Cost-effectiveness analysis of complementary and alternative medicine in treating mental health disorders

Using survey data, this study produces estimates of incremental cost-effectiveness ratios (ICERs) and measures of uncertainty, regarding use of complementary and alternative medicine (CAM) combined with traditional therapies (pharmacotherapy and/or psychotherapy) in the treatment of five common mental health disorders. Background. A recent, large-scale nationally representative survey estimated the 12-month prevalence for any mental health disorder at over 25% of the adult population (Kessler et al., 2005). Traditional therapies include both psychotherapy and pharmacotherapy. Past studies indicate that patients with mental health disorders use complementary and alternative medicines, such as acupuncture, chiropractics, herbal remedies, massage therapy, homeopathy, energy healing, and biofeedback, at a greater rate than the general population. Unlike past cost-effectiveness analyses that use either a narrow definition of CAM or randomized controlled trials, this study uses a broad definition of CAM and survey data. A secondary question is how well this self-reported survey data produces cost-effectiveness analysis results for CAM. Methods. This analysis uses the Medical Expenditure Panel Survey (MEPS), a panel survey of medical use, expenditure, and health status for the civilian noninstitutionalized U.S. population. The primary measure of effect is based on self-perceived mental health status. Secondary measures of effect include functional limitations, instrumental activities of daily living (IADL) limitations, social limitations, and cognitive limitations. Cost-effectiveness is determined by estimation of the incremental cost-effectiveness ratios and construction of bootstrapped cost-effectiveness acceptability curves (CEAC). The incremental net benefit (INB) method is estimated to investigate potential self-selection bias using observable characteristics, propensity score matching, inverse propensity score weighting, and the primary sampling unit complementary and alternative medicine use as an instrument. Results and discussion. The evidence suggests a high probability that CAM is cost-effective for a large range of values of effect for anxiety and neurotic disorders. The evidence strongly suggests that CAM is not cost-effective for depression (NOS) disorders. The evidence suggests that CAM is more cost-effective for user of psychotherapy than for users of pharmacotherapy. The effectiveness of CAM treatment is equivalent to 0.75--1.02 less days of work missed for individuals with neurotic disorders using psychotherapy. The effectiveness of CAM treatment is equivalent to a 2.99--4.64 increase in the EQ-5D Index score for individuals with neurotic disorders using psychotherapy. Attempts to mitigate potential self-selection bias in this observational data indicate an upward bias in the estimate of incremental net benefit for the anxiety and neurotic disorders. However, for these disorders, the psychotherapy samples are least affected by self-selected bias. Conclusion. CAM is a cost-effective additional treatment for anxiety and neurotic disorders. CAM is more cost-effective for psychotherapy users than for pharmacotherapy users. CAM is not a cost-effective addition to treatment for depression (NOS) disorders. The net benefit method for estimating cost-effectiveness analysis provides a useful framework in which common econometric techniques of addresses self-selection can be incorporated. In general, survey data provide another type of data that policy-makers can use to inform decisions regarding adoption of new treatments.