(PS8-23) The Analysis of the Drop out Between Program Interest and Program Initiation Stages in Low-income and Military Couples Seeking Care

In recent years, the utilization of telehealth services has drastically increased (Doraiswamy et al., 2020). This form of treatment, while found to be equally, if not more effective than traditional interventions, has the added benefit of circumventing many of the latter’s structural barriers (Williamson et al., 2019). However, an ongoing challenge of many online interventions is maintaining participant enrollment and engagement (Postel et al., 2010). This issue is only magnified in programs that work with couples, where the active participation of two individuals is required. The present study aims to examine this attrition issue in couples seeking an online program designed to help couples strengthen their relationship.

We examined what traits predict a couple dropping out of the enrollment phase of the OurRelationship (OR) program, an online intervention for couples that has been well-researched and shown to be effective in improving both relationship and individual functioning (Doss et al., 2016, 2020). We analyzed the demographic characteristics of 1,757 low-income and military couples (N= 3,514 individuals; 16% military; 78% low-income; 0.3% both military and low-income) who filled out the initial eligibility screener to enroll in the OR program. Of this sample, 74% (1,299 couples) dropped out after completing the initial eligibility screener but before fully enrolling in the program by attending a Welcome call. The remaining 26% (458 couples) did not drop out during the enrollment phase and completed at least one call with their coach. Using this sample, we ran a logistic regression on six relationship characteristics and four individual characteristics to measure predictors of Welcome call attendance.

Results indicated that, out of the six relationship characteristics measured, same-sex (p = .003, β = .582) couples were 1.790 times more likely to attend their Welcome call than opposite-sex couples and low-income couples (p < .001, β = -.532) were approximately half as likely to attend a Welcome call than couples who did not qualify as low-income. Relationship status, household size, presence of infidelity, and breakup potential were not significant predictors. Out of the individual characteristics measured, couples where both partners indicated military status (p < .001, β = .542) were 1.720 times more likely to attend a Welcome call than civilian couples. Military status was not a significant predictor of Welcome call attendance in couples where only one partner had military status. Additionally, the presence of Intimate Partner Violence, history of injury, and fear of domestic violence were not significant predictors of enrollment.

In this study, the only trait that was a significant indicator of a couple being less likely to enroll in the OR program was low-income status. These results highlight the population that may be the most vulnerable and experience the most barriers to care. Consideration of these findings could be valuable in improving the OR recruitment and enrollment phases and providing more support to vulnerable populations in an effort to increase the reach of care.