Psychosocial intervention to promote fruit and vegetable intake among at-risk women: moderation by intention stability

Abstract

Background: Intention stability can influence implementation intentions’ (II) efficacy to increase physical activity. The objective was to verify if this applies to an II intervention promoting fruit and vegetable intake among women at risk for gestational diabetes mellitus. Methods: The intervention was a randomised controlled trial with 3- and 6-month follow-ups. Women were randomised to an II (n=24) or a question-behaviour effect (QBE, n=26) condition. Fruit and vegetable intake was measured using a validated questionnaire. Intention stability was operationalised as the absolute difference in mean scores of intention between 3 months and baseline and between 6 months and baseline. Participants whose intention differed (>0) between assessments were considered unstable intenders. Mixed model linear regression was used to perform the moderation analysis with participants specified as the random variable. Findings: At baseline, the II and QBE groups were equivalent in terms of intention and vegetable intake (p>0.05). At 3 and 6 months, the II group significantly increased its vegetable intake compared to baseline (p=0.0102 and p=0.0223, respectively). Intention instability was higher in the II compared to the QBE group at 3 months (0.79±0.87 and 0.33±0.37, p=0.0313), but not at 6 months (0.51±0.52 and 0.36±0.34, p=0.2451). There was a significant time×intention stability interaction (β=0.77, p=0.0354), indicating that stable and unstable intenders differed in their vegetable intake across time. At 3 months, unstable intenders (3.22±1.29 servings/day) had a higher vegetable intake compared to stable intenders (2.66±1.15 servings/day). Discussion: It seems unstable intenders would most benefit from a psychosocial intervention to improve vegetable intake.