Screening, enrollment, and assessment in the SMART-ED study.

Implementation of a complex clinical trial in the emergency department (ED) is challenging, as the value of rigorous assessment and internal validity must be balanced against practical constraints of the setting. This presentation describes key features of implementation in the ED for the National Drug Abuse Treatment Clinical Trials Network’s “Screening Motivational Assessment and Referral to Treatment in Emergency Departments” (SMART-ED, CTN-0047) study, including staffing, timing of assessments, integration with the ED, screening procedures and data collection, and provides baseline data on the screened and randomized samples, including drug and alcohol use data and demographic characteristics. Participants were screened using a composite measure including items from the Heavy Smoking Index, the AUDIT-C, and the DAST-10. Participants were excluded unless they had a DAST score of 3 or more and reported use of their self-identified primary problem substance in the past 30 days. Participants in the SAR and BI-B groups received further assessment using the NM-ASSIST and time-line follow-back. Hair testing was used as an objective measure of substance use. 14,972 participants completed the screening instrument, of whom 4005 (27%) reported past 30-day drug use and 1285 (8.6%) were randomized. Within the randomized sample, primary problem substances were cannabis (44%), cocaine (27%), opioids (22%), methamphetamine (4%), sedatives-hypnotics (2%), and hallucinogens (1%). These frequencies varied markedly among the six sites. Participants used their primary substance an average of 16 out of the past 30 days. Of the randomized participants, 70% were male, and mean age was 36 +/- 12. Fifty-percent where white, 35% were black, and 24% were Hispanic. Educational level was low, with 32% having 1-11 years of schooling and only 9% being college graduates. Only 9% were married, 19% had full-time jobs, and 42% were unemployed. 63% had household incomes below $15,000.

Conclusions: Study procedures identified a sample that was diverse with respect to substance of abuse and ethnicity, used drugs frequently, and had very low socioeconomic status.

Related protocols: CTN-0047