Gender differences in the trend of colorectal cancer incidence in Singapore, 1968-2002

Inge Driesprong - de Kok, CS Wong, KS Chia, X Sim, CS Tan, LA Kiemeney, HM Verkooijen

Research output: Contribution to journalArticleAcademicpeer-review

40 Citations (Scopus)
20 Downloads (Pure)


Background and aims Over the past decades, incidence trends of colorectal cancer are sharply increased in Singapore. In this population-based study we describe changes in colorectal cancer incidence in Singapore and explore the reasons behind these changes through age-period cohort (APC) modeling. Methods We included all 22,609 colorectal cancer cases reported to the Singapore Cancer Registry between 1968 and 2002. Poisson regression, using age-period (AP) and age-cohort (AC) models was used to determine the effects of age at diagnosis, calendar period, and birth cohort. Results Male colorectal cancer rates between 1968 and 2002 from 20 to 40 per 100,000 person years. The increase was sharpest among older men, for whom there was a significant AC effect. Female colorectal cancer rates increased until 1992 (from 16 to 29 per 100,000 person years) and stabilized afterward. For women under 65 years, we observed a significant AP effect, corresponding to a sudden rise in colorectal cancer incidence around 1978. Conclusions This study demonstrates important gender differences in colorectal cancer incidence in Singapore, with increasing rates among men, and stabilized rates in women. The increase in men is mainly attributable to an incidence increase in the oldest age groups, probably due to increased exposure to dietary and lifestyle risk factors earlier in life. The stabilization in female colorectal cancer risk could be due to lower exposure to lifestyle risk factors and prophylactic removal of precancerous lesions.
Original languageUndefined/Unknown
Pages (from-to)461-467
Number of pages7
JournalInternational Journal of Colorectal Disease
Issue number5
Publication statusPublished - 2008

Research programs

  • EMC NIHES-02-65-01

Cite this