Gender, Pregnancy, and Treatment Completion by Criminal Justice Referral Status

Background. Court-mandated drug treatment has become more common in the United States over the last decades. However its effectiveness has not been investigated nationally in terms of gender and pregnancy status.

Methods. We used the 2006 Treatment Episode Data Set. The primary outcome was treatment completion including transfers to further care. The primary exposure was criminal justice referrals. Demographic and treatment characteristics were compared with chi-squared and t-tests. Confounding was assessed via a backwards elimination method using change-in-estimate criteria. Logistic regression models were stratified by gender and pregnancy. Results are reported as Odds Ratios (OR) with 95% confidence intervals (CI).

Results. Of the 1.5 million treatment admissions, 38% of men, 29% of women, and 35% of pregnant women entered treatment from the criminal justice system. Overall pregnant women were younger, with lower educational achievement and employment, less likely to use alcohol, and more likely to use methamphetamines. Pregnant women were less likely to complete treatment (55%) than non-pregnant women (57%) or men (61%). After controlling for confounders, the subset of pregnant women (OR 1.37 [1.29, 1.46]) and non-pregnant women (1.25 [1.23, 1.26]) who entered treatment via the criminal justice system had a higher odds of treatment completion than men (0.95 [95% CI: 0.94, 0.96]).

Conclusion. Criminal justice referral appears to be an important effect measure modifier in drug treatment completion. The reasons why pregnant women especially do better in this context needs to be further explored.