. Author manuscript; available in PMC: 2013 Mar 20.

Published in final edited form as: J Affect Disord. 2012 Sep 7;146(1):53–57. doi: 10.1016/j.jad.2012.08.037

Table 2.

Odds ratios of Major Depressive Disorder in adult offspring of depressed (high risk group) and nondepressed (low risk group) parents between 10- and 20-year follow-ups by frequency of religious Services Attendance and Social Adjustment (SAS—SR) at year 10 run individually (univariate models) and simultaneously (multivariate models).

Total sample and risk group	Univariate models^a				Multivariate models^b
	Odds ratio	95% CI	Wald χ²	p	Odds ratio	95% CI	Wald χ²	p
Total sample (N = 173)
Frequent attendance at religious/spiritual services	0.454	0.184-1.118	2.950	0.086	0.399	0.157-1.011	3.749	0.053
High social adjustment (SAS—SR)
Overall	0.268	0.114-0.629	9.158	0.002	0.249	0.105-0.591	9.928	0.002
Work role	0.607	0.259-1.423	1.320	0.251	0.653	0.275-1.546	0.940	0.332
Social and leisure	0.467	0.214-1.021	3.639	0.056	0.450	0.203-0.996	3.884	0.049
Extended family	0.364	0.156-0.852	5.426	0.020	0.348	0.148-0.823	5.787	0.016
Primary relationship	0.306	0.106-0.883	4.792	0.029	0.287	0.097-0.848	5.103	0.024
Parental	0.340	0.078-1.483	2.060	0.151	0.445	0.089-2.230	0.970	0.325
Family unit	0.188	0.080-0.441	15.818	0.000	0.192	0.081-0.453	14.178	0.000
High risk group (N = 118)
Frequent attendance at religious/spiritual services	0.553	0.190-1.606	1.186	0.276	0.435	0.141-1.345	2.087	0.149
High social adjustment overall	0.215	0.076-0.604	8.486	0.004	0.193	0.067-0.557	9.271	0.002
Low risk group (N=55)
Frequent attendance at religious/spiritual services	0.180	0.026-1.240	3.032	0.082	0.169	0.024-1.185	3.202	0.074
High social adjustment overall	0.278	0.046-1.671	1.958	0.162	0.247	0.037-1.630	2.110	0.146

For each of the 12 univariate models, the outcome measure is major depressive disorder at year 20 (yes/no) and the primary predictors are the frequency of religious services attendance variable or the dichotomous social adjustment variables at year 10 run individually. Sex, age, history of depression are controlled in all models. Risk group is controlled for in the total sample analyses.

For each of the nine multivariate models, the outcome measure is major depressive disorder at year 20 (yes/no) and the primary predictors are the frequency of religious services attendance variable at year 10 paired with the social adjustment variables at year 10 and run simultaneously. Sex, age, history of depression, and the other dichotomous variable are controlled in both models. Risk group is controlled for in the total sample analyses.