Table 2. Bivariate and multivariate associations (odds-ratios) between childhood adversities and the subsequent first onset of DSM-IV/CIDI disorders¹ (n = 5,692)^†.

	Bivariate‡		Multivariate (Additive)§		Multivariate (Number of CAs) ‖		Multivariate (Interactive)¶
	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)	OR	(95% CI)
I. Maladaptive family functioning
Parent mental illness	1.7*	(1.5-1.8)	1.3*	(1.2-1.4)	-	-	1.4*	(1.3-1.6)
Parent substance	1.8*	(1.6-1.9)	1.3*	(1.2-1.4)	-	-	1.4*	(1.2-1.6)
Parent criminal	1.5*	(1.4-1.7)	1.0	(1.0-1.2)	-	-	1.2*	(1.0-1.4)
Family violence	1.8*	(1.7-2.0)	1.4*	(1.2-1.5)	-	-	1.5*	(1.3-1.7)
Physical abuse	1.8*	(1.7-2.0)	1.2*	(1.1-1.4)	-	-	1.4*	(1.2-1.6)
Sexual abuse	1.8*	(1.6-2.0)	1.4*	(1.3-1.6)	-	-	1.6*	(1.4-1.9)
Neglect	1.9*	(1.7-2.1)	1.2*	(1.0-1.3)	-	-	1.4*	(1.2-1.6)
χ27			411.2*				59.0*
II. Other childhood adversities
Parent died	1.0	(0.9-1.2)	1.0	(0.9-1.1)	-	-	1.1	(0.9-1.2)
Parents divorced	1.1*	(1.0-1.3)	1.0	(0.9-1.1)	-	-	1.1	(0.9-1.2)
Other parental loss	1.5*	(1.4-1.6)	1.2*	(1.1-1.3)	-	-	1.3*	(1.1-1.5)
Serious physical illness	1.3*	(1.2-1.5)	1.3*	(1.2-1.4)	-	-	1.4*	(1.2-1.6)
Family economic adversity	1.3*	(1.2-1.4)	1.0	(0.9-1.1)	-	-	1.1	(1.0-1.3)
χ25			31.7*				21.9*
χ212			884.5*				86.9*
III. Number of childhood adversities
1	-	-	-	-	1.3*	(1.2-1.5)	-	-
2	-	-	-	-	1.8*	(1.6-2.0)	1.1	(0.9-1.3)
3	-	-	-	-	1.9*	(1.7-2.2)	0.8	(0.6-1.1)
4	-	-	-	-	2.4*	(2.1-2.7)	0.8	(0.5-1.1)
5	-	-	-	-	2.8*	(2.5-3.1)	0.6	(0.4-1.0)
6	-	-	-	-	3.4*	(2.8-4.1)	0.6	(0.3-1.1)
7+	-	-	-	-	3.2*	(2.8-3.6)	0.3*	(0.2-0.7)
χ27					822.0*		χ26 = 63.7*

^*Significant at the 0.05 level, two-tailed

^†A separate person-year file was created for each of the 20 disorders and these 20 files were then stacked.

The models were estimated in a discrete-time survival framework with person-year as the unit of analysis using this stacked dataset, thereby forcing the slopes to be constant across the 20 disorders. Each model controlled for person-year, age category, sex, 19 dummy variables for the outcome disorder category (i.e., for the 20 disorders in the stacked dataset), and controls for the prior onset of comorbid conditions that began up to age 17. The 5692 respondents had a total of 11,047 disorder onsets, ranging from a low of 101 onsets for Bipolar I disorder to a high of 1573 onsets for Major Depressive Disorder. A total of 4,700,780 non-case (i.e., not involving one of the 11,047 onsets) person-years existed across all disorders in the stacked dataset. A 10% stratified probability sub-sample of these person-years was used as controls, each with a weight of 10, in order to decrease computation time. No bias in the estimation of ORs is introduced by sampling on the outcome due to the fact that the sampling fraction cancels out in the estimation of Ors.³⁷ Estimates of PARP, though, are biased by sub-sampling. The weight of 10 (i.e., 1/10% = 10) was added to correct for this bias. Data on the prevalence of individual CAs and the distribution of number of CAs separately in person-years with and without onsets of the disorders are available on request. For person-years with an onset, these prevalence estimates range from a low of 9.0% (physical illness) to a high of 28.5% (family violence).

^‡Models were estimated with one adversity at a time in addition to the controls noted in the previous footnote.

^§The model was estimated with all 12 adversities in addition to the controls noted in the first footnote.

^‖The model was estimated with dummy predictors for number of adversities without any information about the types of adversities. The same controls used in earlier models were included as well.

^¶The model was estimated with dummy predictors for number of adversities as well as information about the types of adversities. The same controls used in earlier models were included as well.