Table 1.
Univariate dynamic structural equation models of (hypo)mania, depression and insomnia conditional on baseline covariates
| Model | Covariate | Individual mean μj (95% CI) | Inertia ϕjj (95% CI) | Innovation log(πj) (95% CI) |
|---|---|---|---|---|
| (Hypo)mania (ASRM) | BD-II bipolar subtypea | 0.34 (0.09 to 0.59) p = 0.004 |
0.04 (0.00 to 0.08) p = 0.044 |
0.16 (−0.06 to 0.38) p = 0.070 |
| Female genderb | −0.44 (−0.69 to −0.18) p < 0.001 |
−0.03 (−0.07 to 0.02) p = 0.131 |
−0.07 (−0.30 to 0.16) p = 0.280 |
|
| Older age (⩾55 yrs)c | −0.24 (−0.49 to 0.01) p = 0.028 |
−0.01 (−0.05 to 0.04) p = 0.370 |
−0.40 (−0.61 to −0.18) p < 0.001 |
|
| History of rapid cyclingd | 0.35 (0.05 to 0.64) p = 0.009 |
−0.02 (−0.07 to 0.03) p = 0.193 |
0.66 (0.41 to 0.91) p < 0.001 |
|
| Depression (QIDS) | BD-II bipolar subtypea | 1.72 (0.96 to 2.49) p < 0.001 |
0.04 (0.00 to 0.08) p = 0.015 |
0.16 (−0.01 to 0.34) p = 0.033 |
| Female genderb | 0.49 (−0.31 to 1.31) p = 0.118 |
0.02 (−0.02 to 0.06) p = 0.139 |
0.22 (0.03 to 0.40) p = 0.011 |
|
| Older age (⩾55 yrs)c | −1.86 (−2.62 to −1.10) p < 0.001 |
−0.02 (−0.06 to 0.02) p = 0.181 |
−0.49 (−0.66 to −0.32) p < 0.001 |
|
| History of rapid cyclingd | 1.94 (1.07 to 2.83) p < 0.001 |
−0.01 (−0.06 to 0.04) p = 0.326 |
0.55 (0.35 to 0.74) p < 0.001 |
|
| Insomnia | BD-II bipolar subtypea | 0.25 (−0.07 to 0.56) p = 0.063 |
0.02 (−0.02 to 0.06) p = 0.142 |
−0.02 (−0.18 to 0.14) p = 0.390 |
| Female genderb | 0.03 (−0.30 to 0.36) p = 0.422 |
0.00 (−0.04 to 0.04) p = 0.472 |
0.23 (0.07 to 0.39) p = 0.003 |
|
| Older age (⩾55 yrs)c | 0.21 (−0.11 to 0.52) p = 0.099 |
−0.02 (−0.06 to 0.01) p = 0.089 |
−0.35 (−0.50 to −0.19) p < 0.001 |
|
| History of rapid cyclingd | 0.72 (0.35 to 1.07) p < 0.001 |
−0.03 (−0.07 to 0.02) p = 0.098 |
0.38 (0.20 to 0.56) p < 0.001 |
ASRM, Altman Self-Rating Mania Scale; QIDS, Quick Inventory of Depressive Symptomatology; BD-I, Bipolar disorder type 1; BD-II, Bipolar disorder type 2; p values quoted are one-tailed as is the convention for MCMC models in Mplus.
Estimates (95% Credible Intervals) in the table indicate the effect of the covariates on individual means, inertia and innovation for each of the three models.
a
Reference: BD-I.
b
Reference: Male.
c
Reference: Age <55 years.
d
Reference: No history of rapid cycling.