Thesis (Ph.D.)--University of Wollongong, 2004. / Includes appendices. Typescript. Includes bibliographical references: leaf 154-184.

Thesis (M. Med. Sc.)--University of Hong Kong, 2005.

published_or_final_version / Medical Sciences / Master / Master of Medical Sciences

Background: This study was conducted to determine the association between fiber intake and insulin resistance in 264 women using a cross-sectional design. Methods: Insulin resistance was indexed using HOMA-IR (fasting insulin (μU/mL)×fasting glucose (mg/dL)/405). HOMA-IR values were log transformed. Fiber and energy consumption were assessed using 7-day weighed food records. Fiber was expressed as grams per 1000 kilocalories. Body fat percentage (BF%) was measured using the BOD POD and physical activity (PA) was ascertained using Actigraph accelerometers worn for 7 consecutive days. Results: (Mean±SD) age: 40.1±3.0 years, glucose: 86.7±5.9 mg/dL; insulin: 7.1±4.3 μU/mL; HOMA-IR: 1.5±1.0; fiber intake (g/1000 kcal), total: 9.3±2.9; soluble: 1.7±0.9; insoluble: 3.8±1.9; physical activity: 2.7044 ±0.7842 million counts; BF%: 31.7±6.9; weight (kg): 66.1±10.1; total caloric intake per day (kcal): 2054.1±320.9; and dietary fat intake (% of total kcal): 30.5±0.5. Women with high total fiber intakes had significantly less insulin resistance than their counterparts (F=4.58, p=0.0332), and women with high soluble fiber intakes had significantly lower levels of insulin resistance than other women (F=7.97, p=0.0051). Participants with high insoluble fiber intakes did not differ from their counterparts (F=0.7, p=0.6875). Adjusting for either PA or BF% weakened the relationships significantly. Controlling for BF% nullified the total fiber-HOMA-IR link (F=1.96, p=0.1631), and attenuated the association between soluble fiber and HOMA-IR by 32% (F = 6.86, p=0.0094). To facilitate interpretation of the results, odds ratios were calculated to determine the association between fiber intake and HOMA-IR with both variables treated as categorical. To create dichotomous variables, fiber intake and HOMA-IR were each divided into two categories using the median (Low and High). In women who had high soluble fiber intake (upper 50%), the odds of having an elevated HOMA-IR level was 0.58 (95% CI: 0.36-0.94) times that of women with low soluble fiber intake (lower 50%). And after controlling for all of the potential confounding factors simultaneously, the odds ratio was 0.52 (95% CI: 0.29-0.93). Conclusion: High fiber intake, particularly soluble fiber, is strongly related to lower levels of insulin resistance in women. Part of this association is a function of differences in PA and BF%.