Abstract

Study participation and retention in the time of COVID

Ashlin Ondrusek¹, Adam Viera, MPH¹, Emily L. Vinson, MPH², Jessica Legge Mulienburg, PhD², Trace Kershaw, PhD¹ (1)Yale School of Public Health, (2)University of Georgia College of Public Health

APHA 2022 Annual Meeting and Expo

Aim: To describe factors related to retention in a multisite prospective cohort study in Connecticut and Georgia of return to use among individuals in treatment for substance use disorder who meet criteria for alcohol use disorder. Research shows that there are many sequential factors regarding needs and motivation for treatment that contribute to treatment outcome. Therefore, we’re interested in learning about the ways in which the TCU treatment model applies to participant retention at 3-month follow-up.
Methods: Participants were 136 individuals who completed a baseline visit and were past their 3-month (T2) follow-up visit. Logistic regression models identified the factors associated with attendance of a study visit three months after the baseline visit and participation in daily ecological momentary assessments (EMA). Specifically, predictors assessed included commitment to sobriety, being legally mandated to attend substance use treatment, depression, anxiety, health care utilization, pain rated according to the visual analog scale, substance use treatment commitment and engagement, substance use behavior, perceived COVID susceptibility and severity, as well as demographic variables of age, race, study site, and gender.
Results: Results suggest that participants who were legally mandated to attend treatment were 43% as likely to show up to T2 as compared to individuals who were not legally mandated (B= -.276, [-.444, -.111]). Similarly, older participants appeared to be more likely to attend compared to younger (B= 1.050, [1.014, 1.087]). Unsurprisingly, participants who completed more EMA surveys over the course of the three months were significantly more likely to attend T2 (p < .05, [-.397, -.211]). While all substance use was assessed, only crack use and prescription stimulant use were significantly predictive of participants not attending T2 (B= 2.106, [1.047, 4.237], and B=.451, [.220, 923] respectively). Notably, race and study site appeared to be significantly predictive of T2 attendance (both p < .05). The variables of commitment to sobriety, treatment commitment and engagement, depression, anxiety, visual analog scale, gender, and perceived COVID severity and susceptibility did not appear to be significantly predictive of T2 attendance.
Conclusion: These data suggest that participants who were more mature in age, not legally mandated to attend substance use treatment, and completed EMA surveys more often were more likely to attend T2. Future qualitative research aims to assess if participants who viewed the EMA surveys and study participation as parallel to substance use treatment influenced attendance.