Published in:
01-09-2011 | Article
Time-varying incidence of cancer after the onset of type 2 diabetes: evidence of potential detection bias
Published in: Diabetologia | Issue 9/2011
Login to get accessAbstract
Aims/hypothesis
Despite the vast body of epidemiological literature on the risk of cancer in people with diabetes, few studies have examined the pattern of cancer risk during different time windows following diabetes onset. The objective of the study was to examine the risks of site-specific cancer in people with incident type 2 diabetes during different time windows following diabetes onset.
Methods
This was a population-based retrospective cohort study. The study period was 1 April 1994 to 31 March 2006; censoring occurred at 31 March 2006, at death or on departure from British Columbia, Canada. Using linked health databases, we identified incident cohorts with and without diabetes, who were matched by age, sex and index year. Following a minimum 2-year cancer washout period, first site-specific cancers were identified prospectively in both cohorts.
Results
Within 3 months following diabetes onset, participants with diabetes had significantly increased risks of colorectal, lung, liver, cervical, endometrial, ovarian, pancreatic and prostate cancers. After the initial 3-month period, the risks for colorectal (HR 1.15, 95% CI 1.05, 1.25), liver (HR 2.53, 95% CI 1.93, 3.31) and endometrial (HR 1.58, 95% CI 1.28, 1.94) cancers remained significantly elevated compared with those without diabetes. The diabetes cohort remained at increased risk of pancreatic cancer in later years, but followed a different pattern: HR 3.71 at 3 months–1 year, 2.94 at 1–2 years, 1.78 at 2–3 years and 1.65 at 3–10 years (p value for all <0.01). After an initial period of elevated risk, men with type 2 diabetes subsequently had a decreased risk of prostate cancer (HR 0.82, 95% CI 0.76, 0.88).
Conclusions/interpretation
People with type 2 diabetes are at increased risk of select cancers; this risk is particularly elevated at the time of diabetes onset, which is likely to be due to increased ascertainment.