Psychiatric Dispatches

Print Friendly 

Primary Psychiatry. 2007;14(1):14-18

Higher Rates of SSRI Prescriptions Associated with Lower Suicide Rates in Youths

Suicide is the third leading cause of death in youths 10–14 years of age in the United States. Evidence shows that most youth suicide victims have mood disorders that are not being treated at the time of death, suggesting that lack of treatment contributes to suicide risk. Concerns about the effect of some selective serotonin reuptake inhibitors (SSRIs) on the risk of youth suicidal ideation have resulted in a black box warning on SSRIs for all age groups. However, recent findings have suggested that higher levels of SSRI prescriptions are associated with lower suicide rates in youths 5–14 years of age.

Robert D. Gibbons, PhD, at the Center for Health Statistics at the University of Illinois in Chicago, and colleagues, analyzed county-level suicide data of youths 5–14 years of age from all 50 states that was gathered by the Centers for Disease Control and Prevention (CDC) between 1996 and 1998. These data were compared with concurrent county-level antidepressant prescription rate data from the US-based global health data firm IMS Health Inc., which represents >50% of all retail pharmacies in the continental US. A random sample of dispensing data was collected from 20,000 pharmacies for the SSRIs citalopram, fluoxetine, fluvoxamine, paroxetine, and sertraline.

Results of the study found that counties with the highest rates of SSRI prescribing had the lowest suicide rates (0.7 per 100,000) and counties with the lowest SSRI prescription rates had the highest suicide rates (1.7 per 100,000). The positive effect of SSRI prescriptions remained highly significant after adjusting for gender, race, income, and access to quality mental health care.

Gibbons and colleagues concluded that SSRI prescriptions are associated with lower youth suicide rates and may reflect antidepressant efficacy, treatment compliance, better quality mental health care, and low toxicity in the event of a suicide attempt by overdose.

“The black box warning should be interpreted by primary care physicians as a reminder to carefully monitor patients put on antidepressants, but not as a reason to stop prescribing them,” Dr. Gibbons said. “Recent IMS data reveal that the rate of prescription of SSRIs to children has decreased by approximately 30%. Using the statistical model presented in our paper, a 30% reduction in SSRI prescriptions will result in 81 additional deaths by suicide per year in children under 15. This is a huge increase given that the yearly number of suicides in children in this age cohort was 323 for the time period (1996–1998) in our study.”

Dr. Gibbons added that if a similar decrease in SSRI and serotonin norepinephrine prescriptions of 30% resulted from the black box warning that the FDA is considering in their December 2006 advisory meeting, there will most likely be an estimated additional 5,517 suicides per year (there were 30,557 suicides per year in the general population at the time of the 1996–1998 study).

Limitations to the study included that medication estimates were based on outpatient prescriptions as opposed to actual use, and that there may have been uncontrolled variability in the suicide rate data due to regional differences.

“The greatest limitation of the study is that it is an ecological study, and we have no way of knowing whether a particular person who did or did not take an antidepressant did or did not commit suicide,” Dr. Gibbons said. “One way to overcome these limitations is to examine large-scale health databases, such as the VA database in the US or the PHARMO database in the Netherlands, where long-term health records and prescription records are available on the same individuals over many years.”

Although there are limitations in the analysis of this type of naturalistic or observational data as well, Dr. Gibbons pointed out that they give insight to the patient’s own experience in sample sizes that are sufficiently large.

Gibbons and colleagues are currently working with the CDC to obtain data on completed suicide rates in US counties and relate changes in these suicide rates to changes in prescription rates that have occurred since the time of the black box warning.

Funding for this research was provided by the National Institute of Mental Health. IMS Health Inc. data were obtained with the assistance of a grant from Pfizer. (Am J Psychiatry. 163:1898-1904.) —DC

Danish Researchers Find Increased Prevalence of Mental Disorders in First-Time Mothers

Postpartum depression affects 10% to 15% of women after childbirth. New research out of Denmark has found that first-time mothers are more prone to hospitalization, are in greater need of outpatient services, and may be at increased risk for various mental disorders.

Trine Munk-Olsen, MSc, from the University of Aarhus, Denmark, and colleagues, reviewed and analyzed data from 2,357,942 Danish-born persons between January 1973 and July 2005 provided by Danish health and civil service registers. Of these patients, 630,373 women and 547,531 men became first-time parents during this time period. Munk-Olsen and colleagues defined new mothers as a woman who has her first live-born child.

“Overall, we wanted to elaborate on what is already known about postpartum mental disorders, and for the first time we wanted to include men in our study,” Ms. Munk-Olsen said. “The aim was to estimate the risk of postpartum mental disorders necessitating hospital admission or outpatient contact for fathers as well as mothers. We also investigated if parents in general differed from nonparents in the risk of admission for a mental disorder and how this difference varied with age.”

Of first-time parents, 1,171 mothers and 658 fathers were admitted to the hospital with a mental disorder within the first year postpartum. Clinical diagnosis for these disorders was made using International Classification of Diseases, Eight Revision and International Classification of Diseases, 10th Revision (ICD-10) criteria.

At 3 months postpartum, the prevalence of severe mental disorders were 1.03 per 1,000 births for mothers and 0.37 per 1,000 births per fathers. When compared to women that gave birth in the previous 12 months, the highest risk period was found to be between 10 and 19 days postpartum (7.3 times increased risk; 95% CI; 5.44-9.81). The highest risk period for psychiatric outpatient contact was also found to occur during days 10 and 19 postpartum (2.67 times increased risk; 95% CI; 1.99-3.59).

New mothers were at increased risk for schizophrenia, depression, and bipolar disorder. Of the 1,171 mothers admitted during the first year after giving birth to their first child, the researchers found that: 102 mothers had schizophrenia, schizophrenia-like, and schizotypal disorders; 280 mothers had unipolar depressive disorders; 44 mothers had bipolar affective disorders; 242 mothers had adjustment disorders; 119 mothers had puerperal disorders (defined in the ICD-10 as “Mental and behavioral disorders associated with the puerperium, not elsewhere classified”); and 384 mothers had other diagnoses. They also noted that fatherhood did not impact the risk of being admitted to the hospital or having outpatient contact.

Funding for this research was provided by the Stanley Medical Research Institute.  (JAMA. 2006;196:2682-2689). —CN

Gender as a Factor in Posttraumatic Stress Disorder Diagnosis

Posttraumatic stress disorder (PTSD), a common condition for patients who have experienced or witnessed a stressful event or ordeal such as a vehicle accident or natural disaster, is characterized by recurrent and distressing recollections or dreams of the event and intense psychological distress, according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). According to the National Institute of Mental Health, an estimated 5 million adults 18–54 years of age in the United States, or 3.6% of that population, are affected by PTSD. Research has shown that women experience PTSD following trauma in higher rates than men who experience similar trauma.

A recent study led by David F. Tolin, PhD, of the Anxiety Disorders Center at the Institute of Living in Hartford, Connecticut and Edna B. Foa, PhD, of the University of Pennsylvania School of Medicine in Philadelphia sought to examine the rates of PTSD diagnosis among men and women, and compare those rates with the gender-specific risk of experiencing potentially traumatic events. The authors reviewed 290 studies conducted between 1980 and 2005 to determine which gender has a higher risk for potentially traumatic events and a PTSD diagnosis.

Tolin and Foa found that although men, on average, experience more traumatic events than women, the PTSD diagnosis rates are higher for women than for men. In addition, the authors found that the type of traumatic events men and women typically experienced was different, with women experiencing more sexual assault and child sexual abuse, and men experiencing more non-sexual assaults, accidents, natural disasters or fire, incidents of witnessing injury or death, combat, and warfare. Among the reviewed studies, differences in age, population, assessment tool, and type of study did not affect results.

The authors said that the findings show that gender differences in risk of exposure to particular types of trauma can only partially account for the difference between PTSD diagnosis rates in men and women. Tolin and Foa concluded that women’s higher rates of PTSD diagnosis may be caused by an increased emotional reaction to sexual trauma. The authors added that this finding could be due to new trauma causing prior PTSD symptoms to re-affect the patient. However, when the study authors compared PTSD rates among male and female patients who experienced sexual trauma, they found that women continued to have higher PTSD rates.

Tolin and Foa said the DSM-IV criteria for PTSD may also predispose women to receive the diagnosis due to the cognitive and emotional assessments involved in the diagnosis. Men may manifest reaction to traumatic events in forms that do not qualify them for PTSD diagnosis, including reacting through violence or drug abuse. In the reviewed studies, male patients reported less anxiety or depression but were more likely to be angry and have other behavioral problems following a traumatic event.

“We considered whether men may be less willing than women to endorse the symptoms of PTSD, although they may still have them,” Dr. Tolin said. “This remains an open question.”
The authors concluded that understanding that different patients can have different reactions to traumatic events—including reactions that are not included in PTSD diagnosis—will help clinicians better determine if a patient has experienced and is being affected by trauma.

“PTSD is often difficult to diagnose accurately, in large part because different people respond to events quite differently,” Dr. Tolin said. “It is important to recognize that we can’t predict PTSD based on the traumatic event itself. Many people can endure severe trauma without developing long-lasting PTSD symptoms, whereas others can develop severe PTSD following events that many of us would not consider particularly serious.”

The authors also called for clinicians to perform more complete assessments following a patient’s traumatic event to assess whether PTSD will develop. (Psychol Bull. 2006;132(6):959-992.) —CP

Researchers Find Few Differences in Effectiveness of First- and Second-Generation Antipsychotics for Schizophrenia Patients

Originally developed in the 1950s, first-generation antipsychotics (FGAs) were the standard treatment for schizophrenia for decades. Second-generation antipsychotics (SGAs) were introduced in the 1990s and soon replaced FGAs as the standard treatment option as they were thought to be safer and more effective. New research from the United Kingdom is challenging this long-held theory.

Peter B. Jones, MD, PhD, from the University of Cambridge in England, and colleagues, conducted a multicenter, randomized, controlled trial of 227 patients between 18–65 years of age diagnosed with schizophrenia, schizoaffective disorder, or delusional disorder as per Diagnostic Statistical Manual of Mental Disorders, Fourth Edition, criteria. A change in each patient’s drug regimen was being considered due to drug ineffectiveness or harmful side effects. One hundred eighteen patients were randomly assigned to receive FGAs and 109 patients were randomly assigned to receive SGAs. Each patient was assessed by the researchers to determine their individual therapies. The Study took place in five university centers and involved 73 clinicians.     

“This study was undertaken to test whether SGAs as a class, led to clinically meaningful improvements in quality of life, over 1 year, compared to FGAs,” Dr. Jones said. “The design aimed to replicate routine clinical care, other than the randomization to class of drugs and the blind, independent rating of outcome”

Eighty-one percent of patients were followed at baseline, 12 weeks, 26 weeks, and 52 weeks using the Quality of Life Scale (QLS) as the primary outcome measure. The QLS is a 45-minute semi-structured interview mainly used in psychopharmacologic treatment trials for schizophrenia. It has 21 items, rated from 0 to 6 with various descriptive anchors. The higher the score, the more normal the patient’s functioning is. QLS scores are detailed in the Table. Contrary to expectation, the five-point QLS benefit for SGA drugs was excluded with a high degree of confidence.

At a 1-year follow-up, the researchers interviewed 185 study participants (100 from the FGA arm and 85 from the SGA arm). A total of 39 patients (17%) died (three in each arm); were lost to follow-up (11); or withdrew from the study (22).

Overall, Jones and colleagues found that patients receiving FGAs had greater overall improvement in symptoms, although this did not reach statistical significance (P=.24). These results lead the researchers to believe that there is little to be gained by switching medications for schizophrenic patients.

“The results do not suggest that FGAs are superior to SGAs; merely that the expected advantage of the latter drugs could not be found,” Dr. Jones said.  “Moreover, benefits of SGAs were not apparent in other domains, including patient preference and side effects, ie extrapyramidal motor side effects.”

According to Jones, this suggests that clinicians can use either class of drugs in patients with schizophrenia, so long as they prescribe carefully and aim for minimal side effects. He also pointed out that very few clinicians and patients chose the FGA haloperidol, in contrast to many efficacy studies of SGA drugs. Further research is certainly necessary in order to further explore these important findings.

Funding for this research was provided by project grant 96/19/06 from the Secretary of State for Health under the United Kingdom National Health Service Health Technology Assessment Program. (Arch Gen Psychiatry. 2006;63:1079-1087.) —CN

Depression Screening and Care in the Workplace May Be Cost Effective

Depression affects patients’ health and well-being as well as productivity level in many areas. The illness costs tens of billions of dollars annually in the United States due to reduced work productivity by employees who are depressed. Available treatments and screening methods may reduce costs but are currently underused.

A cost-effectiveness and cost-benefit analysis conducted by Philip S. Wang, MD, DrPH, at Brigham and Women’s Hospital and Harvard Medical School in Boston, Massachusetts, and colleagues, reported that additional depression screening and care may potentially save employers approximately $2,900 per 1,000 employees over the course of 5 years. The computerized cost-benefit analysis examined data from an enhanced depression care program that screened all employees for depression once and provided telephone counseling and other depression-management strategies with positive results. The model consisted of 40-year-old hypothetical employees, as that is the median age of US workers. Employees were divided into six groups based on percentages of each group in the general population, including those who were never depressed, depressed but not in treatment, depressed and in treatment, recovered but not in treatment, recovered and in treatment, and deceased. The model included both adequate and substandard depression treatment to reflect realistic quality variations. Costs of treatment and care management were based on averages from clinical literature and health plan data. Cost of depression screening administration was based on the cost of implementing new health risk assessments for companies that did not offer them and adding questions to existing health risk assessments for companies that did currently administer them.

Data analysis from a societal perspective assigned model employees to either usual care or to the enhanced program and followed them until death. Approximately $20,000 spent on the enhanced depression management program, including medication costs and time spent in treatment, was shown to provide one employee with one depression-free year. Data analysis from an employer perspective consisted of a 5-year simulation that kept the employee base at a constant size by replacing employees who died or otherwise left the company. Over this time period employers saved $2,895 per 1,000 employees.

“It is somewhat counterintuitive that interventions developed to increase the use and intensity of treatments for depression actually may end up saving employers money because of recovered absenteeism, presenteeism, and worker turnover,” Dr. Wang said.

Dr. Wang added that the greatest limitation to the study was that the data were collected largely from studies conducted in primary care populations, and not necessarily from workers. He noted that effectiveness trials conducted in working populations that assess work outcomes will be needed to verify these findings.

Funding for this study was provided by the National Institute of Mental Health and the Robert Wood Johnson Foundation. (Arch Gen Psychiatry. 2006;63:1345-1353.) —DC

Elderly Patients With Depression Benefit More from Psychosocial Treatment than Antidepressants

Major depressive disorder (MDD) and other forms of depression can affect patients of any age. However, such disorders are particularly common in patients ≥55 years of age. MDD, characterized by one or more depressive episodes that develop over days to weeks, often occurs comorbid with other physical and mental health disorders. Along with other depressive disorders, MDD can exacerbate the severity other health disorders or illnesses and increase suicide risk. Prior studies have shown that estimates of depressive symptoms in the elderly, who are often treated by primary care physicians (PCPs) or in community care settings, ranges from 10% to 25%.

Karyn Skultety, PhD, of the Veterans Affairs Palo Alto Health Care System in Menlo Park, California, and colleagues, recently evaluated findings from various studies that compared “usual care” of depression in the elderly, including treatments using antidepressants, with several psychosocial interventions to treat depressive symptoms. In the study, Skultety and colleagues sought to review and evaluate the efficacy of different depression treatments in older adults.

The authors evaluated eight randomized, controlled, English-language studies published from January 1994 to April 2004. Included studies had to be comparisons of typical MDD and depressive symptom treatment for adults ≥55 years of age and psychosocial-centered treatments. Each study focused on a particular psychological assessment or treatment in the context of an examination into a specific physical disease or reduction of disease risk. The number of patients in the included studies ranged from 96–1,801.

Skultety and colleagues found that most of the included studies utilized two types of psychosocial depression intervention. The first type, geriatric evaluation and management clinics, are inpatient facilities where PCPs and mental health workers examine a patient’s medical, rehabilitation, and psychosocial problems. The second type of intervention, integrated health care models, treat a patient’s mental and physical disorders or illnesses simultaneously.

Psychosocial interventions included counseling and education initiatives provided by PCPs, nurses, psychologists, counselors, and social workers. “Usual care” in each study involved PCPs diagnosing MDD and prescribing treatment.

The review authors found that patients in both psychosocial depression treatment models showed significant improvements in depressive symptoms and better long-term outcomes than patients receiving usual depression treatment, including taking antidepressants. Although these results varied between studies, the authors concluded that the results show that cognitive-behavioral therapies are effective in depression treatment in the elderly and are the desired model for primary care.

Skultety and colleagues also found that patients preferred psychosocial treatments, which have been used previously to treat illnesses including diabetes, to typical treatment through a PCP. The authors said that one deterrent to elderly patients receiving interdisciplinary care is cost and the additional time commitment required for treatment. The authors added that, while cost is a factor, psychosocial treatments could reduce the number of visits an elderly patient makes to a PCP over time. 

Skultety and colleagues cite the inability to combine results across studies due to variation and the low number of reviewed research as study limitations. In addition, the authors believe researchers should create clear guidelines for future investigations comparing traditional and non-traditional MDD treatment in the elderly and report the number of patients who do not continue with alternative treatment programs, which can be more time involving than traditional treatment. (Health Psychol. 2006;25(6):665-674.) —CP

Psychiatric Dispatches is written by Dena Croog, Christopher Naccari, and Carlos Perkins, Jr.