Objectives: This study investigated, on 53 Australians consuming a typical Western diet, the relationship between dietary intake, faecal excretion of carbohydrate and changes in faecal markers believed to be relevant to colon cancer risk, for example faecal output, transit time and concentrations of phenols, ammonia and butyrate. Design: Fifty-three subjects consuming their usual diet were asked to record and weigh all food consumed for a seven day period, and to collect faeces for three days during this period. Setting: Geelong, Victoria, Australia. Subjects: All volunteers were either staff and students of the university, or associates of the authors. Interventions: None. Results: Volunteers had the following dietary intakes of carbohydrate (g/d; mean AE s.d.); starch 131 AE 41 (including resistant starch (RS), 5 AE 2), sugar 108 AE 37 and non-starch polysaccharides (NSP) 14 AE 7. Daily faecal output was 127 AE 70 g and transit time 47 AE 19 h. Analysis of faecal samples found 0.8 AE 1.2 g RS and 5.6 AE 3.6 g NSP were excreted daily. Dietary starch intake was the only dietary carbohydrate to show a signi®cant relationship with the concentration (mmol/L) of butyrate excreted in faeces (r 0.34, P`0.05). Dietary intake of RS was associated with higher concentrations of faecal ammonia (r 0.34, P`0.05), but this association was reversed when RS was combined with NSP in the diet (r 0.07, NS). In contrast to dietary intake, the faecal excretion of RS was negatively related to faecal ammonia concentration (r 7 0.40, P`0.01) and positively related to faecal output (r 0.64, P`0.01). Individuals who consumed more NSP in their diet (19 AE 7 g/d) excreted more than 150 g faeces per day and had higher quantities of faecal-RS and -NSP; faster transit times; higher concentrations of short chain fatty acids and lower concentrations of potentially harmful ammonia and phenols. Conclusions: The combination of RS and NSP in the colon may be required to achieve an optimal luminal environment conducive to`colonic health'. The results also support the suggestion that faecal output (`or b 150 g/d) may provide a useful index of colon cancer risk. High faecal outputs are achieved through higher intakes of NSP (the major component of dietary ®bre). Sponsorship: This work was supported by a scholarship to A. Birkett from the Australian Postgraduate Research Council.

Pacific Island populations exhibit a remarkably high prevalence rate of obesity and type 2 diabetes and represent a unique population for genetic studies of obesity. In the present study we have revealed that a specific combination of alleles in OB and OB-R, two candidate genes for obesity, may confer an increased risk for the development of insulin resistance in Nauruan males.