Cause-Specific Mortality in a Population With Diabetes
South Tees Diabetes Mortality Study
- Nick A. Roper, MRCP12,
- Rudy W. Bilous, MD12,
- William F. Kelly, MD1,
- Nigel C. Unwin, MRCP3 and
- Vincent M. Connolly, MD1
- 1Diabetes Care Centre, Middlesbrough General Hospital, Middlesbrough, U.K.
- 2Department of Diabetes, University of Newcastle, Newcastle upon Tyne, U.K.
- 3Department of Epidemiology and Public Health, University of Newcastle, Newcastle upon Tyne, U.K.
OBJECTIVE—To describe the mortality of a population with diabetes compared with the local nondiabetic population, using age-, sex-, and cause-specific death rates and relative and absolute differences in death rates.
RESEARCH DESIGN AND METHODS—A population-based cohort of 4,842 people with diabetes living within South Tees, U.K., was identified and followed from 1 January 1994 to 31 December 1999. Causes of death were obtained from death certificates, and mortality rates were compared with the nondiabetic population of the same area for the same time period.
RESULTS—There were 1,205 deaths (24.9%) in the study population during the 6 years of study. For type 2 diabetes, mortality from cardiovascular causes was significantly increased in both sexes and at all ages. Relative death rates for the age band 40–59 years were 5.47 (95% CI 4.18–7.15) for men and 5.60 (3.44–9.14) for women. The relative death rates declined with age for both sexes, but absolute excess mortality increased with age. There were no consistent differences in noncardiovascular death rates, other than for renal disease. Similar outcomes were found for type 1 diabetes, although these results were limited by a much smaller population size. People with diabetes and renal impairment had significantly higher mortality than people with diabetes alone, with a rate ratio of 7.27 for people with type 2 diabetes aged 40–59 years.
CONCLUSIONS—In an area of the U.K. with high cardiovascular death rates, people with diabetes had significantly higher cardiovascular death rates than people without diabetes. Interventions targeted at cardiovascular risk factors should be used to try and reduce this excess premature mortality, which is especially high in those with renal impairment.
Address correspondence and reprint requests to Dr. Vincent M Connolly: Diabetes Care Centre, Middlesbrough General Hospital, Middlebrough TS5 5AZ, U.K. E-mail:.
Received for publication 20 September 2001 and accepted without revision 29 September 2001.
A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.