Risk of a daughter having a first induced abortion (IA) as a teenager, in relation to total number of IAs that her mother had between 4 years before and 12 years after daughter’s birth date
| No. of maternal IAs between −4 yr and +12 yr of daughter’s birth date | No. (%) of daughters with IA | Incidence rate at time of first IA, per 1000 person-years (95% CI) | HR (95% CI) | |
|---|---|---|---|---|
| Crude* | Adjusted*† | |||
| 0 (n = 358 105) | 10 108 (2.8) | 4.2 (4.1–4.3) | 1.00 (ref) | 1.00 (ref) |
| 1 (n = 46 441) | 2622 (5.6) | 8.6 (8.2–8.9) | 2.10 (2.01–2.19) | 1.77 (1.69–1.85) |
| 2 (n = 14 812) | 1063 (7.2) | 11.0 (10.4–11.7) | 2.76 (2.59–2.94) | 2.04 (1.91–2.18) |
| 3 (n = 6134) | 559 (9.1) | 14.1 (13.0–15.3) | 3.58 (3.28–3.91) | 2.39 (2.19–2.62) |
| ≥ 4 (n = 6131) | 636 (10.4) | 16.4 (15.1–17.7) | 4.28 (3.94–4.64) | 2.54 (2.33–2.77) |
Note: CI = confidence interval, HR = hazard ratio.
↵* Results based on Cox proportional hazard models. Robust sandwich variance estimates were used to account for more than 1 daughter clustered within the same mother.
↵† Adjusted for mother’s age when giving birth to her daughter, mother’s world region of origin, daughter’s mental health status in the 2-year period before turning age 12 years, whether daughter had major physical health problem in the 2-year period before turning age 12 years, neighbourhood-level teen IA rate, mother–daughter cohabitation when daughter was 12 years of age, daughter’s residence (rural v. urban) at time of birth and at age 12 years, and daughter’s neighbourhood income quintile at time of birth and at age 12 years.