Influence of organizational and client factors on fidelity of an HIV evidence-based program

The emphasis on evidence-based practice requires attention to fidelity in translation of programs. Few studies have assessed fidelity of HIV/STI behavioral counseling and testing programs, including RESPECT. We examined client fidelity for 26 public health departments (HD) and community-based organizations (CBO) delivering RESPECT across the U.S.. Using anonymous client reports on the counseling session (N=808) we constructed agency-level fidelity scores. Seventy percent of agencies demonstrated high fidelity (range: 0-6). We examined client (i.e., demographics, sexual orientation, risk) and agency (i.e., locale, type) characteristics with regard to fidelity. Variables significant at the univariate level were included in multivariate analyses. Multivariate logistic analyses, controlling for clusters, and client risk (OR=1.9; 95% CI= 1.4, 2.6), showed that type of organization (CBO vs. HD; (OR=2.5; 95% CI= 1.3, 4.7)) and client race/ethnicity were significantly associated with high fidelity ((OR= 2.6, CI= 1.8, 3.8) Hispanics vs. whites). Trends were found for repeat testing (OR=1.4, CI .99-1.9) and being African American vs. white (OR= .16, CI .95-2.8.). As expected, results indicated that program delivery was adapted for low risk clients. Unexpectedly, fidelity was lower for whites and for clients attending HD clinics. Qualitative findings on adaptations reported by program staff (n= 62), offer additional insight into adaptations (e.g., time constraints, client factors). Together these findings suggest that: (a) many agencies deliver RESPECT well, but one-third have moderate to poor fidelity, (b) staff may use client stereotypes in adapting the program in ways that impact fidelity, and (c) organizational factors (e.g., training, time constraints) may influence fidelity.