The following clinical studies analyzed medical expenditures during longitudinal trials, or cross-sectional comparisons,
to estimate both a baseline difference in direct medical costs for depressed and non-depressed individuals as well as a
decrease in these costs for patients after treatment for depression. From these studies, a conservative
range of $1,000 to $2,000 in excess medical costs in
depressed individuals was derived. You can change this default range if desired.
These studies either analyzed claims data in order to compare claims costs for patients with depression with claims costs for all other employees, or in one case (Simon, Revicki et al.), conducted a longitudinal study with depression-affected employees who were tracked for a two-year period in order to identify the impact of treatment for depression.
The four studies present statistical evidence of the correlation between depression and higher health care costs. Although the demonstrated correlations are not the same as demonstrated causation, these links support the inclusion in the Productivity Impact Model of a potential reduction in health care costs. In addition, research on depression indicates that patients with depression are less compliant about treatment regimens for any condition, and are more likely to engage in high-risk behaviors that are linked to other health conditions, which are further theoretical arguments for expecting improvements in depression to result in lower medical costs. (It should be noted that even without any reduction in direct medical costs, the Calculator still shows a net benefit to employers from the treatment of depression.)
The following studies made up the primary evidence used in the Depression Model about the impact of depression treatment on health care costs:
“Recovery from Depression, Work Productivity, and Health Care Costs Among Primary Care Patients”
Gregory E. Simon, MD, MPH, Dennis Revicki, PhD, John Heilgenstein, MD, Louis Grothaus, MA, Michael Von Korff, ScD, Wayne J. Katon, MD, Timothy R. Hylan, PhD
General Hospital Psychiatry 22, 153-162, 2000
This longitudinal study covered about 290 employees in the Seattle area who had been diagnosed with major depression. The study examined health care claims periodically from the beginning of treatment for depression for two years.
The study found a significant reduction in absenteeism for those patients with improvement in their depression. The study also found a potential reduction in medical costs for co-morbid conditions. In year two, medical costs for unimproved patients were more than double those who were improved or remitted:
| |
Unimproved
employees
(n = 35) |
Improved or
remitted
employees
(n = 255) |
|
| Year 2 total health care costs |
$6,365 |
$2,236 - $2,345 |
Þ = .06 |
The authors note:
“Improvement in depression could contribute to reduced medical utilization through at least two mechanisms. First, depression leads to increased perception and reporting of somatic distress of all kinds. Resolution of depression might be expected to reduce utilization associated with medically unexplained somatic symptoms. Second, depression is associated with increased prevalence and worse prognosis of medical conditions such as hypertension, diabetes, and ischemic heart disease. Improvement in depression could reduce health care utilization attributable to well-defined medical conditions.”
Also,
“Although observational data cannot definitively prove any causal relationships, these longitudinal results strengthen previous findings regarding the economic burden of depression on employers and health insurers.”
“The Relationship Between Modifiable Health Risks and Health Care Expenditures: An Analysis of the Multi-Employer HERO Health Risk and Cost Database”
Ron Z. Goetzel, PhD, David R. Anderson, PhD, R. William Whitmer, MBA, Ronald Ozminkowski, PhD, Rodney L. Dunn, MS, Jeffrey Wasserman, PhD
JOEM, October 1998
This was a cross-sectional, retrospective study of risk factors and associated medical expenditures. In it the authors examine whether risk factors, within a population of 46,000 employees from several companies, can be statistically linked to higher expenditures for common medical disorders such as heart disease, diabetes, stroke, psychological problems, etc. Depression was considered a risk factor, along with several other behavioral factors such as stress, exercise (or lack of), cholesterol levels, etc.
The statistical analysis showed that depression, and stress, were actually stronger links to higher medical expenditures for other conditions, compared to other risk factors such as exercise and smoking.
“Among these risk factors, the likelihood of incurring any medical expenditures was highest for employees with self-reported depression and high stress.”
“Mean medical expenditures were much higher for those reporting being depressed.”
“Individuals at high risk for stroke and individuals at high risk for psychosocial problems also had predicted medical expenditures that were much higher than those for employees without these risks.” [About $2,000 more per year.]
The authors speculate on the correlation between depression as a risk factor, and higher medical expenditures:
“... First, depressed and highly stressed individuals may seek medical attention for physical conditions such as unspecified pain, fatigue, or headaches, when time would be better focused on their (perhaps undiagnosed) psychological or social issues. Second, depressed or highly stressed persons may consequently develop more serious illnesses or disorders with psychosocial antecedents (e.g., heart disease). Third, people suffering from serious illness may become depressed and highly stressed as a consequence of their illnesses.”
“At the other end of the expenditure spectrum, individuals with [other] risk factors showed no incremental [medical expenditures].”
“Workplace Burden of Depression: A Case Study in Social Functioning Using Employer Claims Data”
Howard G. Birnbaum, PhD, Paul E. Greenberg, MA, MBA, Mary Barton, MBA, Ronald C. Kessler, PhD, Clayton R. Rowland, PhD, Todd E. Williamson, MSc
Drug Benefit Trends, 11(8):6 BH-12 BH, 1999
In this retrospective study, the authors used claims data from a Fortune 100 manufacturer to examine the relationship between depression and medical expenditures (for conditions other than depression). Claims data for workers diagnosed with depression were adjusted for demographic factors and compared with medical claims for all other workers.
The study found that employees diagnosed with depression were higher users of the health care system, and incurred total health care costs four times that of the overall employee average.
“These outcomes suggest that individuals treated for depression, and particularly those treated for major depression, used health care services to a greater degree than the general employee population. These findings imply that the incremental cost burden associated with this illness is likely to be quite high.”
“Since MD patients were especially high users of the health care system, it is not surprising that their total medical, pharmaceutical, and disability costs were 4.2 times those of the typical beneficiary in the company ($8,709 versus $2,059) for the year.”
“After excluding the cost of [depression] treatment, the remaining annual per capita health care costs (i.e., medical and prescription) of [depression] patients were almost three times those of the overall employee population ($5,092 versus $1,790 respectively). These results are consistent with the findings of Croghan et al. (1998) who report that the bulk of the health care cost of treating patients with depression is for care of co-morbid conditions.”
Robinson RL, Birnbaum HG, Morley MA, Sisitsky T, Greenberg PE, Wolfe F (2004). Depression and Fibromyalgia: Treatment and Cost When Diagnosed Separately or Concurrently. The Journal of Rheumatology, 31:8, 1621-1629.
Claims data from a Fortune 100 company was analyzed. Findings from two groups are important to the Productivity Impact Model: the overall employer sample and MDD-only patients. MDD-only patients had $5,974 in 1998 in health care payments while the overall sample had $1,840. This difference is $4,134.
The Productivity Impact Model is conservative about expectations for reduced medical costs as a result of treatment for depression. The authors in these studies make the case that patients with depression have higher medical costs for co-morbid conditions, and they try to link the cost differential to specific symptoms of depression. The more reasonable use of this evidence is to incorporate it conservatively in the model, rather than disregard the impact altogether.
 |
 |
Back |
