Relative validity of a brief Fat and Fibre Behaviour Questionnaire in a population of overweight and obese breast cancer survivors: A note of caution.

Abstract

Dietary intake assessment is often difficult in research contexts because of time and resource constraints and participant burden. Valid, reliable and brief assessments of dietary behaviour are needed. Additionally, examination of instrument performance in a variety of populations is needed. This study assessed relative validity and responsiveness to change of the Fat and Fibre Behaviour Questionnaire (FFBQ) in a population of breast cancer survivors compared with dietary intake measured by 24-hour recalls. Data were collected at baseline and six months after baseline of a six-month, randomised controlled trial that evaluated a telephone-based behavioural weight loss intervention (n = 45) compared to usual care (n = 45) among overweight and obese breast cancer survivors. The FFBQ's total index, fat index and fibre index were assessed for relative validity against two 24-hour recalls (Pearson's correlations). Responsiveness to change was assessed as intervention group change divided by standard deviation of usual care group change. Pearson's correlations of the fat index with fat intake measures ranged from -0.09 to -0.30 (P < 0.05) at baseline and from -0.19 to -0.28 (P < 0.05) for dietary change. Correlations of the fibre index with unadjusted and energy-adjusted fibre intake measures (0.25-0.32, P < 0.05) were significant at baseline only. Both the FFBQ and 24-hour recall were responsive to fat-related dietary changes and not responsive to fibre-related changes. The FFBQ showed small to medium relative validity against 24-hour dietary recall for assessing fat and fibre dietary behaviours and changes in dietary fat, and was responsive to fat-related dietary changes in this population.