Table 1 Summary of age-period-cohort analysis for pleural mesothelioma incidence in four selected gender/age groups
APC parameters | Males, age 0–74 | Males, age 75+ | Females, age 0–74 | Females, age 75+ |
|---|---|---|---|---|
Period effect: peak PM | ||||
Incidence years | 1978–1982 | 2008–2012 | 1973–1977 | 2008–2012 |
Net drift (95% CI) | − 1.8% (− 2.3%, − 1 3%) | +4.0% (3.2%, 4.7%) | − 1.5% (− 3.3%, 0.3%) | +3.7% (1.7%, 5.7%) |
p value | < 0.0001 | < 0.0001 | 0.11 | 0.0002 |
Period RRs different from 1988 to 1992? | Yes | Yes | No | Yes |
p value | < 0.0001 | < 0.0001 | 0.67 | 0.02 |
Period deviation is non-linear? | Yes | Yes | No | No |
p value | 0.03 | 0.05 | 0.89 | 0.81 |
Cohort effect: peak PM | ||||
Birth cohort years | 1928–1932 | 1928–1932 | 1931–1935 | 1921–1925 |
Local drifts = net drift for all age groups? | No | No | No | Yes |
p value | < 0.0001 | 0.0002 | 0.02 | 0.58 |
Cohort RRs different from referent cohort? | Yes | Yes | No | Yes |
p value | < 0.0001 | < 0.0001 | 0.16 | 0.01 |
Cohort deviation is nonlinear? | Yes | Yes | No | No |
p value | <0.0001 | <0.0001 | 0.16 | 0.91 |
Cross-sectional age trend (95% CI) | 12.5% (11.8, 13.2%) | 2.2% (0.9, 3.5%) | 7.2% (5.9, 8.4%) | − 0.1% (− 4.1, 3.9%) |
Longitudinal age trend (95% CI) | 10.7% (9.9, 11.4%) | − 1.7% (− 3.1, − 0.4%) | 8.7% (6.7, 10.6%) | − 3.7% (− 7.8, 0.4%) |
Longitudinal vs. cross-sectional RR trend | Negative | Positive | Negative | Positive/flat |
Age deviation is non-linear? | No | Yes | No | No |
p value | 0.41 | 0.0002 | 0.46 | 0.07 |