African American Veterans Storytelling: A Multisite Randomized Trial to Improve Hypertension

Abstract

BACKGROUND: Disparities in hypertension control persist for African American Veterans. OBJECTIVE: To enhance cultural relevance of hypertension control self-management education, in a multisite, stratified randomized trial, we tested an interactive Veteran-to-Veteran storytelling digital video disk (DVD) intervention created with Veteran partners, versus an education-only DVD comparison. METHODS: At 3 VA facilities, African American Veterans with uncontrolled hypertension were randomized to storytelling DVD intervention or didactic comparison DVD and followed for 6 months. We hypothesized that follow-up blood pressure (BP) would be lower among Veterans in the intervention group. RESULTS: African American Veterans (N=619) were 92% male, 39% over age 65, most had a high-school education, over 50% of both the intervention and comparison group reported a household income of < $20,000, and 40% had less than adequate health literacy. At baseline, mean intervention and comparison group BPs were 138/80 and 139/81 mm Hg, respectively; at 6 months, mean systolic BP (SBP) increased by 0.1 in the intervention group and by 1.9 for the comparison group; diastolic BP fell by 0.1 in the intervention group and rose by 0.2 in the comparison group. No differences between the intervention and control groups were statistically significant (all P > 0.3). Site differences were large; at one, the intervention group improved while the comparison groups deteriorated, resulting in 6.3 and 3.9 mm Hg more improvement for the intervention group in SBP and diastolic BP (P=0.06 and 0.04), respectively; at the other 2 sites, there were positive and negative changes, all small, in the 2 measures, with minimal differences-one site favored the comparison group and the other, the intervention (these comparisons did not approach statistical significance (all P > 0.20). In a secondary analysis stratified by baseline BP, there was no intervention effect among those with uncontrolled BP, but intervention patients who were in control at baseline were more likely to stay in control, compared with comparison [SBP increase by 6.3 mm Hg (SD=14.4) among intervention, and by 10.9 mm Hg (SD=16.9) in comparison, P=0.02]. CONCLUSIONS: In this multisite trial, we did not find a significant overall storytelling intervention effect (Clinicaltrials.gov Reg. #NCT01276197).