Introduction The leading causes of mortality in Sri Lanka are due to chronic diseases. We describe the mortality patterns in a 35–64-year-old urban cohort resident in Ragama, Sri Lanka and followed over 3 years.

Methods A follow-up study was conducted among 2986 35–64 year olds randomly selected from the Ragama Medical Officer of Health area, Sri Lanka. A baseline survey was conducted from January to September 2007 and a follow-up survey was conducted from March to November 2010. Mortality data were obtained from next of kin and cause of death was verified from death certificates.

Results There were 49 deaths during 9186.46 person years of observations. Of the 49 deaths, 11 were due to myocardial infarctions, 5 were due to strokes, 5 were due to other ischaemic heart disease and the rest included 6 due to cancer and 2 due to train accidents. The increase in mortality in men occurs after 45 years and in females it is observed later on. Mortality among men was more than twice as much as females (RR 7.96 vs 3.17 per 1000 person years). All cause mortality was significantly higher in diabetics. Mortality was not associated with hypertension, dyslipidaemia, smoking, central obesity, obesity or physical activity.

Conclusions Diabetes Mellitus was significantly associated with all cause mortality. Other associations may have not been significant due to the small number of deaths.