Process evaluation of five tailored programs to improve the implementation of evidence-based recommendations for chronic conditions in primary care

BACKGROUND Although there is evidence that tailored implementation strategies can be effective, there is little evidence on which methods of tailoring improve the effect. We designed and evaluated five tailored programs (TPs) each consisting of various strategies. The aim of this study was to examine (a) how determinants of practice prioritized in the design phase of the TPs were perceived by health care professionals who had been exposed to the TPs and whether they suggested other important determinants of practice and (b) how professionals used the offered strategies and whether they suggested other strategies that might have been more effective. METHODS We conducted a mixed-method process evaluation linked to five cluster-randomized trials carried out in five European countries to implement recommendations for five chronic conditions in primary care settings. The five TPs used a total of 28 strategies which aimed to address 38 determinants of practice. Interviews of professionals in the intervention groups and a survey of professionals in the intervention and control groups were performed. Data collection was conducted by each research team in the respective national language. The interview data were first analyzed inductively by each research team, and subsequently, a meta-synthesis was conducted. The survey was analyzed descriptively. RESULTS We conducted 71 interviews; 125 professionals completed the survey. The survey showed that 76 % (n = 29) of targeted determinants of practice were perceived as relevant and 95 % (n = 36) as being modified by the implementation interventions by 66 to 100 % of professionals. On average, 47 % of professionals reported using the strategies and 51 % considered them helpful, albeit with substantial variance between countries and strategies. In the interviews, 89 determinants of practice were identified, of which 70 % (n = 62) had been identified and 45 % (n = 40) had been prioritized in the design phase. The interviewees suggested 65 additional strategies, of which 54 % (n = 35) had been identified and 20 % (n = 13) had been prioritized, but not selected in the final programs. CONCLUSIONS This study largely confirmed the perceived relevance of the targeted determinants of practice. This contrasts with the fact that no impact of the trials on the implementation of the recommendations could be observed. The findings suggest that better methods for prioritization of determinants and strategies are needed. TRIAL REGISTRATION Each of the five trials was registered separately in recognized trial registries. Details are given in the respective trial outcome papers.