Patterns of health care costs associated with depression and substance abuse in a national sample

Psychiatr Serv. 1999 Feb;50(2):214-8. doi: 10.1176/ps.50.2.214.


Objective: The associations between self-reported depressive and substance use disorders and estimated health care costs were examined in a representative national sample.

Methods: Data were from the 1994 National Health Interview Survey (N=77,183). Respondents who reported depressive symptoms or major depression (depressive syndromes) or a substance abuse disorder in the past year were compared with respondents who did not report these conditions. The mean number of inpatient days and outpatient visits in both the general medical and the specialty mental health settings were determined, and costs per individual were calculated based on mean costs of such care in each respondent's geographic region. Multivariate models were constructed to calculate mean costs, controlling for demographic variables, insurance coverage, and physical health status.

Results: Individuals with self-reported depressive syndromes or substance abuse had mean health care costs that were $1,766 higher than costs for individuals without these conditions. Depressive syndromes were associated with increases in both inpatient and outpatient costs. However, substance abuse was almost exclusively associated with increased inpatient expenditures rather than outpatient costs. The magnitude of increased costs associated with mental disorders was substantially larger for patients in fee-for-service plans than for those in health maintenance organizations. Only 14.3 percent of visits made by individuals reporting depressive syndromes or substance abuse were made to specialty health providers (psychiatrists, psychologists, and social workers).

Conclusions: Health care costs of people with self-reported mental illness varied significantly across diagnoses and systems of care. It is crucial that researchers estimating increased costs associated with mental illness account for both diagnostic and system factors that can influence the estimates.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Aged
  • Comorbidity
  • Cost of Illness
  • Cross-Sectional Studies
  • Depression / economics*
  • Depression / epidemiology
  • Female
  • Health Care Costs / statistics & numerical data*
  • Health Care Surveys
  • Health Status
  • Humans
  • Insurance, Health / classification
  • Insurance, Health / economics
  • Insurance, Health / statistics & numerical data
  • Least-Squares Analysis
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Substance-Related Disorders / economics*
  • Substance-Related Disorders / epidemiology
  • United States / epidemiology