Cluster-randomized trial comparing organizationally tailored versus standard approach for integrating an evidence-based cancer control intervention into African American churches

Abstract

The field of implementation science has devoted increasing attention to optimizing the fit of evidence-based interventions to the organizational settings in which they are delivered. Institutionalization of health promotion into routine organizational operations is one way to achieve this. However, less is known about how to maximize fit and achieve institutionalization, particularly in settings outside of the healthcare system. This paper reports on findings from a parallel cluster-randomized trial that compared an organizationally tailored with a standard (core components only) approach for institutionalizing ("integrating") an evidence-based cancer control intervention into African American churches. Churches randomized to the organizationally tailored condition identified three or more implementation strategies from a menu of 20, with an implementation time frame for each. The primary study outcome was assessed through the Faith-Based Organization Health Integration Inventory, a measure of institutionalization of health promotion activities in church settings, completed by pastors at baseline and 12-month follow-up. Seventeen churches were randomized and 14 were analyzed as 3 did not implement the study protocol. Though the percent increase in total integration score was greater in the tailored condition (N = 9; 18%) than in the standard condition (N = 5; 12%), linear mixed-effect models did not detect a statistically significant group × time interaction. Despite the challenges of integrating health promotion activities outside of healthcare organizations, the current approach shows promise for fostering sustainable health promotion in community settings and merits further study.

Trial registration: ClinicalTrials.gov NCT03178383.

Keywords: Adaptation; Cancer; Implementation; Institutionalization; Organizational context; Sustainability.

Fig 1

Project HEAL 2.0 CONSORT flow diagram. ^aNot fully enrolled reflects churches that agreed to participate but did not complete the enrollment paperwork. A church is fully enrolled when they agree to participate, and complete the church/pastor informed consent and baseline church survey. Churches were randomized when fully enrolled. ^bIn addition to the 5 standard condition churches that implemented the intervention, 3 additional churches were randomized to the standard condition, but did not implement the intervention and were excluded from all analyses. Lost to follow-up = study team was unable to contact church for continued study participation.