Validation of a prenatal alcohol and drug screening tool: 4 P's Plus

The purpose of this study was to evaluate the performance of a prenatal substance abuse screening instrument, 4 P's Plus, against a standardized diagnosis of substance abuse, dependence, or problem use, based on DSM-IV criteria, in a population of pregnant women in a managed care setting. The hypothesis was that 4 P's Plus accurately identifies pregnant women at risk for substance use. Pregnant women (N=1,910) were recruited from 2 prenatal clinics: 1,036 who screened positive for substance use on 4 P's Plus and 874 who screened negative. All recruited women completed an in-person assessment with a substance abuse specialist and were categorized as positive or negative for risk of substance use during pregnancy based on standardized criteria. Validation consisted of estimating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (PPV). There were significant demographic differences between cases and controls; cases were more likely to be white with low income and less likely to have completed college and be married. Sensitivity was 87.4% (95% CI 82.1%-91.6%), specificity=49.8% (47.4%-52.2%), positive predictive value(PPV)=17.5% (15.2%-19.9%) and negative predictive value(NPV)=97.0% (95.7%-98.1%). Although the sensitivity and NPV were both high, the specificity and PPV were relatively poor. The 4 P's Plus was purposely designed to have high sensitivity and NPV. Since no amount of substance use is safe during pregnancy, the goal of 4 P's Plus is to identify all pregnant women at risk. When sensitivity and NPV are stressed, specificity and PPV are sacrificed, but from a public health perspective, that is reasonable.