The theory of planned behavior and implementation intentions to describe and improve fruit and vegetable intake in women of low socioeconomic status

Abstract

OBJECTIVE: The Theory of Planned Behavior (TPB) and implementation intentions have been used effectively to explain and influence diet in middle-class, but not exclusively low SES populations. Furthermore, dietary research among low SES populations requires intake measures that are feasible and acceptable. Using three lines of research we evaluated 1) the utility of the TPB to explain fruit and vegetable (FV) intake, 2) efficacy, feasibility and acceptability of an implementation intention intervention to improve FV intake, and 3) agreement, feasibility, and acceptability of 2 dietary intake measures of FV intake in low SES women. DESIGN: Participants were adult female residents of Boston Public Housing. Study 1: Using a cross-sectional survey (n=144), we evaluated the utility of the TPB to explain FV intake. Study 2: We conducted a pilot randomized controlled implementation intention intervention to promote FV intake (n=20), and semi-structured interviews to evaluate feasibility and acceptability of the intervention (n=8). Study 3: We administered 2 24-hour recalls, a food frequency questionnaire (FFQ), and structured interviews (n=36) to determine agreement, feasibility and acceptability dietary intake measures. RESULTS: Study 1: The TPB construct perceived behavioral control significantly predicted intention to consume FV (OR=2.55,95%CI:1.23,5.27) and with BMI, FV intake (βPBC=0.37,t(1)=2.29,p=0.0235; βBMI= -0.02,t(1)= -2.41,p=0.0174; R2=.08,F[2,130]=5.72,p=0.0042). Study 2: Feasibility goals were met for retention and days to follow up, but not recruitment. Participants characterized the intervention as enjoyable. Limited hypothesis testing showed no significant increase in mean FV intake within (control (n=11):+0.50, 95% CI:-0.56,1.58 servings; intervention (n=9):+0.17, 95% CI:-0.85,1.20 servings) or between groups (control group +0.33, 95% CI:-1.06,1.73 servings). Study 3: Feasibility targets were met for contacts and retention, but not for enrollment. There was no significant association between 24-hr recall and FFQ measures for fruit (r=0.32, p=0.09) or vegetable (r=0.16, p=0.40) intake and no marked preference for method (35% FFQ; 31% 24-hour recall). CONCLUSION: The TPB may be useful to explain FV intake. Although acceptable, an implementation intention intervention may not be feasible or effective to influence FV intake. We demonstrated limited feasibility and association but generally equal preference between dietary measures of FV intake in low SES women.