The importance of indirect screening and objective gold standards: a response to <scp>T</scp>erplan (2012)

We wish to thank Dr Terplan for his comments 1 on our recent paper 2, and offer here a brief response. Our paper suggests that self-report during a structured interview is a poor gold standard for screener validation when under-reporting is likely. That is, the validity of direct screeners in such contexts may be falsely inflated by comparisons to an overly similar gold standard, resulting in high accuracy values despite the possibility that many may deny drug use during the interview as well as during screening. Using an objective gold standard with a sample of African American women, we demonstrated that a commonly used direct screener failed to identify the majority of those using drugs during pregnancy, and that a rigorously developed indirect screener identified a far greater proportion of at-risk women. Given our clear distinction between self-report and objective gold standards and the very poor performance of direct screening in our sample, it is surprising that Dr Terplan questions the need for a measure such as the Wayne Indirect Drug Use Screener (WIDUS). The direct screener in our study identified only 7% of women using drugs during pregnancy (37% when using its lowest possible cut score). It is difficult to see how this can be considered anything other than unacceptable. Further, the existence of other screeners with similar predictive accuracy is of limited relevance to the WIDUS if those accuracy values were derived against self-report gold standards. Dr Terplan fails to recognize this crucial distinction. Whether or not we need more screeners validated against self-report, we very much need more research on safe, practical ways to identify actual drug use during pregnancy. Dr Terplan's additional concerns relate to what might follow a positive indirect screen. Anticipating such questions, we wrote originally that: ‘Brief interventions that address substance use in a non-specific way, in the context of several other potential health risks, may be helpful … Future research should evaluate whether it is possible for a brief intervention to reduce drug use without directly presuming the presence of that behavior’ ([2], p. 2105). We are in the midst of a clinical trial testing just such an indirect brief intervention. Although outcome data are not yet available, 46 of 52 intervention group participants thus far (85%) report being more likely to make a personal change because of the intervention, with most specifically indicating a desire to reduce drug use in their home; this despite the lack of any presumption of drug use. Further, indirect screening actually presents far less risk to pregnant women than traditional approaches. Endorsing items such as: ‘Most of my friends smoke cigarettes’ is, by design, considerably less prejudicial than a positive urinalysis or direct admission of drug use. It would be irresponsible to simply ignore the many at-risk pregnant women who choose not to disclose drug use, or to respond only with bromides about education and empowerment. Creative approaches can and should be pursued vigorously. None.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.