(When) Is More Better? Comparative Effectiveness of External Vs External+Internal Facilitation on Site‐Level Uptake of a Collaborative Care Model in Community‐Based Practices That Are Slow to Adopt

Research Objective Implementation scientists generally acknowledge that low‐level implementation support will not suffice supporting adoption of evidence‐based practices (EBPs), but little work has examined the best way to step up support for sites that are slow to implement. Facilitation, that is, interactive strategic thinking support for providers to encourage uptake of EBPs, is often identified as a key component of successful implementation. Several models of facilitation, of varying intensity, have been operationalized, notably models involving external facilitation (EF) versus those combining external with internal facilitation (EF/IF). Relative to EF, IF is thought to better embed the EBP into the practice setting and encourage transformational rather than transactional change. However, few studies have examined whether EF/IF improves uptake of evidence‐based practices relative to EF alone, and/or whether certain sites benefit most from IF. This study examines the comparative effectiveness of EF vs. EF/IF on uptake of an evidence‐based collaborative care model (CCM) at sites not responsive to low‐level implementation support. Study Design Providers at participating sites received training and technical support to implement a CCM using the Replicating Effective Programs (REP) implementation strategy. After 6 months, nonresponding sites (<10 patients receiving “adequate dose” of CCM) were randomized to augment REP with either EF (regularly scheduled calls by an implementation expert emphasizing strategic thinking) or EF/IF (EF + protected time for an onsite clinical manager to work on encouraging CCM uptake). We hypothesized that EF/IF would outperform EF on uptake (number of patients receiving the CCM) and would work best at larger sites that might require more internal advocate‐led support to facilitate adoption and sites that showed little uptake of the CCM under REP alone. Mixed‐effects models evaluated the 12‐month difference in uptake at EF vs. EF/IF sites. Moderators analyses further examined whether comparative effectiveness was dependent on prerandomization uptake, number of trained providers, and site size as moderators. Population Studied The Adaptive Implementation of Effective Programs Trial (ADEPT) tested EF vs EF/IF with providers at 59 community‐based mental health clinics in Michigan and Colorado. Principal Findings Overall, 43 (73%) sites were nonresponsive after 6 months; 21 were randomized to EF and 22 to EF/IF. As hypothesized, EF/IF sites saw more uptake than EF sites after 12 months (ΔEF/IF‐EF = 4.3 patients, 95% CI = 1.8‐6.7). Moderators analyses, however, revealed that it was only sites with no prerandomization delivery that saw significantly more benefit from EF/IF (ΔEF/IF‐EF = 6.9; CI = 3.5, 10.3). For sites where providers delivered the CCM prior to randomization, EF/IF offered no additional benefit (ΔEF/IF‐EF = −1.68; CI = −5.1, 1.7). Providers trained and size were not significant moderators. Conclusions Our findings show that while EF/IF did outperform EF overall, its benefit was limited to those sites that failed to achieve program delivery under REP alone. Once one or more providers were delivering the CCM, additional on‐site personnel did not add value to the implementation effort. Implications for Policy or Practice Implementation efforts often assume that more support will ensure more support. Our findings, however, suggest that this is not necessarily true and that more intensive implementation support might only serve to increase the perceived burden of EBP implementation. Primary Funding Source National Institutes of Health.