Differences in successful treatment completion among pregnant and non-pregnant American women

The present study explores characteristics of successful substance abuse treatment completion of pregnant women through an analysis of retrospective outcomes data. Women without prior treatment admissions, aged 18-44, and not in methadone maintenance therapy were included (N = 678,782). Chi-square tests analyzed significant differences; logistic regression provided predictive probabilities; odds ratios (OR) and risk differences with 95 % confidence intervals represent the effect sizes and clinically meaningful differences. Pregnant women were less likely to successfully complete treatment than non-pregnant women (χ (2) = 321.33, df = 1, p < 0.0001), though the difference was not clinically meaningful (risk difference = 4.75, 95 % confidence interval (CI) = 4.23-5.26). Aside from criminal justice agencies, "other community agencies" refer the greatest percentage of pregnant women to treatment (risk difference = 6.37, 95 % CI = 5.89-6.84). Pregnant women successfully complete treatment more than non-pregnant women in only non-intensive outpatient settings (χ (2) = 10,182.48, df = 7, p < 0.0001). Further attention to referral source and treatment setting for pregnant women may improve successful treatment completion by targeting needs of pregnant women. Referring to non-intensive outpatient and residential hospital treatment settings may help to ameliorate prenatal substance abuse treatment contingent on the primary problem substance.