Fat and fibre behaviour questionnaire: Reliability, relative validity and responsiveness to change in Australian adults with type 2 diabetes and/or hypertension

AimThe aim of this study was to assess the test-retest reliability, relative validity and responsiveness to change of a 20-item fat and fibre behaviour questionnaire (FFBQ) compared with a semi-quantitative food frequency questionnaire (FFQ). MethodsFour hundred thirty-four primary care patients with type 2 diabetes and/or hypertension were randomised to lifestyle intervention or usual care. Data (FFBQ and FFQ) were collected at baseline and 4 months follow-up. A subset of usual care participants (n = 27) completed the FFBQ 2 weeks after baseline. FFQ-measured nutrients included total and saturated fat (percentage of energy; energy-adjusted grams) and fibre (crude grams; energy-adjusted grams). The FFBQ was scored as a total, fat and fibre index. Test-retest reliability was assessed by intra-class correlations (ICCs); relative validity by Pearson's correlations between the FFBQ and FFQ; and responsiveness (mean change in intervention group divided by SD of usual care group change). ResultsTest-retest ICCs (95% CI) were 0.87 (0.62, 0.95) for total index, 0.87 (0.69, 0.95) for fat index and 0.89 (0.69, 0.95) for fibre index. Pearson's correlations for fat index ranged from -0.50 to -0.54, and for fibre index from 0.42 to 0.56. Total index had correlations of 0.5-0.6 (or -0.5 to -0.6) with all FFQ-measured nutrients except crude fibre. The fat index showed similar responsiveness compared with FFQ-measured fat intake while the fibre index was significantly more responsive than FFQ-measured fibre. ConclusionsThe FFBQ may provide a suitable alternative or adjunct to FFQs for use in intervention trials in adults with chronic diseases, because of its ability to detect changes in dietary behaviours.