TABLE 3.

Maternal experience of childhood abuse and offspring’s persistent high depressive symptoms across adolescence and young adulthood, Nurses’ Health Study II and Growing Up Today Study, N mothers = 6,692 N offspring = 8,108 N observations (offspring) = 34,302^a

	N offspring	Risk ratio (95% CI)
		Model 1: adjusted for demographic variables	Model 2: Model 1, further adjusted for maternal mental health indicators	Model 3: Model 2, further adjusted for adolescent socioeconomic status and family structure	Model 4: Model 3, further adjusted for offspring’s experience of abuse
Maternal experience of childhood abuse
0: no abuse	2,102	1.00 (Reference)	1.00 (Reference)	1.00 (Reference)	1.00 (Reference)
1	1,850	1.42 (0.90, 2.23)	1.33 (0.85, 2.10)	1.34 (0.85, 2.12)	1.36 (0.85, 2.17)
2	1,591	1.80 (1.16, 2.78)c	1.59 (1.03, 2.47)b	1.60 (1.03, 2.48)b	1.41 (0.90, 2.23)
3	1,553	2.43 (1.60, 3.69)d	2.07 (1.36, 3.15)d	2.04 (1.33, 3.11)d	1.85(1.18, 2.90)c
4	635	2.04 (1.22, 3.44)c	1.72 (1.01, 2.91)b	1.73 (1.02, 2.92)b	1.28 (0.71, 2.29)
5: severe abuse	377	2.47 (1.37, 4.44)c	1.92 (1.05, 3.54)b	1.85 (0.99, 3.43)b	1.44 (0.76, 2.70)

^aModels 1, 2, and 3 include the same participants; the stated N mothers, N offspring, and N observations applies. Model 4 is restricted to offspring with childhood abuse data; results for Model 1 in this subsample were similar to those in the whole sample (RR range = 1.51, 2.55, RR severe abuse = 2.55, 95% CI = 1.37, 4.76). All models adjusted for offspring’s age, sex, and race/ethnicity, and maternal childhood socioeconomic status. Risk ratios were estimated using generalized estimating equations with a log link and a Poisson distribution, using SAS PROC GENMOD. The sample for this table was restricted to offspring with three or more waves of data on depressive symptoms.

^bP < .05.

^cP < .01.

^dP < .001.