Evaluating two implementation approaches to support a nationally mandated caregiver skills training in veterans affairs: a type III hybrid implementation-effectiveness randomized controlled trial.

BACKGROUND

Few caregiver support interventions have been scaled nationally; thus, gaps remain regarding the optimal level of support needed to successfully adopt caregiver trainings in diverse clinical contexts. We evaluated the effectiveness of two implementation support approaches on implementation outcomes for a family caregiver training (Caregivers FIRST) in partnership with the Veterans Health Administration Caregiver Support Program (CSP).

METHODS

In a hybrid type III effectiveness-implementation trial, we enrolled 25 non-adopter sites six months following the national mandate Caregivers FIRST to compare two different implementation support approaches (foundational support or foundational support with the addition of group external facilitation (enhanced support)). Co-primary site-level penetration outcomes at 12-months were 1) percentage of caregivers attending ≥ 1 class among those with any caregiver referral or enrolled in CSP and 2) number of classes delivered. Secondary outcomes were fidelity and adoption. Fidelity was proportion of expected classes delivered of those promised; adoption was delivering at least one round of classes to 5 or more caregivers. Secondary analysis examined outcomes at six and 18-months separately. We fit generalized linear models to examine differences in outcomes between arms. Survey, interview, and administrative data were used to contextualize findings and assess implementation support costs.

RESULTS

Twelve sites were randomized to foundational support and 13 to enhanced support. At 12-months, the estimated mean percentage of qualifying caregivers attending ≥ 1 class was 8.3% and mean number of classes offered was 9.4 for enhanced support and in foundational support were 6.1% and 5.4, respectively. No differences were found by arm at 12-months in mean percentage of caregivers attending a class (mean difference = 2.2%; 95% CI -1.8%,6.1%; p = 0.26) or mean number of classes offered (rate ratio = 1.7; 95% CI 0.9,3.4; p = 0.11). Neither fidelity nor adoption were different by arm at 12-months, and similar patterns were found at additional timepoints. The added cost of delivering enhanced support per site was $3,678 over a one-year period (median foundational =$14,985 vs. enhanced =$18,663).

CONCLUSIONS

Foundational support allowed most sites (75% foundational, 100% enhanced) to successfully adopt by 18 months. While not reaching a priori benchmarks in magnitude, observed means of implementation outcomes at enhanced support sites were higher at 12 months.

TRIAL REGISTRATION: This study was registered on April 8, 2022 at ClinicalTrials.gov (identifier NCT05319535). https://clinicaltrials.gov/ct2/show/NCT05319535 .