Psychosocial mediators of dietary change among Hispanic/Latina breast cancer survivors in a culturally tailored dietary intervention

Objective: To examine psychosocial mediators of the effect of a culturally tailored dietary intervention on dietary change among Hispanic/Latina breast cancer survivors. Methods: Hispanic/Latina breast cancer survivors (n = 70) were randomized to receive either a 12-week theory-based and culturally tailored dietary change program (intervention group, n = 34), or standard-of-care printed recommendations (control group, n = 36) (NCT01414062). Fruit/vegetable intake (F/V), % calories from fat, and hypothesized psychosocial mediators were assessed at baseline, 6 and 12 months. Analysis of covariance assessed intervention effects on psychosocial mediators at 6 and 12 months. Mediation analysis using the bootstrap method evaluated the indirect intervention effects on dietary intake at 6 and 12 months through changes in psychosocial mediators at 6 and 12 months. Results: Compared with controls, at 6 and 12 months, the intervention group reported greater improvements in stages of change (P < .001, P < .001, respectively), self-efficacy (P = .009, P = .002, respectively), snack preference for F/snack preference for F/V (P = .045, P = .002, respectively); at 12 months, the intervention group reported a decrease in chance-oriented external locus of control (P = .02). At 6 months, mediation analysis showed that the intervention effect was associated with an increase of 1.0 (95% CI, -0.1-2.4) serving/day of F/V, compared with the control group, although no indirect effect through the hypothesized psychosocial mediators was observed. At 12 months, the intervention was associated with an increase in 0.5 serving/dayF/V through improved taste/snack preference for F/V at 6 and 12 months (95% CIs, 0.1-1.3, 0.0-1.4, respectively). Conclusions: Future programs can target improving taste/snack preference for F/V to promote dietary change in Hispanic/Latina breast cancer survivors.