233 size of 210 patients and 42 HCPs. Due to COVID-measures and low recruitment in three organizations, we did not manage to meet this sample size. Because of that, our statistical analyses may have lacked the power to estimate the effect of the intervention. Second, we noticed the health literacy levels of the patients with LHL were often close to the cut-off point of ≤25 at the All Aspects of Health Literacy Scale, our used screener[56]. This indicates participants with very low health literacy, and language and reading barriers were underrepresented. In turn, this may have led to less insight in the barriers of people in these groups in the qualitative studies, and may influence the generalizability of the findings on effectiveness of GoYK in the quasi-experimental study. Third, our sample may have consisted of HCPs with a specific interest in health literacy and of patients who were more motivated. If this is true, the health literacy competences of HCPs at T0 may be overestimated, which suggests GoYK might have a larger effect when it is more widely implemented. For patients, this would explain the ceiling effect we found. Quality of the information The strengths and limitations regarding the quality of information vary for the different methods used. Therefore, below, we discuss the quality of information in our systematic review, and qualitative and quantitative studies separately. In our systematic review, a first strength regards the use of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) checklist to guide our search, study selection, data extraction and reporting[57]. A second strength is that we did an evidence synthesis, which is not always done in reviews. This helps to segregate mediating factors with strong evidence from factors with more uncertainty, which helps identify promising intervention targets. A limitation in our evidence synthesis relates to our classification system. We considered evidence as strong when at least three studies reported consistent findings about one potential mediating factor. A stricter or more liberal classification system potentially would have influenced what we consider factors with strong evidence. Although, we do not expect our main findings to change much. Another limitation is that we did not include grey literature, and only included studies using validated measures on health literacy. With this approach, we may have missed promising intervention targets and strategies.
RkJQdWJsaXNoZXIy MTk4NDMw