, Mendlewicz J. Epidemiological survey of insomniac subjects in a sample of 1,761 outpatients. Neuropsychobiology. 1989;21:109-110.
25.    Buysse DJ, Reynolds CF III, Kupfer DJ, et al. Clinical diagnoses in 216 insomnia patients using the International Classification of Sleep Disorders (ICSD), DSM-IV and ICD-10 categories: a report from the APA/NIMH DSM-IV Field Trial. Sleep. 1994;17:630-637.
26.    Rush AJ, Erman MK, Giles DE, et al. Polysomnographic findings in recently drug-free and clinically remitted depressed patients. Arch Gen Psychiatry. 1986;43:878-884.
27.    Rosenthal NE, Sack DA, Gillin JC, et al. Seasonal affective disorder. A description of the syndrome and preliminary findings with light therapy. Arch Gen Psychiatry. 1984;41:72-80.
28.    Reynolds CF, III, Kupfer DJ. Sleep research in affective illness: state of the art circa 1987. Sleep. 1987;10:199-215.
29.    Thase ME, Fasiczka AL, Berman SR, Simons AD, Reynolds CFr. Electroencephalographic sleep profiles before and after cognitive behavior therapy of depression. Arch Gen Psychiatry. 1998;55:138-144.
30.    Hudson JI, Lipinski JF, Keck PE, et al. Polysomnographic characteristics of young manic patients: comparison with unipolar depressed patients and normal control subjects. Arch Gen Psychiatry. 1992;49:378-383.
31.    Linkowski P, Kerkhofs M, Rielaert C, Mendlewicz J. Sleep during mania in manic-depressive males. Eur Arch Psychiatry Neurol Sci. 1986;235:339-341.
32.    Livingston G, Blizard B, Mann A. Does sleep disturbance predict depression in elderly people? A study in inner London. Br J Gen Pract. 1993;43:445-448.
33.    Chang PP, Ford DE, Mead LA, Cooper-Patrick L, Klag MJ. Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. Am J Epidemiol. 1997;146:105-114.
34.    Dryman A, Eaton WW. Affective symptoms associated with the onset of major depression in the community: findings from the US National Institute of Mental Health Epidemiologic Catchment Area Program. Acta Psychiatr Scand. 1991;84:1-5.
35.    Papadimitriou GN, Linkowski P, Kerkhofs M, Kempenaers C, Mendlewicz J. Sleep EEG recordings in generalized anxiety disorder with significant depression. J Affect Disord. 1988;15:113-118.
36.    Reynolds CF, III, Shaw DH, Newton TF, Coble PA, Kupfer DJ. EEG sleep in outpatients with generalized anxiety: a preliminary comparison with depressed outpatients. Psychiatry Res. 1983;8:81-89.
37.    Mellman TA, Uhde TW. Sleep panic attacks: new clinical findings and theoretical implications. Am J Psychiatry. 1989;146:1204-1207.
38.    Hauri PJ, Friedman M, Ravaris CL. Sleep in patients with spontaneous panic attacks. Sleep. 1989;12:323-337.
39.    Mellman TA, Uhde TW. Electroencephalographic sleep in panic disorder. Arch Gen Psychiatry. 1989;46:178-184.
40.    Ross RJ, Ball WA, Sullivan KA, Caroff SN. Sleep disturbance as the hallmark of posttraumatic stress disorder. Am J Psychiatry. 1989;146:697-707.
41.    Engdahl BE, Eberly RE, Hurwitz TD, Mahowald MW, Blake J. Sleep in a community sample of elderly war veterans with and without posttraumatic stress disorder. Biol Psychiatry. 2000;47:520-524.
42.    Insel TR, Gillin JC, Moore A, Mendelson WB, Loewenstein RJ, Murphy DL. The sleep of patients with obsessive-compulsive disorder. Arch Gen Psychiatry. 1982;39:1372-1377.
43.    Rapoport J, Elkins R, Langer DH, et al. Childhood obsessive-compulsive disorder. Am J Psychiatry. 1981;138:1545-1554.
44.    Hohagen F, Lis S, Krieger S, et al. Sleep EEG of patients with obsessive-compulsive disorder. Eur Arch Psychiatry Clin Neurosci. 1994;243:273-278.
45.    Keshavan MS, Reynolds CF, III, Kupfer DJ. Electroencephalographic sleep in schizophrenia: a critical review. Comprehensive Psychiatry. 1990;31:34-47.
46.    Benson KL, Zarcone VP, Jr. Rapid eye movement sleep eye movements in schizophrenia and depression. Arch Gen Psychiatry. 1993;50:474-482.
47.    Hudson JI, Lipinski JF, Keck PE, Jr., et al. Polysomnographic characteristics of schizophrenia in comparison with mania and depression. Biol Psychiatry. 1993;34:191-193.
48.    Zarcone VP, Benson KL, Berger PA. Abnormal rapid eye movement latencies in schizophrenia. Arch Gen Psychiatry. 1987;44:45-48.
49.    Ganguli R, Reynolds CF III, Kupfer DJ. Electroencephalographic sleep in young never-medicated schizophrenics. Arch Gen Psychiatry. 1987;44:36-44.
50.    van Kammen DP, van Kammen WB, Peters J, Goetz K, Neylan T. Decreased slow-wave sleep and enlarged lateral ventricles in schizophrenia. Neuropsychopharmacology. 1988;1:265-271.
51.    Benca RM, Casper RC. Sleep in eating disorders. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Medicine. 2nd ed. Philadelphia, Pa: W.B. Saunders Company; 1994:927-933.
52.    Delvenne V, Kerkhofs M, Appelboom Fondu J, Lucas F, Mendlewicz J. Sleep polygraphic variables in anorexia nervosa and depression: a comparative study in adolescents. J Affect Disord. 1992;25:167-172.
53.    Levy AB, Dixon KN, Schmidt H. Sleep architecture in anorexia nervosa and bulimia. Biol Psychiatry. 1988;23:99-101.
54.    Gupta MA. Sleep-related eating in bulimia nervosa: an underreported parasomnia disorder. Sleep Res. 1991;20:182.
55.    Schenck CH, Hurwitz TD, Bundlie SR, Mahowald MW. Sleep-related eating disorders: polysomnographic correlates of a heterogeneous syndrome distinct from daytime eating disorders. Sleep. 1991;14:419-431.
56.    Winkelman JW. Clinical and polysomnographic features of sleep-related eating disorder. J Clin Psychiatry. 1998;59:14-19.
57.    Spaggiari MC, Granella F, Parrino L, Marchesi C, Melli I, Terzano MG. Nocturnal eating syndrome in adults. Sleep. 1994;17:339-344.
58.    Brower KJ, Aldrich MS, Robinson EA, Zucker RA, Greden JF. Insomnia, self-medication, and relapse to alcoholism. Am J Psychiatry. 2001;158:399-404.
59.    Obermeyer W, Benca R. Effects of drugs on sleep. Otolaryngol Clin North Am. 1999;32:289-302.
60.    Gillin JC, Smith TL, Irwin M, Butters N, Demodena A, Schuckit M. Increased pressure for rapid eye movement sleep at time of hospital admission predicts relapse in nondepressed patients with primary alcoholism at 3-month follow-up. Arch Gen Psychiatry. 1994;51:189-197.
61.    Rosen C, Mahowald M, Ferber R. Sleepwalking, confusional arousals, and sleep terrors in the child. In: Ferber T, Kryer M, eds. Principles and Practices of Sleep Medicine in Children. New York, NY: W. B. Saunders; 1995:99-106.
62.    Broughton R. NREM arousal parasomnias. In: Kryer M, Roth T, Dement W, eds. Principles and Practices of Sleep Medicine. New York, NY: W. B. Saunders; 2000:693-706.
63.    Minde K, Popiel K, Leos N, Falkner S, Parker K, Handley-Derry M. The evaluation and treatment of sleep disturbances in young children. J Child Psychol Psychiatry. 1993;34:521-533.
64.    Stein MA, Mendelsohn J, Obermeyer WH, Amromin J, Benca R. Sleep and behavior problems in school-aged children. Pediatrics. 2001;107:E60.
65.    Kahn A, Van de Merckt C, Rebuffat E, et al. Sleep problems in healthy preadolescents. Pediatrics. 1989;84:542-546.
66.    Owens JA, Spirito A, McGuinn M, Nobile C. Sleep habits and sleep disturbance in elementary school-aged children. J Dev Behav Pediatr. 2000;21:27-36.
67.    Sadeh A, McGuire JPD, Sachs H, et al. Sleep and psychological characteristics of children on a psychiatric inpatient unit. J Am Acad Child Adolesc Psychiatry. 1995;34:813-819.
68.    Beneto A. Sleep disorders in the elderly. Epidemiology. Rev Neurol. 2000;30:581-586.
69.    Martin J, Shochat T, Ancoli-Israel S. Assessment and treatment of sleep disturbances in older adults. Clin Psychol Rev. 2000;20:783-805.
70.    Ancoli-Israel S. Insomnia in the elderly: a review for the primary care practitioner. Sleep. 2000;23(suppl 1):S23-S30. Discussion S36-38.
71.    Bliwise DL. Sleep in normal aging and dementia. Sleep. 1993;16:40-81.
72.    Vitiello MV, Bliwise DL, Prinz PN. Sleep in Alzheimer’s disease and the sundown syndrome. Neurology. 1992;42:83-93. Discussion 93-94.
73.    Bliwise DL, Tinklenberg J, Yesavage JA, et al. REM latency in Alzheimer’s disease. Biol Psychiatry. 1989;25:320-328.
74.    Prinz PN, Vitaliano PP, Vitiello MV, et al. Sleep, EEG and mental function changes in senile dementia of the Alzheimer’s type. Neurobiol Aging. 1982;3:361-370.
75.    Vitiello MV, Prinz PN, Williams DE, Frommlet MS, Ries RK. Sleep disturbances in patients with mild-stage Alzheimer’s disease. J Gerontol. 1990;45:M131-138.
76.    Henderson S, Jorm AF, Scott LR, Mackinnon AJ, Christensen H, Korten AE. Insomnia in the elderly: its prevalence and correlates in the general population. Med J Aust. 1995;162:22-24.
77.    Obermeyer WH, Benca RM. Effects of drugs on sleep [review]. Neurol Clin North Am. 1996;14:827-840.
78.    Bakshi R. Fluoxetine and restless legs syndrome. J Neurol Sci. 1996;142:151-152.
79.    Dorsey CM, Lukas SE, Cunningham SL. Fluoxetine-induced sleep disturbance in depressed patients. Neuropsychopharmacology. 1996;14:437-442.
80.    Schenck CH, Mahowald MW, Kim SW, O’Connor KA, Hurwitz TD. Prominent eye movements during NREM sleep and REM sleep behavior disorder associated with fluoxetine treatment of depression and obsessive-compulsive disorder. Sleep. 1992;15:226-235.
81.    Vogel GW, Buffenstein A, Minter K, Hennessey A. Drug effects on REM sleep and on endogenous depression. Neurosci Biobehav Rev. 1990;14:49-63.
82.    Bowers M, Kupfer DJ. Central monoamine oxidase inhibition and REM sleep. Brain Res. 1971;35:561-564.
83.    Wyatt RJ, Fram DH, Kupfer DJ, Snyder F. Total prolonged drug-induced REM sleep suppression in anxious-depressed patients. Arch Gen Psychiatry. 1971;24:145-155.
84.    Armitage R, Rush AJ, Trivedi M, Cain J, Roffwarg HP. The effects of nefazodone on sleep architecture in depression. Neuropsychopharmacology. 1994;10:123-127.
85.    Kupfer DJ, Reynolds CF, III, Weiss BL, Foster G. Lithium carbonate and sleep in affective disorders: further considerations. Arch Gen Psychiatry. 1974;30:79-84.
86.    Brannen JO, Jewett RE. Effects of selected phenothiazines on REM sleep in schizophrenics. Arch Gen Psychiatry. 1969;21:284-290.
87.    Kaplan J, Dawson S, Vaughan T, Green R, Wyatt RJ. Effect of prolonged chlorpromazine administration on the sleep of chronic schizophrenics. Arch Gen Psychiatry. 1974;31:62-66.
88.    Kupfer DJ, Wyatt RJ, Greenspan K, Scott J, Snyder F. Lithium carbonate and sleep in affective illness. Arch Gen Psychiatry. 1970;23:35-40.
89.    Bootzin RR, Nicassio PM. Behavioral treatments for insomnia. In: Hersen M, Eissler R, Miller P, eds. Progress in Behavior Modification. New York, NY: Academic Press; 1978:1-45.
90.    Spielman AJ, Saskin P, Thorpy MJ. Treatment of chronic insomnia by restriction of time in bed. Sleep. 1987;10:45-56.
91.    Holbrook AM, Crowther R, Lotter A, Cheng C, King D. Meta-analysis of benzodiazepine use in the treatment of insomnia. CMAJ. 2000;162:225-233.
92.    Walsh JK, Pollak CP, Scharf MB, Schweitzer PK, Vogel GW. Lack of residual sedation following middle-of-the-night zaleplon administration in sleep maintenance insomnia. Clin Neuropharmacol. 2000;23:17-21.
93.    Troy SM, Lucki I, Unruh MA, et al. Comparison of the effects of zaleplon, zolpidem, and triazolam on memory, learning, and psychomotor performance. J Clin Psychopharmacol. 2000;20:328-337.
94.    Benca RM. Eating disorders. In: Kryer M, Roth T, Dement W, eds. Principles and Practices of Sleep Medicine. New York, NY: W. B. Saunders; 2000:1169-1175.
95.    Kupfer DJ, Spiker DG, Rossi A, Coble PA, Shaw DH, Ulrich RF. Nortriptyline and EEG sleep in depressed patients. Biol Psychiatry. 1982;17:535-546.
96.    Scharf MB, Sachais BA. Sleep laboratory evaluation of the effects and efficacy of trazodone in depressed insomniac patients. J Clin Psychiatry. 1990;51(suppl):13-17.
97.    Parrino L, Spaggiari MC, Boselli M, Di Giovanni G, Terzano MG. Clinical and polysomnographic effects of trazodone CR in chronic insomnia associated with dysthymia. Psychopharmacology. 1994;116:389-395.
98.    Nierenberg AA, Adler LA, Peselow E, Zornberg G, Rosenthal M. Trazodone for antidepressant-associated insomnia. Am J Psychiatry. 1994;151:1069-1072.
99.    Roth T, Roehrs T, Koshorek G, Sicklesteel J, Zorick F. Sedative effects of antihistamines.
J Allergy Clin Immunol. 1987;80:94-98.
100.    Wirz-Justice A, Armstrong SM. Melatonin: nature’s soporific? J Sleep Res. 1996;5:137-141.
101.    MacFarlane JG, Cleghorn JM, Brown GM, Streiner DL. The effects of exogenous melatonin on the total sleep time and daytime alertness of chronic insomniacs: a preliminary study. Biol Psychiatry. 1991;30:371-376.
102.    Garfinkel D, Laudon M, Nof D, Zisapel N. Improvement of sleep quality in elderly people by controlled-release melatonin. Lancet. 1995;346:541-544.
103.    Haimov I, Lavie P, Laudon M, Herer P, Vigder C, Zisapel N. Melatonin replacement therapy of elderly insomniacs. Sleep. 1995;18:598-603.
104.    Wurtman RJ, Zhdanova I. Improvement of sleep quality by melatonin. Lancet. 1995;346:1491.
105.    Czeisler CA, Cajochen C, Turek FW. Melatonin in regulation of sleep and circadian rhythms. In: Kryer M, Roth T, Dement W, eds. Principles and Practices of Sleep Medicine. New York, NY: W.B. Saunders; 2000:693-706.
106.    Rush AJ, Armitage R, Gillin JC, et al. Comparative effects of nefazodone
and fluoxetine on sleep in outpatients with major depressive disorder. Biol Psychiatry. 1998;44:3-14.
107.    Hindmarch I, Fairweather DB. Assessing the residual effects of hypnotics. Acta Psychiatr Belg. 1994;94:88-95.
108.    Leger D, Quera-Salva MA, Philip P. Health-related quality of life in patients with insomnia treated with zopiclone. Pharmacoeconomics. 1996;10:39-44.

Dr. Brooks is staff physician and Dr. Kushida is assistant professor at the Stanford University Center of Excellence for Sleep Disorders in California. Reprinted from TEN. 2001;3(9):43-47. 



Narcolepsy is a potentially disabling neurologic disorder characterized by excessive daytime somnolence (EDS) and other symptoms, including cataplexy, sleep paralysis, hypnagogic hallucinations, and disrupted nocturnal sleep. The prevalence of narcolepsy is estimated to be .03% to .05% in the general population. The peak onset of symptoms occurs in adolescence, with a second peak at ~40 years of age. Even with treatment, narcolepsy can have a negative impact on health, social functioning, and performance at school or work. There are reasons to be optimistic that more effective treatments may be forthcoming.


History of Narcolepsy

The clinical manifestations of narcolepsy have been recognized for centuries and have been at times attributed to psychopathology, epilepsy, trauma, tumors, or infection. The history of narcolepsy has been extensively reviewed by Mignot,1 and will be summarized here. Willis described narcolepsy as early as 1672. However, Gélineau, in 1880, was the first to use the term “narcolepsy” (combining the Greek words for “somnolence” and “to seize”) to describe a clinical disorder that included irresistible sleep attacks and episodes of “astasia” (falling).

In 1902, Lowenfeld reported an association between excessive sleepiness and episodes of brief muscle weakness triggered by emotions. In 1926, Adie termed this muscle weakness “cataplexy.” In 1927, Lhermitte and Tournay described an association between narcolepsy and hypnagogic hallucinations. A short time later, Wilson reported an association between sleep paralysis and narcolepsy. Daniels, in 1934, noted the association of EDS, cataplexy, sleep paralysis, and hypnagogic hallucinations. Yoss and Daly termed this constellation of symptoms “narcoleptic tetrad” in 1957. In 1960, Vogel observed sleep-onset rapid eye movement periods (SOREMPs) in narcoleptics. Dement and Rechtschaffen later solidified the idea that disordered rapid eye movement (REM) sleep was of central importance in the pathophysiology of narcolepsy.

In the 1980s, Japanese workers discovered a link between narcolepsy and the human leukocyte antigen (HLA) DR2, suggesting that narcolepsy might represent a genetically determined autoimmune disorder.2 Although this idea remains unproven, further studies have established the concept of a genetic susceptibility to narcolepsy.

Clinical Features of Narcolepsy

Narcolepsy is a complex neurological disorder that disrupts sleep and wakefulness. In the majority of cases, symptoms of narcolepsy begin by the age of 25 years; onset occurs in approximately 10% of cases before age 10, and in 5% of cases after age 50. The disorder, affecting both genders equally, often begins gradually. EDS is usually the initial symptom, followed by other symptoms, including cataplexy, sleep paralysis, and hypnagogic hallucinations which occur over months to years. In a small number of cases, one of the other symptoms may precede excessive sleepiness. Once established, narcolepsy is not progressive, but usually persists for life, although symptoms may abate in some cases. Cataplexy, sleep paralysis, and hypnagogic hallucinations are more likely to disappear than excessive sleepiness.

The primary symptom of narcolepsy is EDS. Patients may report an ongoing feeling of drowsiness, or they may suffer from sleepiness episodes of variable duration. As in normal individuals, the sleepiness of the narcoleptic tends to be worse in the afternoon, in warm environments, and in passive or boring situations. Narcoleptic drowsiness, however, may become irresistible, leading to unwanted periods of sleep during the daytime, often at inappropriate times (such as during a meal, during a conversation, or while driving). This overwhelming urge to sleep may occur very suddenly, leading to what have been called “sleep attacks.” Periods of automatic behavior, a reflection of brief intrusions of sleep (“microsleeps”) into the drowsy state, may also occur. Suboptimal alertness may lead to secondary symptoms, such as difficulties with concentration and memory. Duration of daytime sleep episodes varies from minutes to an hour, and patients often feel transiently refreshed from even brief naps. Not only does excessive somnolence place the narcoleptic at increased risk of accidental injury, but it can also have a serious negative impact on family and social situations and on performance at school or work.

Up to 70% of narcoleptics experience cataplexy, which represents sudden, transient loss or reduction of skeletal muscle tone. Attacks of cataplexy are often triggered by an emotional stimulus (especially joking, laughter, or anger) or even by a memory of an emotionally charged event. Other triggers include stress, meals, and fatigue, but episodes may also occur without obvious provocation. The phenomenon of cataplexy may be subtle or dramatic, ranging from brief sagging of the jaw or knee buckling, collapse, or even injury. The severity of the attack may be maximal at onset, or the weakness may evolve over seconds, sequentially affecting various body parts. The duration of cataplexy episodes varies from a few seconds to half an hour, with episodes commonly lasting no longer than a few seconds to minutes. Consciousness and memory are maintained, although patients may report hallucinations or dream-like imagery during prolonged episodes.

Sleep paralysis refers to episodes of voluntary muscle paralysis occurring at sleep onset or upon awakening. Up to 40% to 50%3 of normal individuals report isolated sleep paralysis at least once in a lifetime (recurring episodes are much less common). The phenomenon recurs in up to 40% of narcoleptics. Episodes of sleep paralysis last for 1 to several minutes and resolve spontaneously or in response to external stimulation, such as the touch of another person. Usually, all voluntary muscles are involved, except those controlling eye movements and respiration. Hypnagogic imagery may occur concomitantly, but the sensorium is generally clear. Understandably, the experience may provoke acute anxiety.

Narcoleptics commonly report hallucinations at sleep onset (hypnagogic) or sleep offset (hypnopompic). These are often described as “waking dreams.” Generally, the hallucinations are visual or auditory, but tactile hallucinations or feelings of movement (such as levitating) are not uncommon. The visual hallucinations may be elaborate (such as animals or people) or simple (changing shapes and colors). Auditory hallucinations may take the form of sounds, words (the perception of one’s name being called is a common experience), or music.

Although one might expect that patients with EDS would be able to sleep soundly at will, this is not the case in narcolepsy. The major sleep period is usually disrupted. Narcoleptics encounter difficulty with sleep initiation or, more commonly, sleep maintenance. The total amount of sleep per 24-hour period is not increased in narcolepsy. The problem is that sleep is inappropriately distributed across the period, with intrusions in the daytime and poor consolidation of nocturnal sleep.

Genetics of Narcolepsy

Honda and colleagues2 first described the associations between narcolepsy and HLA DR2 and HLA DQ6 in the Japanese population. The association has been confirmed in 96% of whites with narcolepsy.4 The incidence of HLA DR2 varies among ethnic groups, with a lower incidence in African Americans. HLA DQ6 (and more particularly DQB1*0602) is a more sensitive marker for narcolepsy across ethnic groups. DQB1*0602 occurs more often in narcoleptics with cataplexy than in those without cataplexy.4,5 There is also a positive correlation between DQB1*0602 positivity and severity of cataplexy.

Most cases of narcolepsy are sporadic, but there are numerous reports of familial occurrence of narcolepsy. The risk of development of narcolepsy with cataplexy in first-degree relatives is 1% to 2% (10–40 times that in the general population). A larger percentage (4% to 5%) of relatives have isolated daytime somnolence. Several DQB1*0602-negative families have been identified, in which narcolepsy with cataplexy appears to be transmitted in an autosomal-dominant pattern with high penetrance.6 These cases may stem from a genetic mutation. Sporadic cases of DQB1*0602-negative narcolepsy with clear-cut cataplexy have also been reported, but these are unusual.

Although the strong HLA associations suggest that an autoimmune process may be responsible for narcolepsy, this has not been established. Overall, the evidence suggests that DQB1*0602 and DQA1*0102 confer susceptibility to narcolepsy.4 It is likely that other predisposing genes will be identified.

The Discovery of the Hypocretin/Orexin Peptides

In 1998, de Lecea and colleagues7 reported finding a hypothalamic-specific mRNA encoding the precursor of a pair of peptides homologous to secretin. They named the peptides hypocretin 1 and hypocretin 2 (Hcrt1 and Hcrt2) to denote their hypothalamic specificity and their resemblance to secretin. In the same year, Sakurai and colleagues8 identified two neuropeptides that bound and activated two related orphan G protein-coupled receptors. These peptides were found to stimulate food intake when administered centrally to rats. Thus, the investigators called them orexin A and orexin B (from the Greek “orexis,” meaning “appetite”). Later, it became clear that the orexins and the hypocretins were identical. Both of the hypocretins bind to two G protein-coupled receptors (Hcrtr 1 and Hcrtr 2), although hypocretin 2 has low affinity for Hcrtr 1. The cell bodies of hypocretin-producing neurons reside in the hypothalamus. They have dense projections within the hypothalamus but project widely to many other brain areas as well, most densely to the locus ceruleus.9 Four major pathways of hypocretin projection have been identified, two projecting toward the cortex and two toward the brainstem. The two descending pathways impinge on structures well known to be involved in sleep-wake regulation and occurrence of rapid eye movement (REM) sleep. The generous distribution of the hypocretin system suggests that these neuropeptides contribute to multiple physiologic functions, such as food intake, thermoregulation, endocrine function, cardiovascular regulation, and the sleep-wake cycle.9 The neuronal group is small and may not contain more than 10,000–15,000 neurons, suggesting a regulatory role for this newly discovered brain system.

New Discoveries in Animal Models of Narcolepsy

There are strains of dogs affected with narcolepsy and/or cataplexy. The canine syndrome is similar to the one observed in humans, and symptoms begin during the equivalent of early adolescence. Systematic investigation of a large dog colony initially led to identification of the chromosome containing the gene responsible for canine narcolepsy. Later, it was demonstrated that canine narcolepsy is due to a mutation of the gene coding for the hypocretin 2 receptor.10 Around the same time, other researchers observed narcoleptic-like behavior in preprohypocretin knockout mice.11

Hypocretin/Orexin in Human Narcoleptics

Nishino and colleagues12 hypothesized that human narcolepsy involves a disruption in hypocretin neurotransmission. They measured cerebrospinal fluid hypocretin in nine narcoleptic patients with cataplexy and eight control subjects. Hcrt1 was detected in all control samples. In seven of nine patients, hypocretin levels were below the assay’s limits of detection. In two patients with unquestionable narcolepsy-cataplexy, hypocretin was detectable; one patient’s level was similar to that of the control subjects, and the other’s level was elevated. These results demonstrate a deficiency in hypocretin function in some patients with narcolepsy, possibly due to a defect in hypocretin production. The patients with detectable hypocretin levels were clinically indistinguishable from the other narcoleptic patients. These two cases may reflect an abnormality of the effector-receptor interaction rather than a lack of hypocretin production. More recently, pathologic studies of brains of narcoleptics (compared with those of age-matched control subjects), performed simultaneously at Stanford and UCLA, have demonstrated absence of hypocretin neurons in the hypothalamus.13,14

These observations in human narcoleptics establish obvious links between narcolepsy and the hypocretin system. These findings, along with the those from canine and murine models, suggest that the hypocretin system is of central importance in the development of narcolepsy and have opened new pathways of inquiry into its pathophysiology.

Pathophysiology of Narcolepsy

The pathogenesis of human narcolepsy remains unclear. It is apparent, however, that narcolepsy represents a complex process. Association with specific HLA alleles and increased prevalence in first-degree relatives suggests a genetic basis. That genetic factors alone are insufficient to explain the disorder, is supported by the infrequent familial cases and the low concordance rate (25% to 30%) of narcolepsy between identical twins.6 Although still attractive, the autoimmune hypothesis remains unproven. Narcolepsy is not associated with oligoclonal bands in cerebrospinal fluid or with typical peripheral markers of autoimmune disease, such as autoantibodies, or elevations of erythrocyte sedimentation rate or C-reactive protein. The clinical expression of the disorder likely depends on the interplay between one or more genetic factors and environmental triggers. There are also numerous case reports of secondary or “symptomatic” narcolepsy, including some with cataplexy. Such cases have been associated with head trauma, stroke, multiple sclerosis, brain tumor, neurodegenerative disorders, and central nervous system infections.

How might the hypocretin system fit into the clinical framework of narcolepsy? Although the neuroanatomic and neurophysiologic underpinnings of human narcolepsy are incompletely defined, one of the disorder’s primary features seems to be abnormal regulation of REM sleep. The phenomena of cataplexy, sleep paralysis, and hypnagogic hallucinations all appear to represent inappropriate intrusions of REM sleep physiology into wakefulness. REM sleep onset is generated by cholinergic neurons in the pons and opposed by monoaminergic neurons in the locus ceruleus and dorsal raphe. Normally, locus ceruleus cells are continuously active during wakefulness but cease firing prior to and during cataplexy and REM sleep. Hypocretin neurons project densely to the locus ceruleus and are excitatory. It is reasonable to suppose that a functional defect in hypocretin neurotransmission might diminish opposition of REM sleep onset, thereby allowing its initiation at inappropriate times.

Hypocretin neurons also project to brain regions known to be important in producing and sustaining arousal. Alteration of neurotransmission in these areas due to a defect in the hypocretin system might help to explain the excessive somnolence that is the hallmark of narcolepsy.

Diagnosis of Narcolepsy

In idiopathic narcolepsy, results of physical and neurologic examinations are normal. Neurologic abnormalities may be present in secondary forms of the disorder, depending on the responsible brain lesion, but findings are generally nonspecific. Taking a patient history is of crucial importance in diagnosing narcolepsy. Patients may relate symptoms in broad or imprecise terms, and it is important for the examiner to differentiate complaints of genuine sleepiness from other symptoms such as physical tiredness or fatigue. EDS is a common symptom encountered in medical practice and is not specific to narcolepsy. EDS may arise from many causes, including sleep deprivation, sleep disruption, licit and illicit drugs, and medical and psychiatric diseases. In most of these cases, the overwhelming “sleep attacks” common in narcolepsy do not occur. Unlike narcoleptics, patients with many of these other conditions do not find brief naps to be rejuvenating. Sleep paralysis and hypnagogic (or hypnopompic) hallucinations, common in narcoleptics, may also occur in normal individuals under certain conditions. The most helpful symptom in diagnosing narcolepsy is clear-cut cataplexy. Some experts even insist that the presence of cataplexy is a prerequisite for the diagnosis of narcolepsy.

Polysomnographic studies are also essential in confirming the diagnosis. Overnight recordings usually demonstrate shortening of REM sleep latency, as well as disruption of sleep architecture with multiple awakenings. Results of the Multiple Sleep Latency Test (MSLT) are abnormal, with mean sleep latencies usually less than 5 minutes and the occurrence of SOREMPs. According to the International Classification of Sleep Disorders,3 the minimal criteria for diagnosing narcolepsy in the absence of cataplexy include EDS, associated features (sleep paralysis, hypnagogic hallucinations, disrupted nocturnal sleep, automatic behavior), mean sleep latency on MSLT of less than 5 minutes, and at least two SOREMPs. Others15 recommend that in cases of EDS without cataplexy, a descriptive diagnosis (eg, “EDS with multiple SOREMPs”) is preferable to use of the term “narcolepsy.” It is important to remember that cataplexy may present up to several years after the onset of EDS; in such cases, the diagnosis becomes clear over time.

Overall, HLA typing is of limited usefulness in diagnosing narcolepsy, because the subtypes of interest occur not infrequently in normal individuals, and the HLA associations are strongest in individuals with cataplexy (who also pose the least diagnostic difficulty). Although uncommon, patients with cataplexy who are negative for HLA DQB1*0602 have been reported. Measurement of hypocretin in cerebrospinal fluid may prove to be useful in diagnosing difficult cases. At the present time, neuroimaging studies are not helpful in diagnosing idiopathic narcolepsy.

Treatment of Narcolepsy

The treatment of narcolepsy now consists of medications, education, support, and behavioral changes. The traditional medical treatment of narcolepsy includes the use of stimulants for excessive sleepiness and antidepressants for the manifestations of disordered REM sleep. The relationship between pharmacology and narcolepsy has been reviewed by some (see Nishino and Mignot16). Two new compounds have been extensively studied, and one of them, modafinil, is currently commercially available.

Modafinil is considered more of a “somnolytic” than a “stimulant” medication.17,18 In adults, it has been found to be less efficacious than amphetamine-like drugs, once subjects have already been treated with the latter, and it would be better to consider it for initial treatment in newly diagnosed cases.19 Modafinil does not control cataplexy and may need to be given with anticataplectic medications if this symptom is an important clinical problem. Because amphetamines have some anticataplectic activity, there may be some rebound cataplexy when switching from amphetamines to modafinil.19 The side effects of modafinil are usually mild at the recommended dosage. Headache is the most common and can generally be avoided if the drug is started at 100 mg in the morning and progressively increased for 3–4 days. The recommended daily dosage is 300–400 mg, given in two divided doses in the morning and at lunch time. Some sleep centers have prescribed modafinil in doses up to 600 mg daily, but the higher doses have produced increased side effects without any gain in alertness.

The use of modafinil combined with traditional stimulants has not been well studied. A review of 22 such cases in our clinic (modafinil 400–500 mg/day and methylphenidate 20 mg in divided doses) suggested a better response with combined medications than with one drug alone. We have not seen the combination of amphetamine with modafinil, and we believe that such a combination is unwise until the mechanism of action of modafinil is better defined. The drug has some effect on dopamine reuptake20,21 and may also act at hypothalamic sites subserving wakefulness.22 More definitive work is needed to define modafinil’s mode of action. The advantages of using low doses of methylphenidate are its quick onset of action and rapid elimination.

None of the therapeutic trials of amphetamines, methylphenidate, or modafinil have demonstrated levels of alertness, as measured by the MSLT or the Maintenance of Wakefulness Test (MWT), similar to those in control subjects.23 The great advantage of modafinil is that it does not produce the many side effects seen with amphetamines, including the vasopressor effect, at the recommended dosage.24 None of the regulatory agencies or drug companies have sponsored a study in children despite the fact that the peak age of onset of narcolepsy is around puberty, and that the combination of puberty, hormonal changes, and narcolepsy places young teenagers at the greatest jeopardy with respect to sleepiness. Teenagers with narcolepsy do not respond to treatment as well as older adults. While it may be tempting to treat teenaged narcoleptics with amphetamines, it is better to avoid this approach, considering the long-term side effects.

For many years, γ-hydroxybutyrate (GHB), also known as sodium oxybate, has been reported to be helpful in patients with narcolepsy.25-27 However, GHB’s mode of action is unclear. Initial work was done in France and Canada, and for years, narcoleptic patients have traveled from the United States to Canada to obtain the medication. GHB has been used recreationally and may be abused, as is the case with some other agents used to treat narcolepsy.28 This has delayed the drug’s approval by US regulatory agencies. Orphan Medical has performed clinical trials with doses of 3–9 g/day. In the patients studied at Stanford, GHB produced increased slow-wave sleep, a finding observed previously in narcoleptic patients who were obtaining the drug in Canada. GHB’s duration of action is short; it is given at bedtime, and a second dose must be taken during the night.

Prior investigations of GHB have demonstrated improved nocturnal sleep with decreased fragmentation in adult narcoleptic subjects.26 Although US drug trials have focused on its beneficial effect on cataplexy, results obtained in foreign countries have indicated a slow but progressive improvement in daytime alertness, which narcoleptic patients regard as the drug’s most important benefit. This may not occur until the agent has been used for several weeks, suggesting that there is a slow reorganization of sleep-wake control over time. Although narcolepsy has been described as a “disease of REM sleep,” there is ample evidence that nonrapid eye movement (NREM) sleep is drastically affected in narcoleptics, and restoration of a normal balance of nocturnal NREM and REM sleep may be one of the benefits of GHB. Preliminary studies performed at Stanford indicate that a small group of patients scored better on MSLT and MWT following daily use of GHB for at least 1 month, than with use of other currently used medications. These interesting preliminary results will require confirmation with larger studies.

Side effects of GHB are partially dose  dependent. Over time, there also appears to be adaptation over time to some side effects. Confusion and disorientation have been reported on awakening within 2 hours of nocturnal drug intake, raising concerns about risk of falling in elderly narcoleptics who awake to urinate during the night. Short-term enuresis has been observed, and abnormal levels of excitation have been associated with high dosage and chronic intake. Publication of the results of the large US trial performed during the past several months will provide important information about GHB. At present, it is not known if, or when, the Food and Drug Administration will approve the use of GHB. In any event, it is likely that physicians will encounter increasing numbers of narcoleptic patients taking the medication on their own, with supplies obtained from foreign countries or Internet sources.

Future Directions

As our understanding of narcolepsy continues to deepen, new and more effective methods for its treatment are likely to develop. Based on recent findings, the development of hypocretin agonists offers theoretical promise. There are potential problems with this approach, however, considering the broad and numerous physiologic roles that the hypocretin system appears to play. Perhaps hypocretin-receptor subtypes will be discovered, which might allow for fine-tuning of the system with more specific hypocretin analogues. Implants of hypocretin-producing cells might also be possible. An understanding of the process responsible for the loss of hypocretin cells in the first place would undoubtedly be helpful. In any event, one cannot help but be optimistic about the prospects for improved narcolepsy treatments, considering the recent dramatic advances in our understanding of this disorder.  PP


1.    Mignot E. A hundred years of narcolepsy research. Arch Ital Biol. 2001;139:207-220.
2.    Honda Y, Juji T, Matsuki K, et al. HLA-DR2 and Dw2 in narcolepsy and in other disorders of excessive somnolence without cataplexy. Sleep. 1986;9:133-142.
3.    International Classification of Sleep Disorders, Revised: Diagnostic and Coding Manual. Rochester, Minn: American Sleep Disorders Association; 1997.
4.    Mignot E, Ling Lin, Rogers R, et al. Complex HLA-DR and DQ interactions confer risk for narcolepsy-cataplexy in three ethnic groups. Am J Hum Genet. 2001;68:686-699.
5.    Mignot E, Hayduk R, Black J, et al. HLA DQB1*0602 is associated with cataplexy in 509 narcoleptic patients. Sleep. 1997;20:1012-1020.
6.    Mignot E. Genetic and familial aspects of narcolepsy. Neurology. 1998;50(suppl 1):16-22.
7.    de Lecea L, Kilduff TS, Peyron C, et al. The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A. 1998;95:322-327.
8.    Sakurai T, Amemiya A, Ishii M, et al. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell. 1998;92:573-585.
9.    Peyron C, Tighe D, van der Pol A, et al. Neurons containing hypocretin (orexin) project to multiple neuronal systems. J Neurosci. 1998;18:9996-10015.
10.    Lin L, Faraco J, Li R, et al. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell. 1999;98:365-376.
11.    Chemelli RM, Willie JT, Sinton CM, et al. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell. 1999;98:437-451.
12.    Nishino S, Ripley B, Overeem S, et al. Hypocretin (orexin) deficiency in human narcolepsy [letter]. Lancet. 2000;355:39-40.
13.    Peyron C, Faraco J, Rogers W, et al. A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nature Med. 2000;6:991-997.
14.    Thannickal TC, Moore RY, Nienhuis R, et al. Reduced number of hypocretin neurons in human narcolepsy. Neuron. 2000;27:469-474.
15.    Moscovitch A, Partinen M, Patterson N, et al. Cataplexy in differentiation of excessive daytime somnolence. Sleep Res. 1991;20:301.
16.    Nishino S, Mignot E. Pharmacological aspects of human and canine narcolepsy. Prog Neurobiol. 1997;52:27-78.
17.    US Modafinil in Narcolepsy Multicenter Study Group. Randomized trial of modafinil as a treatment for the excessive daytime somnolence of narcolepsy. Neurology. 2000;54:1166-1175.
18.    Modafinil for narcolepsy. Med Lett Drugs Ther. 1999;41:30-31.
19.    Guilleminault C, Aftab FA, Karadeniz D, Philip P, Leger D. Problems associated with switch to modafinil—a novel alerting agent in narcolepsy. Eur J Neurol. 2000;7:381-384.
20.    Mignot E, Nishino S, Guilleminault C, et al. Modafinil binds to the dopamine uptake carrier site with low affinity. Sleep. 1994;17:436-437.
21.    Wisor JP, Nishino S, Sora I, Uhl GH, Mignot E, Edgar DM. Dopaminergic role in stimulant-induced wakefulness. J Neurosci. 2001;21:1787-1794.
22.    Scammell TE, Estabrooke IV, McCarthy MT, et al. Hypothalamic arousal regions are activated during modafinil-induced wakefulness. J Neurosci. 2000;20:8620-8628.
23.    Mitler MM, Aldrich MS, Koob GF, Zarcone VP. Narcolepsy and its treatment with stimulants. Sleep. 1994;17:352-371.
24.    Mitler MM, Harsh J, Hirshkowitz M, Guilleminault C. Long-term efficacy and safety of modafinil (Provigil) for the treatment of excessive daytime sleepiness associated with narcolepsy. Sleep Med. 2000;1:231-243.
25.    Broughton R, Mamelack M. The treatment of narcolepsy cataplexy with nocturnal gamma-hydroxybutyrate. Can J Neurol Sc. 1979;12:1-6.
26.    Mamelak M, Scharf MB, Woods M. Treatment of narcolepsy with gamma-hydroxybutyrate. A review of clinical and sleep laboratory findings. Sleep. 1986;9:285-289.
27.    Lammers GJ, Arends J, Declerck AC, et al. Gammahydroxybutyrate and narcolepsy: a double-blind placebo-controlled study. Sleep. 1993;16:216-220.
28.    Boyce SH, Padgham K, Miller LD, Stevenson J. Gamma hydroxybutyric acid (GHB): an increasing trend in drug abuse. Eur J Emerg Med. 2000;7:177sw-181.

Dr. Rogers is research associate and Dr. Dinges is professor of psychology in psychiatry at the Unit for Experimental Psychiatry in the Division of Sleep and Chronobiology at the University of Pennsylvania School of Medicine in Philadelphia.

Acknowledgments: This work was supported in part by NIH grant NR04281, NASA Cooperative Agreement NCC 9-58 with the National Space Biomedical Research Institute, and AFOSR grant 49620-95-1-0388. Reprinted from TEN. 2001;3(9):58-64. 



In today’s global society with work schedules placed across the 24-hour day, shiftwork, particularly night shiftwork, is becoming increasingly common. Despite the almost universal acceptance of the necessity of night shiftwork, the full impact of these schedules on workers is still not fully appreciated. Both short- and long-term effects of night shiftwork have been described, in retrospective, epidemiological, and laboratory and field studies investigating sleep deprivation. Increases in certain disorders, including cardiovascular disease and reproductive problems, have been reported as long-term consequences of shiftwork. In the short term, individuals required to work at night and sleep during the day experience disruption in their circadian systems, due to misalignment with the external environment and time cues, as well as chronic sleep restriction. Potential treatments for this circadian disruption include strategies to promote synchrony and rapid adaptation between the internal body clock and external environment (eg, melatonin, bright-light therapy) and wakefulness during the night shift (eg, napping, wake-promoting compounds).


The Circadian System

Circadian rhythms are biological rhythms that are endogenously produced by an organism, characterized by repetitive oscillations having a frequency of one cycle approximately every 24 hours,1 and sustained in constant environmental conditions (for a review see Czeisler and Khalsa2). In humans the endogenous circadian pacemaker, or biological clock, is located in the suprachiasmatic nuclei (SCN), which are situated in the anterior hypothalamus (for a review see Weaver3). The circadian system oversees the timing of a range of physiological processes, including those controlled by the endocrine, thermoregulatory, immune, cardiovascular, respiratory, gastrointestinal, renal, and sleep-wake systems.

Environmental time cues that are able to entrain the biological clock are termed zeitgebers; the most salient zeitgeber in many animals including humans is the light-dark cycle.2 In addition to entraining the circadian system to maintain an appropriate phase relationship with the environment, zeitgebers are able to shift the timing of the biological clock and, consequently, the circadian rhythms of the functions under its control.

Markers of the Circadian System


Melatonin is an indoleamine produced in the pineal gland. The production of melatonin is under the control of sympathetic innervation of the pineal gland. At night, in the absence of light, melatonin synthesis in the pineal gland is stimulated by signals from neurons originating in the SCN (for a review see Arendt4). The SCN receive light information from the eyes, and send information via the spinal cord to paraventricular nuclei and superior cervical ganglia, which terminate near the pineal gland. During the daylight hours, melatonin synthesis is inhibited due to suppression of activity in the sympathetic neurons coming from the SCN. Onset of melatonin secretion typically occurs between 10pm and 12am.

It is important to note, however, that while the circadian rhythm of melatonin secretion is regulated by the SCN, and thus is strongly influenced by the light-dark cycle, in the absence of a light-dark cycle (ie, in dark:dark), circadian rhythmicity of melatonin secretion is conserved. Additionally, exposure to a new light-dark cycle (eg, during night shiftwork) does not immediately shift the secretory profile of melatonin to a new light-dark cycle phase. Rather, this process is gradual, and hence may take a number of days to weeks to be completed. Although melatonin secretion occurs across the sleep period and has been proposed as a trigger for sleep,5,6 studies have demonstrated that melatonin secretion occurs irrespective of sleep-wake status, even in the absence of nocturnal sleep.7 Consequently, changes in circadian timing can alter the timing of melatonin secretion, independent of sleep-wake state or sleep deprivation.

Melatonin may function as a timekeeper for the circadian system, keeping the pacemaker informed about changes in day length and time of day. In addition, as a circadian-mediated humoral factor circulating in the bloodstream, melatonin may be responsible for relaying information from the circadian pacemaker to the physiological systems that are regulated by the biological clock. Therefore, melatonin may play a role in coordinating various circadian rhythms within the body.

Core Body Temperature

Another robust marker of the circadian system is core body temperature. In a normally entrained person, core body temperature increases gradually throughout the day, and decreases at night, continuing across the sleep period, reaching a nadir in the early hours of the morning (between 4am and 6am). Core body temperature typically begins rising prior to nocturnal sleep offset and continues to increase across the day.

Circadian Disruption

Circadian disruption occurs as a result of misalignment in the timing of circadian rhythms in relation to either the external light-dark cycle or to each other. Circadian disruption is evident in transmeridian travelers, and contributes significantly to the group of maladies collectively referred to as “jet lag,” which is experienced annually by millions of people who fly across time zones. In addition, night-shift workers experience a disruption of the circadian system. The difference between jet lag and night-shift work is that in the former, zeitgebers are available to help the circadian system re-entrain to a new rest-activity phase, while in night-shift work, zeitgebers reinforce the endogenous circadian phase that is out of symmetry with the rest-activity phase required by the night shift. As a result, jet lag is typically transient, lasting 1–14 days, depending on the number of time zones crossed, while circadian disruption from night work can last much longer.

Indeed, as will be explained below, many night-shift workers never fully adjust their endogenous phase to align with the requirement to be awake and alert at night. Circadian realignment can be experimentally induced by using appropriately timed exposure to bright light, by administering melatonin, or by altering the timing or duration of the light-dark cycle. However, to be effective, these interventions must be administered at the appropriate circadian phase, and hence, a reasonable estimate of phase position must be obtained. Consequently, this may limit their utility in some situations.


Shiftwork is defined as any work conducted outside the standard Monday–Friday, 8am–6pm work day. Recent estimates suggest that approximately 15% to 20% of the workforce in industrialized countries can be considered shiftworkers.8-10 There are several detrimental effects associated with working irregular shifts. Epidemiological studies show a high occurrence of several chronic disorders, including cardiovascular complaints and reproductive difficulties, in shiftwork populations. More immediate effects are also evident. These include decreased sleep quantity and quality, decreased neurobehavioral functioning, and increased fatigue, sleepiness, irritability, and illness. These effects are due to a number of factors, including circadian disruption.

Chronic Health Problems

Several researchers have investigated the long-term effects of shiftwork and health. Koller11 described a dose-response relationship between number of shiftwork years and risk for cardiovascular disease in 270 oil refinery workers. Later, Knutsson and colleagues12 reported an association between an increased risk for ischemic heart disease and shiftwork in 504 male blue-collar workers from a paper and pulp manufacturing plant in Sweden. This finding was reported to be independent of both age and cigarette smoking. Knutsson and colleagues13 then reported an increased risk of myocardial infarction in men and women shiftworkers, relative to controls. Again, this finding was independent of age and cigarette smoking, and was also not related to job strain or job educational level. In contrast, Tenkanen and colleagues14 reported an increased risk of coronary heart disease (CHD) in 1,806 Finnish shiftworkers that was significantly associated with smoking and obesity. A similar association was not evident in day workers. The authors suggest that shiftwork may be a trigger for CHD in individuals who have other lifestyle factors that place them at increased risk for CHD.

Shiftwork and health in female workers have also been investigated. Japanese15,16 and French studies17 reported an increased incidence of irregular menstrual cycles in women involved in shiftwork. In addition, shiftwork has been associated with reduced fertility, as assessed by the number of menstrual cycles required for a planned pregnancy to occur.15,18-20

Several studies have reported a relationship between the frequency of spontaneous abortions and shiftwork. Bisanti and colleagues19 reported that night work and three-shift schedules may be related to an increased incidence in spontaneous abortions in a group of Swedish midwives. Further, a number of studies have reported an increased incidence of spontaneous abortion in women working nonstandard work hours.15,21-25 Three of these studies stated that spontaneous abortion was associated with some variety of rotating schedule.22,23,25

In a Canadian study, Fortier and colleagues26 reported an increased risk of preterm births in mothers employed in regular evening or night work. This elevated risk appeared in women who continued with irregular shiftwork following week 23 of their pregnancies. Xu and colleagues27 reported an association between preterm births and mothers involved in rotating shiftwork in Chinese textile workers, and McDonald and colleagues28 reported an inconsistent association between preterm births and shiftwork in Canadian workers. In contrast, Saurel-Cubizolles and Kaminski29 reported no relationship between preterm births and shiftwork in a group of French shiftworking mothers.

Circadian Misalignment

When working during the night and sleeping during the day, shiftworkers are attempting to be active during their inactive phase and vice versa. The circadian systems of shiftworkers generally do not adapt well to the change in working time, usually due to the short duration of the night-shift period and the time taken for the body to adapt. Consequently, shiftworkers can be considered to be in a state of almost constant circadian misalignment. This point is illustrated in a study by Sack and colleagues,30 who monitored melatonin secretion across the 24-hour day in a group of night-shift workers who had been on a shift for 2–5 nights prior to the study. In eight of the nine shiftworkers studied, melatonin secretion occurred at an innappropriate time versus that in normally entrained subjects. The time of melatonin onset for these workers ranged between the hours of 10:28pm and 12:30am. A study of subjects in Antarctica revealed that, in the absence of sunlight, potential synchronizers such as social contact and mealtimes were unable to entrain circadian rhythms.31

Although circadian re-entrainment may be difficult to achieve under many conditions of night work, it may not be impossible. Barnes and colleagues32 examined adaptation to a 2-week night-shift period in oil rig workers, using 6-sulphatoxy melatonin (aMT.6S) excretion as a circadian marker. Urine samples were taken at 2- to 3-hour intervals during wake hours and one sample was collected across the sleep period, in both winter and summer. Adaptation of the aMT.6S rhythm to the new work and sleep schedule occurred within a week during both summer and winter. This phase shift was achieved via a phase delay of the aMT.6S acrophase. The authors suggest that possible entraining mechanisms include forced reversal of work and sleep schedules, timing of meals, and exposure to artificial lighting and sunlight, particularly during the summer.

Potential Treatments for Circadian Misalignment


Appropriately timed exposure to bright light can act as an artificial zeitgeber, and hence, has been suggested as a treatment for circadian disruption. One of the most widely studied areas of the application of light for this purpose is in shiftworkers. This has been accomplished in the field using shiftworkers, and in the laboratory, where normal subjects are placed on “shiftwork schedules.” For example, in a laboratory-based study, Dawson and Campbell33 placed subjects on a day-sleep/night-wake schedule, and investigated the effects of bright-light exposure on sleep, circadian phase (core body temperature), and alertness. Two groups of subjects completed three consecutive sessions of night work, followed by daytime sleep. Both groups of subjects were maintained in dim light, with one group (treatment group) receiving a 4-hour light pulse from 12am–4am on the first night. In both groups, the timing of the temperature nadir was phase-delayed after the night shifts, relative to baseline. Compared with the control group, the light-treatment group achieved a significantly greater delay (355±43 versus 143±60 minutes). Coincident with these changes in the timing of the temperature rhythm, improved sleep efficiency and increased alertness levels were reported for the treatment group. Such findings suggest that appropriately timed bright-light administration to night workers accelerates circadian adaptation to a new sleep-wake schedule, with coincident improvements in sleep and alertness.

Eastman34 reported similar findings following light exposure during a 12-hour shift of the sleep-wake cycle, for a longer period of time (≥8 night shifts). In nearly 90% of subjects, the core temperature rhythm was shifted following appropriately timed bright-light exposure during the first 4 days of the schedule. Light was also administered on the following days, to maintain the new phase position. In all subjects, the average daily shift in temperature was approximately 2 hours. In the subjects who produced a phase delay, there was complete adaptation of the temperature curve to the new sleep-wake schedule by day 7.

While these bright-light-induced phase shifts appear positive, they are not dissimilar to those reported in shiftworkers with no active intervention.32 Barnes and colleagues32 reported that the timing of the aMT.6S rhythm in oil rig workers had entrained to a 12-hour phase shift in their sleep-wake cycle within 7 days. This adaptation period is similar to that reported above by Eastman for the core temperature rhythm following exposure to bright light.34 Barnes and colleagues suggest that the entrainment of the circadian system to the new sleep-wake schedule was likely due in part to exposure to both artificial and natural light.

Therefore, in real life, workers may be able to adjust their circadian systems relatively rapidly without having to use specific bright-light treatments. Alternatively, it may be that an interaction between a number of time markers (eg, meals or forced reversal of work and sleep times, and light exposure) facilitates circadian adaptation.

Gallo and Eastman35 studied the effects of bright-light exposure on circadian adaptation during rapidly rotating changes in the sleep-wake schedule. Sleep times were progressively delayed across 5 days by a total of 10 hours, maintained at this time for 5 days, and then progressively returned to the initial timing.

The timing of the temperature rhythm did not shift equally in all subjects. Some subjects demonstrated a greater phase shift, and their core temperature rhythm approached synchronization to their new sleep-wake schedule. Other subjects did not shift their temperature rhythm to such an extent, and one subject even appeared to free run. These data demonstrate a high degree of intersubject variability in bright-light-induced resynchronization of the core temperature circadian rhythm following large phase shifts.

The authors suggested that the type of illumination used might be a reason why the temperature rhythm did not fully adapt to the new sleep-wake schedules. It is important for subjects to face the light source and be at an optimal distance from it for maximum light intensity to reach the pacemaker.36 While this may be a factor in the results obtained, it may also represent a more real-life situation. Typically, shiftworkers would not be exposed to optimal intensity of light for the correct duration at the appropriate time. Therefore, the results of this study may illustrate what is likely to occur in the real world, and not just in the laboratory.

The magnitude of the acute effects of light on melatonin suppression and temperature elevation can vary depending on the intensity of the light administered.37 Similarly, it is possible that the magnitude of the delayed (phase-shifting) effects of light may also be affected by light intensity. Martin and Eastman38 compared the phase-shifting effects of three different light intensities: low (<250 lux), medium (~1,250 lux), and high (~5,700 lux) for 3 hours/day light intensities for 5 out of 6 nights of work with daytime sleep.38 Following exposure to the high-intensity light, temperature rhythms in all subjects were delayed, such that the nadir occurred during the day-sleep period (ie, full adaptation). This effect was also observed in most (85%) of the subjects exposed to medium-light intensity. In contrast, fewer subjects (42%) exposed to low-light intensity had a temperature rhythm shift of this amount. Therefore, it appears that both timing of light exposure and intensity are important.

Exogenous Melatonin Administration

Exogenous melatonin administration has been proposed to treat persons with circadian disruption disorders, eg, shiftworkers, and with delayed sleep phase syndrome.39,40 Attenburrow and colleagues41 administered melatonin to a group of healthy subjects at 5pm, and reported a significant phase advance in endogenous melatonin rhythm.41 This advance was evident after 7 days of melatonin administration, with no effect after a single day of administration.

In a laboratory-based study, Dawson and colleagues42 compared the effectiveness of light and melatonin in producing phase shifts in subjects on an enforced shiftwork schedule. Following light treatment, a significant phase delay (8.8 hours) was evident. In contrast, the delay following melatonin administration was approximately half (4.7 hours). This delay was not significantly different from that found in the placebo condition (4.2 hours).

Sack and colleagues43 have suggested that lack of response to melatonin may be due to timing of administrations (split dose: two administrations). When comparing the time of administration with the phase response curve for melatonin,44,45 Dawson and colleagues42 apparently administered one dose of melatonin on the advance portion of the phase response curve and the other during the delay portion. Therefore, Sack and colleagues43 concluded that any phase-shifting effect may have been cancelled out.

Folkard and colleagues46 compared the effects of melatonin and placebo on sleep and alertness in a group of night shiftworkers. Subjects took the treatments at bedtime (during the diurnal portion of the circadian cycle). Melatonin increased self-reported sleep quality and duration, relative to placebo and baseline. There was no effect of melatonin on sleep-onset latencies. Melatonin also reportedly increased early-morning alertness. Hypothetically, this increase may have been due to a phase shift in the alertness rhythm, but the authors conclude that this would be unlikely. It may be that increased sleep duration and sleep quality increased alertness levels.

Sack and colleagues47 also noted that shiftworkers demonstrated a high degree of variability in their ability to phase shift, both with and without melatonin treatment. Some subjects shifted after taking placebo, but not after taking melatonin. Other subjects demonstrated an equivalent delay following administration of placebo and melatonin, while a small number of subjects did not shift following either treatment.

This high degree of intersubject variability highlights the difficulties in assessing the effectiveness of melatonin treatment. Much experimental evidence appears to suggest that, in order for melatonin to be effective, regimens that are personalized with regard to timing, dose, and route of administration, are required.

Several studies have reported that exogenous melatonin administration at doses of ≥5 mg produces significant decrements in neurobehavioral performance.48-56 This finding is particularly important if melatonin is to be administered as a chronobiotic rather than a soporific agent. Under these circumstances, it is possible that melatonin would be administered at a time other than directly prior to bedtime, when activities such as driving or operating machinery may be undertaken.

Several research groups have reported the absence of hangover effects following a night’s sleep when melatonin was taken.51,54-56 Taken together, these findings suggest that, if melatonin were administered as a soporific agent, it would be unlikely to produce unwanted neurobehavioral effects. However, if melatonin were administered as a chronobiotic agent during the day, eg, in shiftworkers and transmeridian travelers, these adverse performance effects could be problematic.

In addition, treatment with melatonin and bright-light administration are aimed at improving the adaptation of circadian rhythms to a new sleep-wake schedule, via manipulation of the circadian system. While some of these methods have demonstrated efficacy, the time required for reversal of the phase misalignment is often greater than the time of exposure to the new schedule. Consequently, these treatments are unlikely to be successful in the management of circadian maladjustment in shiftworkers who do not remain on new sleep-wake schedules for substantial periods of time (eg, 2 weeks or more).

Wake-Promoting Compounds

Treatment regimens for shiftwork have begun to move toward more immediate-acting therapies, such as wake-promoting compounds and napping strategies.


A number of studies have examined caffeine’s effectiveness in night-shift workers in promoting neurobehavioral functioning and alertness during sleep deprivation. Rosenthal and colleagues57 reported significant alerting effects of caffeine 75 mg and 150 mg following both a normal (8 hours) and a restricted (5 hours) night’s sleep. Penetar and colleagues58 showed significant alerting effects of caffeine 150 mg, 300 mg, and 600 mg using subjective and objective measures of alertness during a 49-hour period of sustained wakefulness.58 Kamimori and colleagues59 had similar findings after administering low (2.1 mg/kg), medium (4.3 mg/kg), or high (8.6 mg/kg) doses of caffeine during 49 hours of sustained wakefulness.

In subjects deprived of sleep for 45.5 hours, Wright and colleagues60 reported that caffeine 200 mg, administered at 8pm and 2am, enhanced neurobehavioral performance. They used a battery of performance tasks, including reaction time, vigilance and memory tasks, and alertness, assessed using the Maintenance of Wakefulness Test. Caffeine coupled with bright-light exposure, from 8pm–8am, produced the greatest improvement in neurobehavioral performance and alertness, relative to a control condition (dim light plus placebo). The greatest improvement in alertness and performance was observed during the early morning hours, around the time of the circadian nadir.

Lagarde and colleagues61 assessed the efficacy of three doses (150 mg, 300 mg, and 600 mg) of slow-release caffeine on wakefulness and performance, using questionnaires, the multiple sleep latency test (MSLT), and performance tasks including attention, grammatical reasoning, tracking, and memory, across 32 hours of sleep deprivation. Significant increases in MSLT sleep-onset times and decreased subjective sleepiness were reported for the three caffeine doses. This finding is consistent with increased nocturnal sleep latencies noted by Sicard and colleagues62 following administration of 600-mg caffeine using a slow-release formulation. In addition, there was an increase in performance on some neurobehavioral measures relative to placebo, which lasted for 13 hours after administration.


Dinges and colleagues63 investigated the effects of 2-hour naps during 56 hours of sustained operations. The naps took place at the circadian peak and trough, following 6, 18, 30, 42, and 54 hours of wakefulness. Neurobehavioral performance and subjective mood were assessed throughout. There was a significant improvement in neurobehavioral performance following the naps, but no effect on subjective measures of mood. The benefit of naps on neurobehavioral performance may not be apparent immediately if the nap occurs prior to an accumulation of sleep loss, ie, a prophylactic nap. In contrast, it appears that the duration of the effect may be limited (up to 12 hours) if the nap occurs following a period of sleep loss.

Therefore, despite a reported benefit on performance following all the nap periods, the timing of a nap relative to the amount of sleep loss appears to be an important determinant of a nap’s effect on neurobehavioral functioning. Hence, it would appear that prophylactic naps, despite having both lower sleep quantity and quality versus naps occurring later in the deprivation period, provide significant benefits in neurobehavioral functioning for many hours following waking.

Naps Plus Caffeine

A small number of studies have examined the combined effects of caffeine and naps on alertness and neurobehavioral functioning. Bonnet and Arand64 reported a greater improvement in alertness and performance following a 4-hour afternoon nap combined with nocturnal caffeine administration relative to four 1-hour naps across the night. They also reported positive effects on neurobehavioral functioning and alertness during 24 hours of sleep deprivation when prophylactic naps were used in conjunction with caffeine administration.65

Dinges and colleagues66-68 investigated the effects of 88 hours of total sleep deprivation, with and without two 2-hour naps in 24 hours, and with and without sustained low-dose caffeine (0.3 mg/kg/hour) for 66 hours. Caffeine, without naps, reduced the frequency of psychomotor vigilance task (PVT) lapses for up to 22 hours following the first administration (at 22 hours of sleep deprivation), compared with the placebo/no-nap condition. The positive effects of caffeine on neurobehavioral functioning lasted for approximately 22 hours, coinciding with the rising slope of the plasma caffeine pharmacokinetic curve. The greatest benefits in neurobehavioral functioning were evident in the subjects who received caffeine combined with naps. In addition, in these subjects, caffeine was apparently able to significantly attenuate the effects of sleep inertia on neurobehavioral functioning following the 2-hour naps.68


Modafinil is currently approved for use in the management of excessive sleepiness in patients with narcolepsy, a central nervous system disorder characterized by excessive daytime sleepiness, and other symptoms.69 In addition, recent studies have examined modafinil’s wake-promoting effects during sleep deprivation and simulated shiftwork settings.

While its exact mechanism of action has yet to be fully described, modafinil appears to be a novel wake-promoting agent that is chemically and pharmacologically distinct from other central nervous system psychostimulants (eg, amphetamine) and lacks their more serious adverse effects (eg, locomotor agitation). In a randomized, double-blind, crossover study comparing the subjective effects of modafinil (300 mg) to dextroamphetamine (15 mg), caffeine (300 mg), and placebo, the subjective effects of modafinil were similar to those of caffeine, and markedly different from those of dextroamphetamine.70

A number of studies have investigated modafinil’s effectiveness in maintaining alertness in the face of excessive sleepiness due to pathological disorders (eg, narcolepsy) and sleep deprivation (eg, as experienced by shiftworkers). In rested, nonsleep-deprived subjects, morning administration of modafinil 200 mg was reported to enhance vigilance recorded by an electroencephalogram, daytime sleep latency, concentration, complex reactions, subjective alertness, and performance on search and memory tests.71

Two recent studies investigated the effects of modafinil during 64-hour total sleep deprivation.72,73 Pigeau and colleagues73 compared modafinil 300 mg to dextroamphetamine 20 mg or placebo in a double-blind, crossover trial. Both modafinil and dextroamphetamine decreased subjective fatigue and sleepiness and improved reaction time, logical reasoning, and short-term memory performance.73 Although modafinil generally improved all aspects of performance in this study, the drug has also been associated with mild adverse effects on the efficiency of communication74 and subjective estimates of cognitive capability.75

Sleep polysomnography before and after drug administration and sleep deprivation demonstrated that modafinil was associated with fewer sleep disturbances compared to amphetamine.76 Modafinil-treated subjects were better able to sleep than those given amphetamine, and their sleep closely resembled that of subjects in the placebo group. Additionally, the amount of recovery sleep required following 64 hours of sustained wakefulness appeared to be less in those subjects receiving modafinil. This finding is consistent with results from animal work by Edgar and Seidel77 showing that, unlike other psychostimulants, modafinil reduces rebound sleepiness and could reduce the requirement for recovery sleep.

Modafinil’s effect on sleepiness associated with four 24-hour periods of simulated night shiftwork was assessed recently in a controlled laboratory environment.78 During night shifts, subjects were randomized to receive either modafinil 200 mg or placebo. Placebo-treated subjects reported increased subjective feelings of sleepiness and demonstrated significantly attenuated performance on the PVT. In contrast, subjects receiving modafinil reported significantly higher subjective alertness, with a coincident improvement in performance on the PVT. In addition, modafinil did not significantly disrupt daytime sleep periods.

Modafinil’s positive effects on neurobehavioral functioning do not appear to be associated with changes in circadian physiology (Process C)79 or in circadian-mediated hormones (melatonin, cortisol, growth hormone). Rather, its effects appear to involve changes in waking and sleeping processes that reflect the homeostatic drive for sleep (Process S).79 This is evidenced by the fact that modafinil not only reverses physiological, behavioral, and subjective sleepiness, but it does so without disrupting recovery sleep following sleep deprivation. It actually appears to attenuate the need for slow-wave sleep and prolonged recovery sleep following deprivation.76


Shiftwork produces a range of effects on a number of physiological systems, in both the short and long term. Epidemiological studies report decreased health in shiftworkers, including increased cardiovascular complaints and reproductive dysfunctions in night workers. In addition, immediate effects on sleep and neurobehavioral functioning may be evident after only 1 night of shiftwork. Potential treatments for deficits in neurobehavioral functioning include those directed at improving adaptation of the circadian system to new sleep-wake schedules (eg, melatonin and bright light), and more acutely, at improving alertness and neurobehavioral performance directly (eg, naps and caffeine). PP


1.    Czeisler CA, Duffy JF, Shanahan TL, et al. Stability, precision, and near-24-hour period of the human circadian pacemaker. Science. 1999;284:2177-2181.
2.    Czeisler CA, Khalsa SBS. The human circadian timing system and sleep-wake regulation. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia, Pa: W.B. Saunders Company; 2000:353-376.
3.    Weaver DR. The suprachiasmatic nucleus: a 25-year retrospective. J Biol Rhythms. 1998;13:100-112.
4.    Arendt J. Melatonin and the pineal gland: influence on mammalian seasonal and circadian physiology. Rev Reprod. 1998;3:13-22.
5.    Myers BL, Badia P. Changes in circadian rhythms and sleep quality with aging: mechanisms and interventions. Neurosci Biobehav Rev. 1995;19:553-571.
6.    Dawson D, Encel N. Melatonin and sleep in humans. J Pineal Res. 1993;15:1-12.
7.    Morris M, Lack L, Barrett J. The effect of sleep/wake state on nocturnal melatonin excretion. J Pineal Res. 1990;9:133-138.
8.    Australian Bureau of Statistics. Working Arrangements Australia; 1993. Report no. Cat. 63420.
9.    U.S. Congress Office of Technology Assessment. Biological Rhythms: Implications for the Worker; 1991. Report no. OTA-BA-463.
10.    Wedderburn A. Bulletin of European Studies on Time. Dublin, Ireland: European Foundation for the Improvement of Living and Working Conditions; 1993.
11.    Koller M. Health risks related to shiftwork. Int Arch Occup Environ Health. 1983;53:59-75.
12.    Knutsson A, Akerstedt T, Jonsson BG, Orth-Gomer K. Increased risk of ischaemic heart disease in shift workers. Lancet. 1986;2(8498):89-92.
13.    Knutsson A, Hallquist J, Reuterwall C, Theorell T, Akerstedt T. Shiftwork and myocardial infarction: a case-control study. Occup Environ Med. 1999;56:46-50.
14.    Tenkanen L, Sjoblom T, Harma M. Joint effect of shift work and adverse life-style factors on the risk of coronary heart disease. Scand J Work Environ Health. 1998;24:351-357.
15.    Uehata T, Sasakawa N. The fatigue and maternity disturbances of night workwomen. J Hum Ergol. 1982;11(suppl):465-474.
16.    Miyauchi F, Nanjo K, Otsuka K. Effects of night shift on plasma concentrations of melatonin, LH, FSH and prolactin, and menstrual irregularity. Japanese J Ind Health. 1992;34:545-550.
17.    Messing K, Saurel-Cubizolles MJ, Bourgine M, Kaminski M. Menstrual-cycle characteristics and work conditions of workers in poultry slaughterhouses and canneries. Scand J Work Environ Health. 1992;18:302-309.
18.    Ahlborg G Jr., Axelsson G, Bodin L. Shift work, nitrous oxide exposure and subfertility among Swedish midwives. Int J Epidemiol. 1996;25:783-790.
19.    Axelsson G, Ahlborg G, Bodin L. Shift work, nitrous oxide exposure, and spontaneous abortion among Swedish midwives. Occup Environ Med. 1996;53:374-378.
20.    Bisanti L, Olsen J, Basso O, Thonneau P, Karmaus W. Shift work and subfecundity: a European multicenter study. European Study Group on Infertility and Subfecundity. J Occup Environ Med. 1996;38:352-358.
21.    Axelsson G, Lutz C, Rylander R. Exposure to solvents and outcome of pregnancy in university laboratory employees. Br J Ind Med. 1984;41:305-312.
22.    Axelsson G, Rylander R. Outcome of pregnancy in women engaged in laboratory work at a petrochemical plant. Am J Ind Med. 1989;16:539-545.
23.    Hemminki K, Kyyronen P, Lindbohm ML. Spontaneous abortions and malformations in the offspring of nurses exposed to anaesthetic gases, cytostatic drugs, and other potential hazards in hospitals, based on registered information of outcome. J Epidemiol Commun Health. 1985;39:141-147.
24.    Infante-Rivard C, David M, Gauthier R, Rivard GE. Pregnancy loss and work schedule during pregnancy. Epidemiology. 1993;4:73-75.
25.    McDonald AD, McDonald JC, Armstrong B, et al. Fetal death and work in pregnancy. Br J Indust Med. 1988;45:148-157.
26.    Fortier I, Marcoux S, Brisson J. Maternal work during pregnancy and the risks of delivering a small-for-gestational-age or preterm infant. Scand J Work, Environ Health. 1995;21:412-418.
27.    Xu X, Ding M, Li B, Christiani DC. Association of rotating shiftwork with preterm births and low birth weight among never smoking women textile workers in China. Occup Environ Med. 1994;51:470-474.
28.    McDonald AD, McDonald JC, Armstrong B, et al. Prematurity and work in pregnancy. Br J Ind Med. 1988;45:56-62.
29.    Saurel-Cubizolles MJ, Kaminski M. Pregnant women’s working conditions and their changes during pregnancy: a national study in France. Br J Ind Med. 1987;44:236-243.
30.    Sack RL, Blood ML, Lewy AJ. Melatonin rhythms in night shift workers. Sleep. 1992;15:434-441.
31.    Kennaway DJ, Van Dorp CF. Free-running rhythms of melatonin, cortisol, electrolytes, and sleep in humans in Antarctica. Am J Physiol. 1991;260 (6 Pt 2):R1137-R1144.
32.    Barnes RG, Deacon SJ, Forbes MJ, Arendt J. Adaptation of the 6-sulphatoxymelatonin rhythm in shiftworkers on offshore oil installations during a 2-week 12-h night shift. Neurosci Lett. 1998;241:9-12.
33.    Dawson D, Campbell SS. Timed exposure to bright light improves sleep and alertness during simulated night shifts. Sleep. 1991;14:511-516.
34.    Eastman CI. High-intensity light for circadian adaptation to a 12-h shift of the sleep schedule. Am J Physiol. 1992;263(2 Pt 2):R428-R436.
35.    Gallo LC, Eastman CI. Circadian rhythms during gradually delaying and advancing sleep and light schedules. Physiol Behav. 1993;53:119-126.
36.    Dawson D, Campbell SS. Bright light treatment: are we keeping our subjects in the dark? Sleep. 1990;13:267-271.
37.    Trinder J, Armstrong SM, O’Brien C, Luke D, Martin MJ. Inhibition of melatonin secretion onset by low levels of illumination. J Sleep Res. 1996;5:77-82.
38.    Martin SK, Eastman CI. Medium-intensity light produces circadian rhythm adaptation to simulated night-shift work. Sleep. 1998;21:154-165.
39.    Arendt J, Deacon S. Treatment of circadian rhythm disorders—melatonin. Chronobiol Int. 1997;14:185-204.
40.    Arendt J. Jet-lag and shift work: (2). Therapeutic use of melatonin. J R Soc Med. 1999;92:402-405.
41.    Attenburrow ME, Dowling BA, Sargent PA, Sharpley AL, Cowen PJ. Melatonin phase advances circadian rhythm. Psychopharmacology. 1995;121:503-505.
42.    Dawson D, Encel N, Lushington K. Improving adaptation to simulated night shift: timed exposure to bright light versus daytime melatonin administration. Sleep. 1995;18:11-21.
43.    Sack RL, Hughes RJ, Edgar DM, Lewy AJ. Sleep-promoting effects of melatonin: at what dose, in whom, under what conditions, and by what mechanisms? Sleep. 1997;20:908-915.
44.    Lewy AJ, Ahmed S, Jackson JM, Sack RL. Melatonin shifts human circadian rhythms according to a phase-response curve. Chronobiol Int. 1992;9:380-392.
45.    Lewy AJ, Bauer VK, Ahmed S, Thomas KH, Cutler NL, Singer CM, et al. The human phase response curve (PRC) to melatonin is about 12 hours out of phase with the PRC to light. Chronobiol Int. 1998;15:71-83.
46.    Folkard S, Arendt J, Clark M. Can melatonin improve shift workers’ tolerance of the night shift? Some preliminary findings. Chronobiol Int. 1993;10:315-320.
47.    Sack RL, Lewy AJ. Melatonin as a chronobiotic: treatment of circadian desynchrony in night workers and the blind. J Biol Rhythms. 1997;12:595-603.
48.    Rogers NL, Phan O, Kennaway DJ, Dawson D. Effect of daytime oral melatonin administration on neurobehavioral performance in humans.
J Pineal Res. 1998;25:47-53.
49.    Lieberman HR, Waldhauser F, Garfield G, Lynch HJ, Wurtman RJ. Effects of melatonin on human mood and performance. Brain Res. 1984;323:201-207.
50.    Wynn VT, Arendt J. Effect of melatonin on the human electrocardiogram and simple reaction time responses. J Pineal Res. 1988;5:427-435.
51.    Waldhauser F, Saletu B, Trinchard-Lugan I. Sleep laboratory investigations on hypnotic properties of melatonin. Psychopharmacology. 1990;100:222-226.
52.    Dollins AB, Lynch HJ, Wurtman RJ, Deng MH, Kischka KU, Gleason RE, et al. Effect of pharmacological daytime doses of melatonin on human mood and performance. Psychopharmacology. 1993;112:490-496.
53.    Dollins AB, Zhdanova IV, Wurtman RJ, Lynch HJ, Deng MH. Effect of inducing nocturnal serum melatonin concentrations in daytime on sleep, mood, body temperature, and performance. Proc Nat Acad Sci U S A. 1994;91:1824-1828.
54.    Deacon S, Arendt J. Melatonin-induced temperature suppression and its acute phase-shifting effects correlate in a dose-dependent manner in humans. Brain Res. 1995;688:77-85.
55.    Zhdanova IV, Wurtman RJ, Lynch HJ, et al. Sleep-inducing effects of low doses of melatonin ingested in the evening. Clin Pharmacol Ther. 1995;57:552-558.
56.    Zhdanova IV, Wurtman RJ, Morabito C, Piotrovska VR, Lynch HJ. Effects of low oral doses of melatonin, given 2-4 hours before habitual bedtime, on sleep in normal young humans. Sleep. 1996;19:423-431.
57.    Rosenthal L, Roehrs T, Zwyghuizen-Doorenbos A, Plath D, Roth T. Alerting effects of caffeine after normal and restricted sleep. Neuropsychopharmacology. 1991;4:103-108.
58.    Penetar D, McCann U, Thorne D, et al. Caffeine reversal of sleep deprivation effects on alertness and mood. Psychopharmacology. 1993;112:359-365.
59.    Kamimori GH, Penetar DM, Headley DB, Thorne DR, Otterstetter R, Belenky G. Effect of three caffeine doses on plasma catecholamines and alertness during prolonged wakefulness. Eur J Clin Pharmacol. 2000;56:537-544.
60.    Wright KP Jr, Badia P, Myers BL, Plenzler SC. Combination of bright light and caffeine as a countermeasure for impaired alertness and performance during extended sleep deprivation.
J Sleep Res. 1997;6:26-35.
61.    Lagarde D, Batejat D, Sicaed B, Trocherie CD, Enslen M, Chauffard F. Slow-release caffeine: a new response to the effects of a limited sleep deprivation. Sleep. 2000;23:651-661.
62.    Sicard BA, Perault MC, Enslen M, Chauffard F, Vandel B, Tachon P. The effects of 600 mg of slow release caffeine on mood and alertness. Aviat Space Environ Med. 1996;67:859-862.
63.    Dinges DF, Whitehouse WG, Carota Orne E, Orne MT. The benefits of a nap during prolonged work and wakefulness. Work Stress. 1988;2:139-153.
64.    Bonnet MH, Arand DL. Impact of naps and caffeine on extended nocturnal performance. Physiol Behav. 1994;56:103-109.
65.    Bonnet MH, Arand DL. The use of prophylactic naps and caffeine to maintain performance during a continuous operation. Ergonomics. 1994;37:1009-1020.
66.    Dinges DF, Doran SM, Mullington J, et al. Neurobehavioral effects of 66 hr of sustained low-dose caffeine during 88 hr of total sleep deprivation. Sleep. 2000;23(suppl 1):A20-A21.
67.    Rogers NL, Price NJ, Szuba MP, Van Dongen HP, Dinges DF. Effect of sustained caffeine on core body temperature during 88 hours of sustained wakefulness. Sleep. 2001;24(suppl 1). In press.
68.    Van Dongen HP, Doran SM, Mullington JM, et al. Sustained low-dose caffeine administration reduces sleep inertia after nap sleep during 88h extended wakefulness. Sleep. 2000;23(suppl 1):A120-A121.
69.    Lafon LL. Application Dossier Modafinil. France; 1998.
70.    Warot D, Corruble E, Payan C, Weil JS, Puech AJ. Subjective effects of modafinil, a new central adrenergic stimulant in healthy volunteers: a comparison with amphetamine, caffeine and placebo. Eur Psychiatry. 1993;8:201-208.
71.    Lyons TJ, French J. Modafinil: the unique properties of a new stimulant. Aviat Space Environ Med. 1991;62:432-435.
72.    Baranski JV, Cian C, Esquivie D, Pigeau RA, Raphel C. Modafinil during 64 hr of sleep deprivation: dose-related effects on fatigue, alertness, and cognitive performance. Milit Psychol. 1998;10:173-193.
73.    Pigeau R, Naitoh P, Buguet A, et al. Modafinil, d-amphetamine and placebo during 64 hours of sustained mental work. I. Effects on mood, fatigue, cognitive performance and body temperature. J Sleep Res. 1995;4:212-228.
74.    Bard EG, Sotillo C, Anderson AH, Thompson HS, Taylor MM. The DICEM map task corpus: Spontaneous dialogue under sleep deprivation and drug treatment. Speech Commun. 1996;20:71-84.
75.    Baranski JV, Pigeau RA. Self-monitoring cognitive performance during sleep deprivation: effects of modafinil, d-amphetamine and placebo. J Sleep Res. 1997;6:84-91.
76.    Buguet A, Montmayeur A, Pigeau R, Naitoh P. Modafinil, d-amphetamine and placebo during 64 hours of sustained mental work. II. Effects on two nights of recovery sleep. J Sleep Res. 1995;4:229-241.
77.    Edgar DM, Seidel WF. Modafinil induces wakefulness without intensifying motor activity or subsequent rebound hypersomnolence in the rat. J Pharmacol Exp Ther. 1997;283:757-769.
78.    Hughes RJ, Van Dongen HPA, Dinges DF, et al. Modafinil improves alertness and performance during simulated night work. Sleep. 2001;24(suppl 1):A200.
79.    Borbely AA. A two process model of sleep regulation. Hum Neurobiology. 1982;1:195-204.

Dr. Quan is director of the Sleep Disorders Center, associate director of the Arizona Respiratory Center, and professor of medicine and anesthesiology at the University of Arizona College of Medicine in Tucson. Reprinted from TEN. 2001;3(9):39-42.



Sleep is a required biological function that outwardly appears to be a simple process, but is in actuality quite complex. There are over 80 recognized conditions that can affect sleep and waking behavior. The most common is insomnia, which affects up to 30% of the general population. However, obstructive sleep apnea is the sleep disorder that usually prompts a patient’s referral to a sleep disorders center. These centers offer comprehensive evaluation and treatment in a single location. Accreditation by the American Academy of Sleep Medicine offers assurance that certain standards related to medical care, facilities, equipment, and technician staffing have been met. Polysomnography, the most common sleep laboratory procedure, is usually performed to determine if a person has obstructive sleep apnea. Although evaluation and treatment in a sleep disorders center are expensive, most insurance carriers provide reimbursement for polysomnography. Sleep disorders centers are costly to develop and maintain, but can be economically viable and beneficial to patients.



Sleep, as defined by The American Heritage Dictionary,1 is “a natural, periodically recurring physiologic state of rest, characterized by relative physical and nervous inactivity, unconsciousness, and lessened responsiveness to external stimuli.” This relatively simple definition belies the complexity of a physiologic function that occupies approximately one third of our time. Far from being a homogeneous state, a night of sleep generally consists of four sequential units containing episodes of nonrapid eye movement (NREM) and rapid eye movement (REM) sleep (Figure). During polysomnography (simultaneous monitoring of several physiologic parameters during sleep), REM sleep is distinguished from NREM sleep by rapid movements of the eyes. Awakening from REM sleep is associated with dream recall. Although most metabolic functions decline with onset of sleep, there is a paradoxic increase in these activities during REM sleep.


Sleep, by its very nature, is tightly intertwined with wakefulness. Alterations in sleep affect wakefulness and vice versa. Although the requirement for sleep is a biologic imperative, the purpose of it remains unknown.2 Some propose that the function of sleep is restorative. During sleep, there is active synthesis of proteins, which is important for regeneration of stores depleted during wakefulness. Others suggest an ethological or conservation theory: Because we live in an environment with periodic cycles of changing light and temperature, species survival would be enhanced if our activity cycle were synchronized to these cycles.

Most adults require 7–8 hours of sleep per night. However, many individuals can function well with less than this amount, and others require more. Nevertheless, as one becomes increasingly sleep deprived, the biological urge to sleep becomes more irresistible.

Epidemiology of Sleep Disorders

Given the complexity of sleep, it is not surprising that symptoms of poor sleep or excessive daytime sleepiness are widespread. In surveys of the adult general population, the prevalence rate of insomnia complaints ranges from 17.4% to 37.8%, with higher levels in women and the elderly.3 Excessive daytime sleepiness is common as well, with rates ranging from 4% to 12%, and higher percentages in older individuals.4,5 In addition to female gender and advancing age, other commonly reported risk factors for poor sleep include snoring and chronic medical and mental health problems.3 Sleep problems are not limited to adults. One recent study in children 3–14 years of age found insomnia in 16.8% and excessive daytime sleepiness in 4.0%.6

Types of Sleep Disorders

In its diagnostic and coding manual, the American Academy of Sleep Medicine lists over 80 disorders of sleep and waking behavior.7 Most of these adversely impact individual quality of life. Some are potentially life-threatening to the affected individual and can endanger public safety by causing motor vehicle or industrial accidents.

The most common sleep disorder is insomnia: Over 30% of American adults may be affected. Insomnia refers to trouble initiating or maintaining sleep. Insomnia can develop as a result of a number of medical, social, or environmental factors. Many occurrences are transient and directly attributable to an obvious cause. In some chronic cases, however, there does not appear to be any apparent predisposing factor. In these situations, the condition is classified as psychophysiologic insomnia.

In addition to sleep impairment, insomnia may produce significant problems with daytime functioning. Insomniacs have greater difficulty with concentration and coping with minor problems; they also have more motor vehicle accidents.8 Although many self-medicate with over-the-counter or complementary and alternative medications, most insomniacs do not seek medical treatment from their physicians.9

Although insomnia is the most prevalent sleep disorder, obstructive sleep apnea (OSA) is the one that most commonly results in a medical evaluation. OSA is characterized by apnea, or cessation of breathing, during sleep. The oropharyngeal structures collapse during inspiration, resulting in an obstructed airway. In addition to having such episodes witnessed by a bed partner, loud snoring and excessive daytime sleepiness are the other prominent symptoms. OSA occurs frequently: The estimated prevalence is 4% in middle-aged men and 2% in middle-aged women, with higher rates in the elderly.10 Children can also have OSA and it can contribute to poor school performance and hyperactivity.11,12 The excessive daytime sleepiness associated with OSA may be quite severe and as a result, individuals with OSA have a high rate of motor vehicle accidents.13 Studies have linked OSA as a causal factor in the development of hypertension, coronary heart disease, and stroke.14-16

Given that the prevalence of coronary heart disease is ~5% and that sleep disorders adversely impact quality of life and mortality, identification and treatment of sleep disturbances should be an important public health mandate. Unfortunately, this is not the case. It was not until 1992 that the National Commission on Sleep Disorders Research highlighted the inadequacy of patient and professional education and research in sleep in the United States. This led to the creation of the National Center on Sleep Disorders Research within the National Institutes of Health, with a mandate to conduct and support research and educational activities related to sleep disorders. Nevertheless, because recognition and treatment of sleep disorders by primary care providers remains deficient, thus, sleep disorders centers play an important role in the delivery of health care.

Historical Evolution and Function of Sleep Disorders Centers

Sleep disorders centers provide a centralized location where patients with sleep problems can receive comprehensive evaluation and treatment. The concept of the sleep disorder center developed from the visionary leadership of Dr. William Dement at Stanford University, who along with several other clinicians and sleep investigators, founded the Association of Sleep Disorders Centers (ASDC) in 1975.17 These pioneers thought that diagnosis and treatment of sleep disorders should be performed at a central location where there was specific expertise and interest. The nascent ASDC sparked further interest and publicity in sleep disorders, resulting in an exponential growth in the number of sleep disorders centers. They are now found not only in academic medical centers, but also in community hospitals in both large- and medium-sized population centers.

All sleep disorders centers provide comprehensive evaluation and treatment services for the entire spectrum of sleep disorders. However, the most common problem evaluated at most centers is OSA.18 This represents an evolution from 20 years ago when the number of centers was much smaller and the diversity of patients evaluated was greater.19

Generally, an evaluation begins with a history and physical examination obtained by a sleep specialist in a clinic or office setting. For some sleep problems, such as restless legs syndrome, the history and physical examination may be sufficiently diagnostic so that no further extensive testing is required and treatment can be started. In many other cases, additional diagnostic tests are needed. Questionnaires that help evaluate the patient’s degree of sleepiness or assess the presence of depression, and sleep logs, which provide an indicator of a patient’s sleep pattern, are often used. Polysomnography is the diagnostic test used most often.

Polysomnography is performed in the sleep laboratory of sleep disorders centers. Patients are asked to sleep in the laboratory during their normal sleep time and simultaneous recordings of electroencephalogram, electrooculogram, chin electromyogram, electrocardiogram, leg electromyogram, ventilatory function, and oxygen saturation are obtained. Originally, these parameters were recorded simultaneously on an analog chart recorder. However, most laboratories now use computerized digital acquisition systems for both display and data processing.

Polysomnography is usually performed when the diagnosis of sleep apnea is being considered, but it also provides useful information in the evaluation of other sleep disorders. It is often performed when nasal continuous positive airway pressure (CPAP) is being titrated to treat sleep apnea. In CPAP, above-ambient airway pressure is applied using a nasal mask. This produces pneumatic splinting of the upper airway and prevents obstruction. The amount of pressure required varies from patient to patient. Therefore, it is current clinical practice to increase the level of CPAP until sleep disruption, snoring, and apneic events are abolished. This requires concurrent polysomnography.

Two other tests commonly performed in sleep laboratories are the Multiple Sleep Latency Test and the Maintenance of Wakefulness Test. Both provide an objective measure of sleepiness. The Multiple Sleep Latency Test consists of four or five sessions during the day, each separated by 2 hours, during which the patient is asked to fall asleep with electroencephalographic, electrooculographic, and chin electromyographic monitoring. The average time taken to fall asleep is the mean sleep latency and an objective index of daytime sleepiness. The structure of the Maintenance of Wakefulness Test is similar to the Multiple Sleep Latency Test, but patient is asked to stay awake instead of to fall asleep. Thus, the Maintenance of Wakefulness Test provides a measure of daytime vigilance.

After all diagnostic testing is completed, sleep disorders centers provide a diagnostic impression and recommendations for therapy. Therapeutic interventions may include medication, behavioral strategies, oral appliances, and nasal CPAP. Although all sleep disorders centers have the capability to provide treatment and follow-up, insurance contracts and long-distance travel for patients often dictate that therapeutic regimens be initiated by the patient’s primary care provider.

Most sleep disorders centers provide comprehensive evaluation and treatment for all sleep problems. Some, however, only evaluate and treat a narrow spectrum of sleep problems—most commonly, sleep-disordered breathing or sleep apnea. Still others focus on pediatric sleep disorders.

Many sleep disorders centers, especially those affiliated with an academic healthcare system, are actively engaged in clinical research. Pharmaceutical companies sponsor a large proportion of this research, which usually involves clinical testing of new therapeutic compounds for various sleep disorders. Often, patients evaluated at these sleep disorders centers are offered the opportunity to participate in clinical trials. Some centers are also involved in research funded by federal agencies such as the National Institutes of Health and the Department of Defense. These projects are usually more focused on the pathophysiology of various sleep disorders.

Accreditation and Organizational Structure of Sleep Disorders Centers

The American Academy of Sleep Medicine (www.aasmnet.org), which is the successor to the ASDC and the principal professional society representing sleep clinicians and researchers, accredits sleep disorders centers and specialty laboratories meeting its standards.2 These standards represent what the academy believes are the minimum required for quality care in the evaluation and treatment of sleep disorders. The standards contain requirements for facilities, equipment, and technician staffing. A major requirement is that each center must have a diplomat of the American Board of Sleep Medicine (www.absm.org) on staff, to insure that the center’s clinical evaluations occur under a sleep medicine specialist’s general oversight.

Sleep disorders centers and specialty laboratories may voluntarily apply for accreditation. After an application fee, the accreditation process involves review of the center’s materials, to see how the center or laboratory meets the academy’s standards, and a site visit by a team of sleep specialists. The duration of accreditation is for 5 years, after which reaccreditation is required. While there are currently over 400 centers and laboratories accredited by the American Academy of Sleep Medicine, there are many unaccredited facilities. Some of these facilities function very similarly to accredited centers, while others are sleep centers in name only, providing laboratory testing with varying degrees of supervision by sleep medicine professionals. In these facilities, the sleep specialist generally does not see the patient prior to testing and treatment is not provided.

The physical facility for a sleep center must provide an area where patients can be interviewed and examined by a clinician, and a sleep laboratory where polysomnography can be performed. In some small facilities, the bedrooms function as examination rooms during the daytime. In others, patients are interviewed in clinics or offices separate from the sleep laboratory. Most sleep centers have at least two bedrooms, and some have four or more. There should be a sufficient number of technologists to perform the polysomnograms. The American Academy of Sleep Medicine recommends a ratio of no more than two patients to one polysomnographic technologist.20 Most centers employ at least one technologist who has passed the registry examination administered by the Board of Registered Polysomnographic Technologists (www.brpt.org).

Sleep Center Economics

The costs of developing and running even a small sleep disorders center can be substantial. The first requirement is a location with adequate space for the sleep laboratory. Many centers are “free standing,” ie, not located within a medical center, and rent payments can be substantial. Second, the monitoring equipment is a significant capital expense. The cost to equip a center with two bedrooms may approach $100,000 if the newest technology is purchased. Third, there are personnel costs. Polysomnographic technologists earn $25,000–$45,000 per year, depending on the location. In addition, there are costs for secretarial/reception and billing services. For institutions that lack the expertise to develop or run their own center, consulting firms can be contracted to assist in development and maintenance. In addition, there are companies specializing in sleep laboratory management.

Despite the high initial start-up costs and the ongoing expenses of managing a sleep center, the number of both accredited and unaccredited facilities has been growing. This growth has been fueled by a demand for services and by insurance carriers’ recognition that evaluation and treatment of sleep disorders are medically indicated and reimbursable. Medicare covers polysomnography for the diagnosis of sleep apnea, narcolepsy, parasomnias, and male impotence, but not for insomnia. Billing for sleep center physician professional charges related to outpatient visits is accomplished using standard Current  Procedural Terminology evaluation and management codes. Fees for polysomnography, MSLT, and MWT testing are billed as outpatient procedures even though patients spend the night in the laboratory. Some centers, particularly free-standing ones, bill a global fee that encompasses both the technical expense of performing the procedure and a professional interpretation. Physicians interpreting the studies are then paid by the center on a contractual basis. Other centers bill only a technical fee and the physician bills for the interpretation separately.

The total fees charged for sleep laboratory procedures vary from one facility to the next, depending on the area of the country, operational expenses, collection rates, and type of facility (free-standing versus hospital-based). Global fees generally cost $1,100–$2,000. Insurance reimbursement is highly variable, but generally ranges from 50% to 80% for most commercial carriers. The current Medicare relative value units for sleep laboratory procedures are shown in the accompanying Table. Insurance companies often calculate their reimbursement rates as a percentage of the Medicare allowable. Many health maintenance organizations and preferred provider organizations contract with only certain sleep centers at substantial discounts. Some insurance carriers also distinguish between accredited and unaccredited sleep centers, allowing reimbursement only for those accredited by the American Academy of Sleep Medicine.



There is increasing recognition on the part of the public and the medical community that sleep disorders are common and that they represent a largely unmet public health need. As comprehensive diagnostic and treatment facilities, sleep disorders centers are filling this void. Under our current pluralistic healthcare system, they are both economically viable and contribute to better health care for many Americans.  PP


1.     The American Heritage Dictionary. 2nd ed. Boston, Mass: Houghton Mifflin; 1982.
2.    Riley TL. Historical overview and introduction. In: Riley TL, ed. Clinical Aspects of Sleep and Sleep Disturbance. Boston, Mass: Butterworth; 1975:4-6.
3.    Klink ME, Quan SF, Kaltenborn WT, Lebowitz MD. Risk factors associated with complaints of insomnia in a general adult population. Arch Intern Med. 1992;152:1634-1637.
4.    Roth T, Roehrs TA, Carskadon MA, Dement WC. Daytime sleepiness and alertness. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. 2nd ed. Philadelphia, Pa: W.B. Saunders; 1994:40-41.
5.    Klink M, Quan SF. Prevalence of reported sleep disturbances in a general adult population and their relationship to obstructive airways diseases. Chest. 1987;91:540-546.
6.    Camhi SL, Morgan WJ, Pernisco N, Quan SF. Factors affecting sleep disturbances in children and adolescents. Sleep Med. 2000;1:117-123.
7.    Diagnostic Classification Steering Committee. International Classification of Sleep Disorders—Diagnostic and Coding Manual. Westchester, Ill: American Academy of Sleep Medicine; 1990.
8.    Balter MB, Uhlenhuth EH. New epidemiologic findings about insomnia and its treatment. J Clin Psychiatry. 1992;53(suppl 12):34-39.
9.    Gallup Organization. National Sleep Foundation: Sleep in America—1995 Gallup Poll. Available at:?www.stanford.edu/~dement/95poll.html. Accessed 1995.
10.    Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230-1235.
11.    Gozal D. Sleep-disordered breathing and school performance in children. Pediatrics. 1998;102:616-620.
12.    Guilleminault C, Winkle R, Korobkin R, Simmons B. Children and noctural snoring: evaluation of the effects of sleep related respiratory resistive load and daytime functioning. Eur J Pediatr. 1982;139:7.
13.    Findley LJ, Unverzagt ME, Suratt PM. Automobile accidents involving patients with obstructive sleep apnea. Am Rev Respir Dis. 1988;138:337-340.
14.    Shahar E, Whitney CW, Redline S, et al. Sleep-disordered breathing and cardiovascular disease: cross-sectional results of the Sleep Heart Health Study. Am J Respir Crit Care Med. 2001;163:19-25.
15.    Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med. 2000;342:1378-1384.
16.    Nieto FJ, Young TB, Lind BK, et al. Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. Sleep Heart Health Study. JAMA. 2000;283:1829-1836.
17.    Dement WC. History of sleep physiology and medicine. In: Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine. Philadelphia, Pa: W.B. Saunders; 1994:11-15.
18.    Punjabi NM, Welch D, Strohl K. Sleep disorders in regional sleep centers: a national cooperative study. Coleman II Study Investigators. Sleep. 2000;23:471-480.
19.    Coleman RM, Roffwarg HP, Kennedy SJ, et al. Sleep-wake disorders based on a polysomnographic diagnosis. A national cooperative study. JAMA. 1982;247:997-1003.
20.    Clinical Practice Review Committee. American Acadmey of Sleep Medicine: AASM Standards “Technologist Staffing.” Available at: www.aasmnet.org/PDF%20Files/techstaffing.pdf. Accessed 2000.

Dr. Thompson is associate clinical professor of psychiatry in the Department of Psychiatry at the UCLA School of Medicine in Los Angeles.

Acknowledgments: The author reports no financial, academic, or other support of this work.



Which patients suffering from anxiety respond to dynamic therapy and how does dynamic therapy effect change?  Dynamic therapy can be uniquely helpful to selected patients with anxiety disorders who are often inadequately treated with medication or time-limited treatments. Psychodynamic insight-oriented treatment conceptualizes anxiety not as an isolated symptom, but as a “signal” indicating deeper, more enduring maladaptive character traits and internalized interpersonal reactive patterns. Anxiety is examined concurrently as the patient’s dominant preoccupations, concerns, and relational patterns are identified.  Memories and emotions are explored, real versus fantasized dangers are clarified, and different interactive patterns offering more adaptive responses to current life challenges and opportunities are elaborated.



While Freud and other early psychoanalysts were fascinated by anxiety and other neurotic symptoms, the domain of the modern psychodynamic clinician is the study and treatment of personality. Today, insight is achieved not so much by the archeological search for deeply buried memories, but by attempts to understand personality. Personality, contributing to interpersonal relationships with aspects of desire, self-conceptions, views of others, and anticipations, is endlessly played out in interpersonal dramas of everyday life. These internal patterns of interaction, expectation, and defense typically only reinforce existing internal expectations and fears, often resulting in enduring anxiety.

The modern psychodynamic clinician recommends medication or time-limited cognitive-behavioral therapy for many types of anxiety. Yet, many anxious patients respond incompletely or not at all to medication or time-limited therapies. Numerous patients who are left with enduring anxiety inevitably find their way to psychodynamic therapy. There is no shortage of such patients, who come at personal expense, and apply hemselves to a treatment that requires time, commitment, and labor.


Psychodynamic Treatment

Clearly, not all patients will benefit from psychodynamic therapy. The challenge is identifying which ones will. Psychodynamic treatment has been woefully overprescribed in the past, and while a detailed description of patient selection for psychodynamic therapy is beyond the scope of this article, clinical experience and research would suggest that individuals with chronic maladaptive character traits and personality disorders can be greatly helped by psychodynamic therapy, whereas those with Axis I disorders cannot.

Systematic outcome studies of psychodynamic therapy have been difficult because of the complexity of variables inherent in psychodynamic therapy, the time and financial support necessary to conduct such studies, and a seeming lack of interest on the part of psychodynamic clinicians to develop and participate in such studies. In spite of these limitations, existing studies of psychodynamic therapy, comprehensively reviewed by Gabbard,1,2  suggest that it helps improve global functioning, decrease maladaptive character traits, and reduce the frequency of diagnosable personality disorders.3-8 Thus, insight-oriented therapy offers a viable treatment option when anxiety is interwoven with maladaptive personality traits and character patterns. Conversely, a patient who has no obvious personality difficulties, as evidenced by a healthy capacity to work, love, play, and interact, is likely to be more expediently treated with a time-limited treatment or medication.

How does the psychodynamically oriented clinician view anxiety? Anxiety is conceived of as the proverbial “tip of the iceberg.” Thus, anxiety is a barometer reflecting both temperamental and biologic vulnerabilities, but also complex, enduring inner struggles (eg, lack of inner harmony, anticipation of danger, evidence of deprivation, attachment struggles, or consequence of past trauma). Freud,9 who hypothesized that anxiety was the “signal affect,” posited that when a human being is faced with danger in the real world (eg, an enemy troop cresting the ridge), the individual has two options: flight (run with great urgency), or fight (reach for the nearest weapon).

Anxiety serves as a signal for inner, anticipated danger, which in turn activates an internal psychological response—the use of defenses or coping mechanisms.10-15 Once activated, defenses become the means to protect—a way to manage emotions, perceptions, impulses, desires, and, importantly, a way to deal with internal conflict surrounding significant objects in the individual’s internal world or current relationships.16 For instance, seeing one’s alcoholic mother realistically may be too painful; thus, the child subconsciously chooses not to see the mother’s brutality (denial). Should a friend ask the individual whether his mother has a problem with alcohol, the individual may transiently experience anxiety, which in turn activates a defense of denial. The individual shrugs off the question, minimizes the possibility suggested by the friend, and switches the topic to another subject. Any number of reflexive behaviors from simple to complex, whether changing the subject, leaving the room, having a drink, going for a jog, getting busy cleaning, or having an affair, can be defensive reactions.

How is anxiety addressed in the practice of psychodynamic therapy? First, anxiety must be examined in the context of the “company it keeps,” meaning anxiety does not occur in isolation and, thus, all of the patient’s words, concerns, and preoccupations must be considered in relation to how each might be related to the patient’s experience of anxiety. Freud’s notion of “psychic determinism” hypothesizes that every word, thought, and action has meaning. Free association, words, fantasies, thoughts, and particularly nonverbal communications, attitudes, anticipations, and desires, as expressed toward the therapist, are thought to tell a story. For example, the hopeful expectant look, the fearfully lowered eyes, or the angry sullen silence directed toward the therapist (transference) conveys information about the patient’s history. The story enacted in the transference relationship typically finds the same recurring conclusion. The sullen silence eventually arouses the therapist’s frustration and anger, yet at the same time, the patient’s internal narrative seeks a different ending. At some level, the patient hopes the therapist will not respond with anger and will care instead.

How does change in anxiety occur? The therapist offers a benign, empathic, nonjudgmental, supportive relationship and models and encourages curiosity regarding the patient’s feelings and thoughts, without attempting to confront, “reality test” away (discourage or minimize the patient’s perception of reality), or behaviorally negate any of the patient’s perceptions or feelings. The initial goal of dynamic therapy is to find emotional “room” for the patient to allow for and contemplate inner experiences that he or she may otherwise feel is too shameful or guilt-producing to touch. Gradually, a process unfolds where cognitions, visceral reactions, and responses become more available for consideration. As the patient’s ideas, feelings, and perceptions are shared, they can be gradually clarified. The following physician comments may indicate that the patient’s dominant concerns are initially identified: “you seem to be talking about your kids;” “you seem to be sad;” “you seem to be talking about how heartless people can be;” or “you seem to be upset.” Repeatedly clarifying that the patient has a particular feeling or is thinking about a certain topic, allows both the patient and therapist to ponder further.

As the clarification process occurs, attitudes and reactions toward the therapist become more apparent, and the therapist concurrently becomes more aware of his or her attitudes and feelings toward the patient. Wants, expectations, and fears gradually become articulated and tested in the therapeutic relationship. Old internalized interpersonal schemas (eg, rebellious child, angry retaliating parent, needy child, or indifferent/contemptuous parent) are replayed, relived, and hopefully experienced in less painful, maladaptive, and traumatic ways than previously enacted. Fearful anticipations can be addressed and different interactive patterns can be experimented with.

Old experiences often become strikingly clear both in content and in terms of internal persistence and power. For example, one patient always seemed anxious and fearful, particularly in the moment when the therapist began to make a comment. The therapist stated that he “thought the patient felt as if the therapist was going to attack him.” The comment startled the patient, who found himself instantaneously raising his arms as if to protect his face in anticipation of being hit. The patient then immediately recalled how his mother would slap him as a child, often when he would enter the kitchen after coming home from school, and he began to cry.  With this incident, the patient saw with evident clarity how fearful he really was.17

Two broad elements of psychodynamic treatment contribute to change. First is the therapist’s interpretation of meaning, ideally in the context of felt experience like the above example (eg, you fear this, you anticipate this, you miss this, you want this). Second, is the noncondemning, empathic relationship, which vigorously attempts not to repeat past painful or traumatic interpersonal ways of relating. This is not easy, as the power of the patient’s expectations, the tendency to indirectly test, challenge, and ultimately sound out the therapist while always keenly anticipating and looking for the therapist to fulfill old frustrating and disappointing patterns, often perpetuates this cycle.

The focus of psychodynamic therapy is the stuff of life, the drama of interpersonal and familial experience, that always occurs in the context of temperament, biologic givens, and one’s own unique developmental experience. These are the concerns of life that are reflected in literature and art. This may be the deprivation of childhood that results in the bitterness and stinginess of Charles Dickens’ Ebenezer Scrooge.18 Or, for example, this may be the toddler’s traumatic life experience who, from a helpless position, witnesses his mother’s death by asphyxiation from tubercular hemoptysis, who is ever after left with a preoccupation with death, dying, and the living death. Thus was the life experience of the writer Edgar Allen Poe, as delineated by Terr,19 which apparently led to his adult preoccupation and repetitive expression in his creative works. 

The terrors and fears of human life become woven into the fabric of an individual’s psyche. The goal of psychodynamic therapy is to confront these fears in the light of the present, and in the context of an empathic, desensitizing, and understanding relationship.  Decreasing enduring anxiety typically requires time and patience. On occasion, an insight may offer immediate and sustained relief. For instance, an otherwise high-functioning professional woman who was normally mildly anxious, would become intensely anxious, to the point that she would be unable to sleep, when her husband was away on business trips. Through psychodynamic therapy, it became clear that being alone activated ill-defined fears that touched an old traumatic experience where, when alone in the middle of the night as a college student, the patient’s apartment was entered by a stranger and she was raped at knifepoint. As this old trauma was talked about and related to her current situation, her anxiety diminished and she was able to sleep in her husband’s absence. 

More typically, as illustrated in the following case, anxiety is not relieved so quickly and simply. A married, 35-year-old mother of two came to therapy complaining of an ill-defined sense of malaise and chronic anxiety.  After several months of therapy, she began to share a series of erotic fantasies about several male coworkers.  She deemed these fantasies as unacceptable, yet could not help thinking about how appealing these men were.  She became more and more anxious, worrying whether she could contain her desire, which made little sense to her as she thought she was in love with her husband, in spite of having been irritated with him recently for unclear reasons.

In the therapy sessions she also appeared increasingly anxious, but simultaneously seemed to flirt with her male therapist. She commented on his attractive style of dress, his tie, and wondering out loud how much fun he must be to be with. The clinician made a tentative comment that she seemed to find her therapist appealing but that at the same time she seemed anxious, and he wondered if this attraction and her anxiety were somehow related. At this remark, the patient appeared visibly more anxious and changed the subject. At the next visit, she reported several elements with little conscious awareness that they might be interrelated. She talked of recent anger with her husband after he approached her sexually and wondered why she felt so intensely about this interaction. She delineated a series of sexual inhibitions she had struggled with for many years—something she had not discussed previously. She then related a memory that she noted she had not thought of in 15 years. The memory was of her father, lying in her bed on a camping trip and fondling her genitals. The memory was evidently distressing to her as she relayed the incident.

While somewhat more dramatic than many memories that surface in the context of therapy, this sequence is fairly typical of interpersonal patterns, anxiety, and memories that become alive and more accessable in insight-oriented therapy. Latent feelings, desires, and fears are rekindled, in part by the proximity of the therapeutic relationship. Certain patterns are activated and addressed, resulting in reappearance of  old memories that enlighten the current dilemma. Yet, retrieval of memories alone is usually insufficient for change to occur. For this patient, retrieval of the old disturbing and traumatic memory was only the beginning. Previously, she had idealized her father, who had since come to a sudden and traumatic death.

In her relationships with men, she was often flirtatious, deferential, yet silently angry and self-defeating. She compulsively played out a pattern where she was subtly sexually provocative and would toy with the idea of an affair, yet never followed through, and ultimately felt guilty, anxious, and remained sexually inhibited. The memory described above marked a turning point for her, as her disappointment with her father and with men in general, began to crystallize, and her sexualized, yet self-demeaning pattern of interacting could now be examined openly.19 These patterns with her therapist were enacted with twists and variations as she both mourned the absence of a more supportive and nonabusive father, and experimented with different and more assertive ways of interacting in the context of therapy, with a concomitant decrease in her anxiety. In the process of therapy, the patient discovered feelings, memories, attitudes, and anticipation that heretofore she was only marginally aware of.



In dynamic therapy, the process of discovery occurs repeatedly before awareness of feelings, memories, and inner reactive patterns become fluid and can be accessed with ease. This repetitive process is the work of therapy, allowing for a gradual desensitization of old fears and concerns. Sorting out fantasized dangers from the real dangers of life and ultimately facing the true dangers is an important goal of psychodynamic therapy. An anxious patient may obsess about the risk of contracting acquired immunodeficiency syndrome, which may be a completely unrealistic danger for him. Yet, he may still need to face the true risk—that of being rejected and hurt, should he ask a woman out on a date. Life presents real risks that must be faced to handle adaptive living of life. Dealing with enduring psychological patterns and fears is a daunting task. Psychotherapy cannot undo all traumas and cannot always provide what has been neglected. Nevertheless, dynamic therapy can help in important ways to understand and offer patients a more benign laboratory for new experiences and opportunities to live life in less fearful and less maladaptive ways.  PP



1. Gabbard GO. Psychoanalysis and psychoanalytic psychotherapy. In: Livesley WJ, ed. Handbook of Personality Disorders: Theory, Research, and Treatment. New York, NY: The Guilford Press; 2001:359-376.
 2. Gabbard GO, Lazar SG, Hornberger J, Spiegel D. The economic impact of psychotherapy: A review. Am J Psychiatry. 1997;154:147-155.
3. Bateman A, Fonagy P. Treatment of borderline personality disorder with psychoanalytically oriented partial hospitalization: An 18-month follow-up. Am J Psychiatry. 2001;158:36-42.
4. Kernberg OF, Burstein ED, Coyne L, Appelbaum A, Horwitz L, Woth H. Psychotherapy and psychoanalysis: Final report of the Menninger Foundation’s Psychotherapy Research Project. Bull Menninger Clin. 1972;36:87-275.
5.    Meares R, Stevenson J, Comerford A. Psychotherapy with borderline patients: I. A comparison between treated and untreated cohorts. Aust N Z J Psychiatry. 1999;33:467-472.
6. Monsen J, Odland T, Faugli A, Daae E, Eilersten DE. Personality disorders and psychosocial changes after intensive psychotherapy: A prospective follow-up study of an outpatient psychotherapy project, 5 years after end of treatment. Scand J Psychology. 1995a;36:256-268.
7. Monsen J, Odland T, Faugli A, Daae E, Eilersten DE. Personality disorders and psychosocial changes after intensive psychotherapy: Changes and stability after intensive psychotherapy focusing on affect consciousness. Psychotherapy Res. 1995;5:33-48.
8.     Stevenson J, Meares R. An outcome study of psychotherapy for patients with borderline personality disorder. Am J Psychiatry. 1992;149:358-362.
9.   Freud S. Inhibitions, Symptoms and Anxiety. Standard edition, vol. 20. London, England: Hogarth Press; 1926:77-178.
10. Freud S. Introductory Lectures on Psycho-analysis. Standard edition, vol. 15. London, England: Hogarth Press; 1933:13-80.
11. Freud S. The Ego and the Id. Standard edition, vol. 19. London, England: Hogarth Press; 1923:3-107.
12. Compton A. A study of psychoanalytic theory of anxiety. 1. The development of Freud’s theory of anxiety. J Am Psychoanalytic Assoc. 1972;20:3-44.
13. Compton A. A study of the psychoanalytic theory of anxiety: II. Developments in the theory of anxiety since 1926. J Am Psychoanalytic Assoc. 1972;20:341-394.
14. Compton A. A study of the psychoanalytic theory of anxiety: III. A preliminary formulation of the anxiety response. J Am Psychoanalytic Assoc. 1980;28:739-773.
15. Compton A. An investigation of anxious thought in patients with DSM-IV agoraphobia/panic disorder: rationale and design. J Am Psychoanalytic Assoc. 1998;46:691-721.
16. Vaillant GE, Vaillant LM. The role of ego mechanisms of defense in the diagnosis of personality disorders. In: Darron J, ed. Making Diagnosis Meaningful: Enhancing Evaluation and Treatment of Psychological Disorders. Washington, DC: American Psychological Association; 1999:139-158.
17. Dickens C. A Christmas Carol. New York, NY: Washington Square Press; 1962.
18. Terr L. Childhood trauma and the creative production. Psychoanalytic Study Child. 1987;42:545-572.
19. Thompson JM, Cotlove C. A Clinical Introduction to the Therapeutic Process.  Northvale, NJ: Jason Aronson Inc. Publishers. In press.