Evaluating women veterans seeking care from substance use treatment facilities
Open AccessBackgroundPrevalence of substance use disorder (SUD) among women veterans is increasing, as well as an association with comorbidities; post-traumatic stress disorder (PTSD), anxiety, depression, concurrent benzodiazepine use, chronic pain, military sexual trauma (MST), sleep disorders, and suicide have been found to be associated with SUD. Women are the fastest growing group among veterans, but there are few treatment studies that focus on the women veteran population. As a result, the need for research to better understand this population is necessary to identify programs and resources needed to accommodate these women’s healthcare needs, especially at the Department of Veterans Affairs (VA). The objectives of this study are the following: compare the differences in treatment enrollment in women veterans compared to non-veteran women; compare the differences in veteran treatment enrollment in women compared to men; and evaluate the factors associated with completion of a treatment program among women veterans. Methods The Treatment Episode Data Set (TEDS) data set is an annual collection of data on substance abuse treatment programs, which has not been previously evaluated in the literature for differences based on veteran status and sex. Deidentified data sets are publicly available through Substance Abuse and Mental Health Services Administration (SAMHSA) for the years 2006-2017 and include over 19 million records. Descriptive statistics for the population, as well as treatment use, and treatment completion was calculated with chi-square tests to identify significant differences between groups by veteran status among women and sex among veterans. Logistic regression models tested sex and veteran status and the association with variables for pre-enrollment factors, demographic information, and discharge information. The following variables were included as predictors for the model: age, race/ethnicity, marital status, education, employment status at admission, pregnancy, substance use type, length of stay in treatment, treatment referral source, number of days waiting to enter treatment, previous substance use treatment episodes, frequency of use of primary substance at admission, health insurance at admission, homeless at admission, co-occurring mental and substance use disorders, number of arrests in 30 days prior to admission, and age at first use of primary substance. A multivariate logistic regression tested veteran and non-veteran treatment completion rate differences among women, as well as sex differences in treatment completion among veterans. The overall fit of the models was tested and used to select the final model. An interaction between age and veteran status was tested based on the possible age differences between veteran and non-veteran women, and an interaction between age and sex was tested based on the possible age differences between men and women veterans. The sample size of the study includes 58,163 women veterans and a power analysis indicated there is a sufficient sample for this study. A path analysis was performed to identify patterns of correlation among variables and explain variable variance within the model for completion of SUD treatment programs. The model was specified and identified, and structural equation modeling (SEM) estimated and tested the model parameters and coefficient estimates. The model was tested for fit using a Chi-squared goodness-of-fit test and other measures of overall fit. Additional fit analysis and modification indices were calculated, and the model was re-specified. Results After controlling for other covariates, a multivariate model found women veterans have 1.16 (95% CI: 1.12 – 1.20) times greater odds of completing SUD treatment compared to non-veteran women. Increased length of stay in treatment was also associated with completing SUD treatment. Women in a program for 181 days to a year had 4.44 (95% CI: 4.38 – 4.49) times the odds and more than a year had 4.05 (95% CI: 3.99 – 4.12) times the odds of completing treatment compared to women in treatment for 30 days or less. Women with co-occurring mental health and substance use disorders had 14% reduced odds of completing SUD treatment compared to women without a co-occurring mental health and substance use disorders (OR = 0.86, 95% CI: 0.85 – 0.86). After controlling for other covariates, the adjusted model found women veterans have 0.96 (95% CI: 0.93 – 0.99) times fewer odds of completing a treatment program compared to men veterans. While the results of this study found sex was significantly associated with SUD treatment completion among veterans, the small effect size suggests there is no clinical relationship between completing SUD treatment and sex among veterans. Veterans in a program for 181 days to a year had 4.60 (95% CI: 4.42 – 4.79) times the odds and more than a year had 4.12 (95% CI: 3.90 – 4.34) times the odds of completing treatment compared to veterans in treatment for 30 days or less. Veterans with co-occurring mental health and substance use disorders had 19% reduced odds of completing SUD treatment compared to veterans without a co-occurring mental health and substance use disorders (OR = 0.81, 95% CI: 0.80 – 0.83). The path analysis measures of overall fit for the final model indicated an acceptable fit for the theoretical model. The final model confirms parts of the manifest variables contribute to length of stay in treatment, which plays a significant role in the completion of treatment. Conclusion While veteran women were more likely to complete SUD treatment compared to non-veteran women, when compared to veteran men, women veterans had slightly lower odds of completing SUD treatment. This significant but small difference demonstrates the sex difference among veterans enrolled and completing SUD treatment. Though women-only treatment for veterans may not be clinically better than combined-gender treatment programs, understanding the underlying characteristics of treatment enrollment and demographic variables may help health systems and programs better address the specific needs of women veterans. By including data outside of the VHA, the overall burden of SUD among women veterans can be truly understood and estimated, while also potentially encouraging cooperation between VA and community SUD treatment programs. An integrated approach to primary care and specialty care; mental health and substance abuse treatment; and referrals between VA and community facilities is likely to improve women veteran’s access to the SUD treatment they need and their likelihood to complete treatment.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.