Table 3

 Ever regular daily smoking among men and women by five year birth cohort: the observed prevalences and those expected on the basis of the model* and extrapolation†. Corrected for the bias attributable to declining response rate

MenObserved (%)Model (%)Extra- polation (%)Model-ExtrapolationWomenObserved (%)Model (%)Extra- polation (%)Model-Extrapolation
*The model: age profile + three dummies for the deviant cohorts + continuous cohort trend + TCA + TCA×continuous cohort (see table 2). †In the extrapolation the model was used so that the effects of the three dummies and the effect of the TCA were assumed to be zero. ‡The observed smoking prevalence was lower than the estimated prevalence because almost all members in this cohort were 19–24 year old. The observed points where number of cases was less than 200 were excluded.
Observed: age group 50–64 year old, in the models age = 57 year old
1916–2072.074.468.4+6.01916–2013.915.611.0+4.6
1921–2575.273.667.6+6.01921–2520.119.613.9+5.7
1926–3066.666.866.801926–3019.317.417.40
1931–3563.063.165.9−2.81931–3521.921.021.6−0.6
1936–4064.465.065.001936–4024.326.426.40
1941–4566.664.264.201941–4532.631.931.90
1946–5067.065.363.3+2.01946–5040.938.237.9+0.3
Observed: age group 35–49 year old, in the models age = 42 year old
1931–3563.963.866.6−2.81931–3521.323.023.6−0.6
1936–4065.065.765.701936–4026.928.728.70
1941–4565.964.864.801941–4535.734.434.40
1946–5066.166.064.0+2.01946–5041.040.940.6+0.3
1951–5563.063.163.101951–5548.347.247.20
1956–6060.759.562.2−2.71956–6047.548.553.8−5.3
1961–6556.753.961.3−7.41961–6541.840.660.3−19.7
Observed: age group 19–34 year old, in the models age = 24 year old
1946–5064.864.162.0+2.11946–5040.139.839.5+0.3
1951–5562.261.161.101951–5549.246.046.00
1956–6056.857.460.2−2.81956–6044.547.352.6−5.3
1961–6549.951.859.3−7.51961–6538.439.459.2−19.8
1966–7050.050.458.4−8.01966–7040.539.465.4−26.0
1971–7547.949.157.4−8.31971–7539.139.471.1−31.7
1976–8040.5‡47.756.5−8.81976–8033.0‡39.476.3−36.9