Table 4.
Childhood Psychotic Symptoms and Risk of Clinical and Functional Problems at Age 18 Years Additionally Adjusted for Family History of Mental Illness
Young-Adult Outcomes | RR (95% CI) | P Value |
---|---|---|
Mental health | ||
Psychotic experiences | 1.39 (1.09, 1.78) | .009 |
Psychotic symptoms | 3.98 (1.69, 9.36) | .002 |
Depression | 1.69 (1.24, 2.31) | .001 |
Anxiety | 2.29 (1.39, 3.79) | .001 |
Suicide attempt | 2.82 (1.38, 5.73) | .004 |
Self-harm | 1.56 (1.04, 2.36) | .033 |
Substance use | ||
Tobacco dependence | 2.09 (1.30, 3.35) | .002 |
Cumulative score for mental health outcomes | 1.59 (1.27, 2.00) | <.001 |
Psychosocial | ||
Low life satisfaction | 1.54 (1.11, 2.15) | .010 |
Loneliness | 1.43 (1.11, 1.86) | .007 |
Social isolation | 1.36 (0.98, 1.88) | .066 |
Parenthood | 4.87 (1.38, 17.20) | .014 |
Physical health | ||
Risky sexual behaviors | 1.30 (0.97, 1.75) | .080 |
Overweight | 1.36 (1.02, 1.83) | .039 |
Sleep problems | 1.34 (1.10, 1.63) | .004 |
Cumulative score for functional outcomes | 1.33 (1.11, 1.59) | .002 |
Note: This analysis focuses only on those outcomes that were statistically significant at P < .05 following adjustment for covariates in tables 2 and 3. Here we adjust for family psychiatric history and maternal psychotic symptoms as well as the nonindependence of twin observations, gender, age-5 IQ, family socioeconomic status, and other psychopathology at age 12. The N within each model is restricted to participants with non-missing data on all variables included in the multivariate models. The comparison group is those who did not have psychotic symptoms at age 12. The cumulative score for mental health outcomes was derived by summing all of the dichotomized scores for the different mental health and substance use problems. The cumulative score for functional outcomes was derived by summing all of the dichotomized scores for the different functional outcomes. Statistically significant results (P < .05) are presented in bold text. CI, confidence interval; RR, risk ratio derived using Poisson regression.