Abstract B41: Effect of community health worker (CHW) interventions on rates of screening mammography: A systematic review and meta-analysis

Abstract

Abstract Background: Community health workers (CHWs) are lay individuals who are specially trained to serve as liaisons between members of their communities and healthcare providers and services. Research conducted on the effectiveness of CHW programs in promoting screening mammography has reported conflicting results (i.e., some studies favored the use of CHWs while some do not). The aim of this study was to conduct a synthesis of all prospective studies on the effectiveness of CHW programs in improving screening mammography rates using the techniques of systematic-review and meta-analysis. It was hypothesized that study participants who received a CHW intervention would be more likely to receive a screening mammogram when compared to participants who did not receive a CHW intervention. Methods: Any prospective study conducted in the U.S. assessing the effect of a CHW intervention in a controlled setting and enrolling a minimum of 10 participants was eligible for inclusion. A comprehensive search of CINAHL, MEDLINE, PsycINFO, and Web of Science databases along with the reference lists of published articles was performed to identify potentially relevant studies conducted and published between 1980 and January 31, 2008. All retrieved citations were reviewed by two researchers for inclusion. Data extraction was performed using a standardized data collection form. Dichotomous data (receipt of mammogram) were summarized as risk ratios (RRs) for each study. The summary results from each study were pooled under a random effects model. Heterogeneity was assessed both by visual inspection of forest plots for each outcome, as well as by formal statistical testing for heterogeneity using the chi square and I2 test. Robustness of findings was assessed according to characteristics of CHW interventions, methodological quality of included studies, or study design. Results: Initial search yielded 228 citations of which 24 studies met the pre-determined inclusion criteria. Of the 24 studies, 18 had extractable data. Altogether these 18 studies included a total of 26,660 participants at baseline. In a combined sample of RCTs (n=10) and quasi-experimental studies (n= 8), participation in a CHW intervention was associated with a small, but significant increase in receipt of screening mammography [RR: 1.06 (favoring intervention); 95% Confidence interval (CI: 1.02, 1.11), p = .003]. The effect remained when pooled data from only the RCTs were included in the meta-analysis (RR: 1.07, 95% CI: 1.03, 1.12, p < .001), but was not present using pooled data from only the quasi-experimental studies (RR: 1.03, 95% CI: 0.89, 1.18, p = .71). Sensitivity analyses using data from the RCTs indicated that CHW programs where participants were recruited from medical settings (RR: 1.41, 95% CI: 1.09, 1.82, p = .008), programs conducted in urban settings (RR: 1.23, 95% CI: 1.09, 1.39, p = .001), and programs where CHWs were matched to intervention participants on race or ethnicity (RR: 1.58, 95% CI: 1.29, 1.93, p < .001) demonstrated stronger effects on increasing mammography screening rates. Conclusions: Results from the meta-analysis indicate that CHW interventions are an effective strategy for increasing screening mammography, but intervention effectiveness depends on the research design of studies, as well as characteristics of the CHW intervention. Citation Information: Cancer Epidemiol Biomarkers Prev 2010;19(10 Suppl):B41.