Dietary change through African American churches: baseline results and program description of the eat for life trial

Abstract

Background: Eat for Life, a multicomponent intervention to increase fruit and vegetable (F & V) consumption among African Americans, is delivered through African American churches.

Methods: Fourteen churches were randomly assigned to one of three treatment conditions: 1) comparison; 2) culturally-sensitive multicomponent intervention with one phone call; and 3) culturally-sensitive multicomponent intervention with four phone calls. The intervention included an 18-minute video, a project cookbook, printed health education materials, and several "cues" imprinted with the project logo and a 5 A Day message. A key element of the telephone intervention was the use of motivational interviewing, a counseling technique originally developed for addictive behaviors. Major outcomes for the trial included total F & V intake, assessed by food-frequency questionnaires (FFQs) and 24-hour recalls, and serum carotenoids. Psychosocial variables assessed included outcome expectations, barriers to F & V intake, preference for meat meals, neophobia, social support to eat more F & V, self-efficacy to eat more F & V, and nutrition knowledge.

Results: Baseline mean F & V intakes across the three FFQs ranged from 3.45 to 4.28 servings per day. Intake based on a single 24-hour recall was 3.0 servings. Variables positively correlated with F & V intake included self-efficacy, outcome expectations, and a belief that F & V contain vitamins. Factors negatively correlated with intake include perceived barriers, meat preference, neophobia, and high-fat cooking practices. The completion rate for the first telephone counseling call was 90%. Completion rates for the remaining three calls ranged from 79% to 86%.

Conclusion: The recruitment and intervention methods of the Eat for Life study appear promising. The telephone intervention based on motivational interviewing is potentially useful for delivering dietary counseling.