Our study population was racially mixed (76% white), most had attended or completed college, and more than half were age 50 years or older at the time of interview (). Those reporting 6 or 7 family strengths were significantly more likely than those reporting fewer family strengths to be age 65 years or older at interview and to be unemployed or retired. Although more than half of those with a high number of childhood family strengths (6 or 7) reported having 1 or more ACE, they were significantly less likely than the group with fewer family strengths (0–5) to report a history of ACE (57% vs 85%; p < 0.0005).
Distribution of demographic and behavioral characteristics by childhood family strengths
Exposure to each of the seven categories of childhood family strengths was associated with a significant 30% to 40% decreased risk of adolescent pregnancy (data not shown). Compared with women who experienced family strengths never, rarely, or sometimes (“no” in ), those reporting such experiences often or very often (“yes” in ) had reductions in teen pregnancy for each family strength: 37% reduction for protection (35.3% vs 22.0%; RR, 0.63; 95% CI, 0.56–0.71); 42%, support (33.5% vs 19.6%; RR, 0.58; 95% CI, 0.53–0.65); 34%, closeness (30.7% vs 20.1%; RR, 0.66; 95% CI, 0.59–0.73); 37%, loyalty (33.6% vs 21.0%; RR, 0.63; 95% CI, 0.56–0.70); 29%, feeling important (29.0% vs 20.8%; RR, 0.71; 95% CI, 0.63–0.77); 34%, feeling loved (31.9% vs 21.0%; RR, 0.66; 95% CI, 0.59–0.74); and 33%, responsiveness to health care needs (33.4% vs 22.2%; RR, 0.67; 95% CI, 0.57–0.78). Furthermore, a significant trend effect on adolescent pregnancy was observed as the frequency of each childhood family strength increased, from “never/rarely” to “sometimes” to “often” to “very often” ().
Risk of adolescent pregnancy according to characterization of childhood family strengths.
As the number of childhood family strengths increased, the risk of adolescent pregnancy decreased significantly (). Adjusted odds ratios (OR) for adolescent pregnancy were 1.0, 0.86, 0.74, 0.59, 0.55, 0.48, and 0.46, respectively, among those with 0 to 1, 2, 3, 4, 5, 6, and 7 categories of family strengths. The absolute percentage of adolescent pregnancies for women with 7 family strengths (19%) was about half that for women with 0 or 1 family strengths (39%). We found that the magnitude of protective effect of childhood family strengths on adolescent pregnancy was significantly altered by the cofactor of ACE, which functioned as an effect modifier. Among those reporting one or more ACE, there was a highly significant protective (p < 0.000001) trend effect of childhood family strengths against adolescent pregnancy. Adolescent pregnancy rates among those women with ACE decreased from 42% to 33%, 26%, and 24%, respectively, for those reporting 1 to 2, 2 to 3, 4 to 5, and 6 to 7 categories of childhood family strengths (). After adjustment, we observed a 46% reduction in adolescent pregnancy rates (adjusted OR = 0.54) among with both high family strengths (6 or 7 categories) and coexisting ACE, compared with women with low childhood family strengths (0 or 1 category) and coexisting ACE.
Association between numbers of childhood family strengths and risk of adolescent pregnancy
Numbers of childhood family strengths and adolescent pregnancy until age 18 among women with adverse childhood experiences and those without adverse childhood experiences
Women without ACE, regardless of their family strengths, were at lower risk of adolescent pregnancy than the reference group (ACE, 0 or 1 family strengths; ). No significant difference was seen in the risk of adolescent pregnancy among those with 6 or 7 family strengths (14%) and those with 0 to 5 (combined rate of 17.3%).
We also examined whether childhood family strengths were associated with delays in initiation of sexual activity in analyses that simultaneously considered ACE (). Among women with ACE, significant protective trends (p < 0.005) against initiation of sexual activity either before age 15 years or at ages 15 to 19 years (compared with initiation of sexual activity at ages 20 years and older) were seen with greater numbers of childhood family strengths; however, the lowest risk of early initiation of sexual activity was consistently observed among women without ACE and was consistent across the 0 to 7 range of childhood family strengths. Of those with ACE and 0 or 1 family strength, 67.6% initiated sex at age 15 to 19 years; among those with ACE and 6 or 7 family strengths, 57.5% initiated in this age range; among those without ACE and 0 to 7 family strengths, 40.4% of women initiated sexual activity at age 15 to 19 years.
Childhood family strengths and adverse childhood experience status as predictor of age of initiation of sexual activity
Finally, we analyzed long-term psychosocial consequences, which were measured at the time of interview, when the interviewees were at a mean age of 56 years. We found significant positive trends for childhood family strengths and each of the psychosocial outcomes considered, including serious or disabling problems with jobs, family, or finances, high stress, or uncontrollable anger (). After adjusting for age, race, education, adolescent pregnancy, and history of coexisting childhood abuse or family dysfunction, we found that a high number of family strengths (6 or 7) led to a significant protective effect against job, family, and financial problems, as well as uncontrollable anger. These findings did not vary by whether ACE were reported.
Numbers of childhood family strengths and long-term psychosocial problems
In analyses stratified by age cohort (19–34, 35–49, 50–64, and ≥65 years), we found a significant trend for each age cohort when we compared the number of childhood family strengths with the rate of adolescent pregnancy (data not shown). For 0 or 1, to 2 or 3, 4 or 5, and 6 or 7 family strengths, we found that the risk of adolescent pregnancy was as follows (p for trend for each group < 0.0001): 19–34 years: 42.9%, 25.6%, 28.1%, and 16.1%; 35–49 years: 38.2%, 36.6%, 26.4%, and 21.0%; 50–64 years: 39.7%, 35.2%, 26.3%, and 24.0%; and ≥65 years: 36.6%, 22.7%, 18.0%, and 16.4%. Also, for each age cohort, the odds of psychosocial consequences decreased as numbers of family strengths increased (data not shown).