Lack of association between cassava consumption and tropical pancreatitis syndrome.

Previous studies suggesting an association between use of tubers of cassava and tropical pancreatitis have been weak and conflicting. To test a possible association the cassava consumption of 40 consecutive cases of tropical pancreatitis syndrome were compared with age-matched and sex-matched healthy hospital visitors. The sociodemographic characteristics of the two groups were comparable. There was no association between cassava consumption and tropical pancreatitis syndrome (odds ratio = 0.56; 95% confidence interval = 0.21-1.45). Controlling for the possible confounding effects of low socio-economic status and vegetarian diet did not alter the odds ratio. Testing for interaction also failed to show any effect modification of the association of cassava by economic status, chilli consumption or vegetarian diet. A significantly higher number of cases gave a positive family history of diabetes compared to the controls (odds ratio = 4.11; 95% confidence interval = 1.04-16.30; P = 0.04). In this case-control study which had sufficient power to detect an odds ratio > or = 3.5, there was no association between cassava consumption and tropical pancreatitis syndrome.