UCLA

The coronavirus (COVID-19) pandemic resulted in unprecedented challenges for children due to marked emotional distress, social isolation, and substantial grief and loss. Unfortunately, the rise in mental health challenges for youth due to the pandemic could exacerbate the already high rate of unmet treatment needs of youth (Whitney & Peterson, 2019). The pandemic also transformed the field of mental healthcare due to the rapid expansion of telehealth services to slow the spread of COVID-19. Telehealth or telemental health (TMH) services are defined as services that occur remotely through videoconferencing platforms or telephone calls. Although pre-pandemic research on telehealth has been largely positive, the influence of telehealth on youth treatment engagement is unclear due to inconsistent study findings, use of unidimensional measures of engagement, and low utilization of telehealth services in schools (Brooks et al., 2013; Georgeson et al., 2020; Love et al., 2019). Thus, the COVID-19 pandemic offered an excellent opportunity to examine the influence of telehealth on multiple dimensions of engagement within a traditional care setting for youth.

The objective of this two-study dissertation was to investigate the influence of the pandemic and rapid expansion of telehealth on youth treatment engagement using a multidimensional measurement framework. These studies occurred within large urban school-based mental health program serving youth and families with well documented logistical, cognitive, and systemic barriers to engaging in mental health services. The first study examined whether telehealth circumstances facilitated or degraded engagement when measured using a multidimensional and multi-perspective lens. We compared two demographically matched cohorts of families receiving services prior to the pandemic with those enrolled during the phase of the pandemic when stay-at-home orders were in effect, and schooling and non-essential work occurred remotely (hereafter referred to as the mid-pandemic lockdown period). Results revealed that pre- and mid-pandemic ratings of engagement were similar with the exception that caregivers showed greater risk for low attendance and youth had lower expectations about the benefits of mental health treatment within the mid-pandemic lockdown context. We additionally found that the type of delivery mode (i.e., telephone services, videoconferencing services, or a combination of both) was associated with some aspects of engagement for youth, whereas factors unrelated to telehealth were associated with caregiver ratings of engagement during the mid-pandemic lockdown period. The overall results of Chapter 1 suggest that utilization of telehealth services may represent an option for families that does not appear to negatively impact their engagement in services.

Chapter 2 was exploratory and used a mixed methods approach to investigate school-based mental health providers’ early impressions of youth treatment engagement for families enrolled in mental health services during the mid-pandemic lockdown period. A coding system was applied to describe providers’ impressions of low and high engagement indicators based on a multidimensional framework. Multilevel logistic regression models were conducted to examine providers’ abilities to detect engagement challenges and pandemic factors that may influence provider reports of engagement. The qualitative results revealed that providers’ impressions of engagement were substantially more positive than what is described in the literature. The most frequent indicators of engagement reported by providers were associated either with overt behaviors or with ambiguous descriptions that were difficult to code. Use of a multidimensional measurement framework revealed that providers generally appeared to under-detect engagement challenges and use behavioral observation to make engagement inferences. These findings underscore that the telehealth context does not appear to facilitate or degrade providers’ detection abilities, which are generally low, possibly due to a lack of shared vocabulary of engagement indicators, suboptimal assessment strategies regardless of the service context, and/or an over-reliance on the delivery mode as an indicator of engagement in the mid-pandemic lockdown context. Overall, the results of Chapter 2 suggest that providers will benefit from receiving additional supports to enhance early identification of engagement problems and reduce the alarmingly high rates of premature termination.