Background: Colorectal cancer incidence has shown substantial temporal variation across populations, reflecting changes in risk factor exposure, diagnostic practices, and prevention strategies. Age–period–cohort (APC) analysis enables separation of age-related effects from calendar-period and birth-cohort influences.
Objectives: This study aimed to examine sex-specific APC patterns of colorectal cancer incidence in ?zmir, Türkiye, from 1998 to 2017 among individuals aged 45–74 years.
Methods: Data including only microscopically confirmed cases were obtained from the Global Cancer Observatory detailed database. Small numbers of cases with unknown age were redistributed proportionally. APC models were fitted separately for females and males using five-year age groups (45–74 years), four calendar periods, and corresponding birth cohorts. Longitudinal and cross-sectional age curves, age, period, and cohort deviations, rate ratios (RRs with references for birth cohort 1948, for period 2005.5), local drifts, and net drift (% annual change) were estimated with 95% confidence intervals (CIs). Statistical significance was assessed using likelihood ratio tests. Male and female results were visualized in the same figures using R version 4.5.0.
Results: Incidence increased steeply with age in both sexes, with consistently higher rates in males. In males, longitudinal age-specific incidence increased from 14.9 per 100,000 (12.6–17.6) to 275.7 per 100,000 (247.0–307.7). Among females, longitudinal age-specific incidence increased from 11.2 per 100,000 (95% CI: 9.5–13.3) at age 47.5 to 132.0 per 100,000 (116.7–149.3) at age 72.5. Cross-sectional age curves showed similar gradients, with male rates exceeding female rates at all ages (e.g., age 67.5: 127.2 vs 65.8 per 100,000). A significant upward temporal trend was observed in both sexes. The net drift was 3.24% per year in males (95% CI: 2.66–3.83; p<0.001) and 2.75% per year in females (95% CI: 2.15–3.36; p<0.001). Male local drifts were consistently higher, between about 2.1% and 3.7% per year, while female local drifts showed moderate age-specific variation, ranging approximately from 1.6% to 3.3% per year, while. Period effects were similar across both sexes. In males, period RRs increased monotonically from 0.74 (0.66–0.81) in 1998–2002 to 1.21 (1.12–1.30) in 2013–2017 (p<0.001). In females, period RRs increased from 0.81 (95% CI: 0.73–0.90) in 1998–2002 to 1.20 (1.10–1.31) in 2013–2017, with statistically significant overall period deviations (p=0.003). Cohort effects showed clear sex-specific differences. In males, cohort deviations were not statistically significant (p=0.74) and cohort-related gradients were more pronounced, with cohort rate ratios increasing steadily from 0.47 to 1.60 across successive birth cohorts (p<0.001). In females, cohort deviations were also not statistically significant overall (p=0.29), although cohort rate ratios showed a gradual increase across birth cohorts, from 0.66 to 1.71 (p<0.001).
Conclusions: Colorectal cancer incidence increased significantly in ?zmir between 1998 and 2017, with higher rates and faster growth in males. APC analyses reveal sex-specific dynamics, characterized by pronounced period effects in both sexes. These findings underscore the importance of sex-sensitive prevention strategies and timely implementation of colorectal cancer screening.

Age–period–cohort pattern of colorectal cancer in ?zmir, 1998–2017