Logo of nihpaLink to Publisher's site

Table 4

Simulated effects of childhood adversities on proportional increase in mean duration between time of interview and time of most recent episode in sub-samples defined by the cross-classification of disorder type and respondent age at interview

OverallAges 18-29Ages 30-44Ages 45-59Ages 60+
Meanr*MeanuDiff %MeanrMeanuDiff %MeanrMeanuDiff %MeanrMeanuDiff %MeanrMeanuDiff %
Mood3.83.74.91.41.36.22.62.312.56.05.84.011.811.35.1
Anxiety10.910.80.62.02.0-1.59.39.11.515.615.7-0.632.232.03.7
Substance7.47.32.11.81.82.26.96.63.311.311.02.715.213.710.7
Disruptive Behaviora9.29.3-1.24.74.8-2.511.512.3-6.5------
Any8.48.31.63.02.91.48.17.92.711.411.22.020.019.71.3
*Mean number of years in the restricted model, mean number of years in the unrestricted model, and the % difference. The restricted model is one in which the ORs associated with CAs were restricted to be 1.0, simulating a situation in which CAs were completely unrelated to duration between time of interview and time of more recent episode. The unrestricted model is one in which the empirically observed associations between CAs and the outcome were retained. If CAs are associated with more recent episodes, we would expect the estimated mean duration in the restricted models to be larger than in the unrestricted model; that is, for the amount of time since most recent episode to be longer in the absence of CAs. This is, in fact, the general pattern in the table, with differences between Meanr and Meanu being mostly positive.
aDisruptive behavior disorders are restricted to those <= 44 years of age at interview