Mood Disorders and Workplace Performance: Half a Loaf

Mood disorders have far-reaching consequences. Recent studies have demonstrated their impact on the children of depressed parents (1, 2) . Two papers in this issue of the Journal focus on the impact of mood disorders in the workplace, where they are common and where they cause substantial disruption to job performance. An epidemiological study by Kessler and his colleagues clearly documents those facts. What then is the relationship between job performance and symptoms for workers with depression? A prospective study by Adler and his colleagues, also in this issue of the Journal, concludes that the relationship is complex. Although symptom remission is associated with improved job performance, deficits remain in workplace performance. Treatment seems to confer benefits but is limited. This represents half a loaf.

Kessler and his colleagues find that mood disorders are common among workers, and they cause substantial problems with workplace performance. From scores on the WHO Comprehensive International Diagnostic Interview (CIDI), investigators identified individuals who met diagnostic criteria for major depressive disorder or bipolar disorder. Among these individuals, they identified a subgroup of individuals that were employed at least 20 hours per week. Among these employed individuals, they found annual prevalences of 6.4% and 1.1% for major depressive disorder and bipolar disorder, respectively. These individuals were asked to report their absences from work as well as assess their performance at work on a scale in which 100 represents fully effective work performance and zero represents no productive work. The ratings of absenteeism and “presenteeism” (i.e., low performance while at work) were combined to estimate annual days of lost productivity and the costs associated with those losses.

The study estimated that major depressive disorder was associated with 8.7 days absent and 18.2 days of lost productivity per year at a cost of $4,426 per person annually. The comparable figures for bipolar disorder were 27.7 days absent and 35.3 days of lost productivity annually at a cost of $9,619 per person with bipolar disorder. When these data from the representative sample were inflated to the entire adult population of the U.S., the lost productivity was estimated at $36.6 billion for major depressive disorder and $14.1 billion for bipolar disorder. The magnitude of lost productivity from absenteeism and presenteeism for workers with mood disorders is substantial, according to this well-designed probability sample. Of course, those out of work completely are presumed to be even more impaired and costly.

There are limitations to the design of this study, but they do not change the main conclusion about the importance of recognizing the workplace performance problems and economic losses associated with major depressive disorder and bipolar disorder. One can be somewhat skeptical of self-reports on performance from which presenteeism assessments are derived, since they are likely biased by mood. The authors assure us that these measures have been validated against the ratings of job supervisors, but self-reported presenteeism is inherently subjective. A bigger problem is the failure to compare the rates of absenteeism and presenteeism in their sample with individuals with other conditions and with normal subjects. What are the average annual losses due to absenteeism and presenteeism for all workers? How much is lost each year while we “normal” workers stare out the window or play solitaire on our computers?

It is clear that losses due to mood disorders in the workplace are substantial. Kessler and his authors realize, however, that the key question is “What are the benefits of reducing symptoms over time and with treatment?” They ask, “What is the ‘return on investment’ of programs to identify and treat mood disorders? “

The paper by Adler and his colleagues (4) focuses on one aspect of those questions as well. They report on a prospective study of workers who are depressed. The study compares their job performance with healthy subjects and with workers with rheumatoid arthritis. The study observes the relationship between symptom remission and various measures of work-related performance. The investigators find that various aspects of job performance are affected by depression. They conclude that as symptoms remit, job performance improves, but that depressed workers whose clinical status improves continue to have job performance deficits. They remain more impaired than healthy subjects and even more impaired than those recovering from rheumatoid arthritis.

There are some limitations to the Adler et al. study. As with the Kessler paper, there are concerns about the accuracy of self-report on the work measure. The study also does not identify some of the specific impairments associated with depression-related problems with job performance, such as cognitive deficits (3) . Medical comorbidity could also explain lingering problems with workplace functioning for those workers with a primary diagnosis of depression. Those limitations aside, this study provides important insights into the world of work and mental impairment.

The authors suggest that while effective treatments are critical, they are not sufficient. Specific workplace interventions are needed, such as employee assistance programs and occupational health interventions. They also suggest that supported employment strategies that have been successful with individuals with severe psychiatric disabilities (4) might be adapted for use with depressed workers. We concur with this conclusion. People with residual symptoms, however subtle, likely need supports and accommodations to work effectively.

Only an experimental trial will answer this question. In the meanwhile, clinicians and patients should avail themselves of vocational rehabilitation services, if they can find them.

Here we have summarized the findings of these studies and discussed their limitations and implications and conclude that treatment for mood disorders might hold promise for improving work performance, but more than treatment is required. These two studies cast important light on the prevalence of mood disorders among workers and the associated disruption of work performance. Many days of work productivity are lost, but time and treatment seem to reduce these losses. Unfortunately, symptom remission associated with natural recovery and treatment is not sufficient to return depressed workers to full productivity. If full recovery and productive work are the objective, additional workplace interventions and specific rehabilitation efforts must be added to effective treatment for mood disorders.