Article Abstract

Article Abstract

Online ISSN: 1099-176X Print ISSN: 1091-4358
The Journal of Mental Health Policy and Economics
Volume 10, Issue 4, 2007. Pages: 207-219
Published Online: 18 Dec 2007

Cost-Effectiveness of a Mixed-Gender Aftercare Program for Substance Abuse: Decomposing Measured and Unmeasured Gender Differences

Hyong Suk Yeom,¹ Donald S. Shepard²

¹Ph.D., Department of Social Work, JamesMadisonUniversity, Harrisonburg, VA, USA
²Ph.D., HellerSchool of Social Policy and Management, BrandeisUniversity, Waltham, MA, USA

* Correspondence to: Hyong Suk Yeom, MSC 4303, Health & Human Services Room 2143, Department of Social Work, James Madison University, Harrisonburg, VA 22807, USA.
Tel.: +1-540-568-8087
Fax: +1-540-568-7896
E-mail: yeomhs@jmu.edu

Source of Funding: Supported by NIDA Grant DA 08739 to BrandeisUniversity.

Abstract

Most substance abuse treatment programs are mixed-gender programs targeted largely at male substance abusers. This study analyzed gender differences in the cost-effectiveness of a mixed-gender aftercare substance abuse program in Westborough, MA, where 78 females and 141 males enrolled from 1994 to 1996. The analysis first used the generalized estimating equations (GEE) model to assess gender differences on the outcome of ‘any substance use’ and the ‘overall cost.’ It then used an extension of multivariate cost-effectiveness analysis (CEA) that decomposed the overall gender-based differences into differences on baseline characteristics (measured characteristics) and a pure gender effect (unmeasured characteristics of the clients and the treatment system). This new approach could be used to assess the impact of gender, race, or ethnicity on outcomes and costs in other programs. This study found that the mixed-gender program was more cost-effective for men than women, although the program was reasonably cost-effective for both genders.

Background: Most substance abuse programs are mixed-gender programs targeted largely at male substance abusers. Studies on gender differences in the cost-effectiveness of mixed-gender aftercare programs are rare. Little is known about how beneficial and efficient various aftercare treatments are for either female or male substance abusers in mixed-gender programs.

Aims of the Study: This study analyzed gender differences in the cost-effectiveness of a mixed-gender relapse prevention aftercare program for substance abuse at Spectrum in Westborough, MA. Its aims are to contribute to methods of gender research by decomposing gender effects into measured and unmeasured characteristics, and to guide policy on addressing needs of female and male substance abusers in aftercare programs.

Methods: This study examined data from a NIDA-funded project which enrolled 78 females and 141 males in a mixed-gender aftercare program in MA from 1994 to 1996. It first used the generalized estimating equations (GEE) model and the multivariate cost-effectiveness analysis to assess gender differences in the cost-effectiveness of treatment over two years of follow-up. The approach then allocated impacts on cost and outcomes to each measured characteristic (as the product of its coefficient times the gender difference) and attributed all remaining gender differences to unmeasured characteristics, termed the ``pure'' gender effect.

Results: This aftercare treatment was effective for both men and women on proportion of days with any substance use compared to pre-treatment levels (-0.62 for men and -0.51 for women), but more reduction appeared in men than in women. When the costs were standardized with the multivariate regression including client characteristics covariates, women cost more in aftercare treatment ($4435) than men ($4284). This greater cost for women was due to baseline characteristics associated with higher costs (e.g. less likely to have a profession); female gender per se actually reduced the overall cost for women. Finally, the numerically smaller cost-effectiveness ratio for men ($6894 per abstinent year) than for women ($8620 per abstinent year) implies that aftercare treatment of men was more cost-effective than that of comparable women.

Discussion: After standardization, this study of a mixed-gender aftercare program found that women tended to show less improvement than men in the outcome of any substance use and to be more costly to treat. Overall, aftercare substance abuse treatment of women seems to be less cost-effective than that of men.

Implications for Health Care Provision and Use: This mixed gender program was more cost-effective for men than women, although results were reasonably favorable for both genders. This study's findings might enhance the discussion about which program (gender-specific or mixed-gender) is more cost-effective and how the currently prevalent mixed-gender programs could meet the each gender's needs more effectively.

Implications for Further Research: By identifying a pure gender effect, this study has widened the application of the methods of multivariate cost-effectiveness analysis. This study's approach could be used to assess the impact of a number of client and program characteristics on outcomes. Among client characteristics beyond gender, the ethnic, cultural, or age groups could be compared on the cost-effectiveness of their treatment. Among program characteristics, mixed- or separate-gender programs, faith-based or secular, multi-modality versus single-modality, or larger versus smaller programs could all be compared.

Received 19 June 2006; accepted 8 November 2007