PAD Completion Correlates: Cross-Site Analyses of Characteristics Associated With Having PADs
| Bivariate Associations | Domain Models | Final Model | ||||
|---|---|---|---|---|---|---|
| Common OR | 95% CI | Adjusted OR | 95% CI | Adjusted OR | 95% CI | |
| Model 1: Demographics and social environment | ||||||
| Age above median (>44 y) | 1.09 | (0.66–1.81) | ||||
| Female | 0.89 | (0.54–1.48) | ||||
| Nonwhite | 0.75 | (0.44–1.27) | ||||
| Cohabitation | 0.99 | (0.53–1.88) | ||||
| Some college education | 1.93 | (0.87–4.30) | ||||
| Marital status/involuntary commitment interaction | ||||||
| Single and no involuntary commitment [comparison] | [1.00] | |||||
| Married and no involuntary commitment | 0.23 | (0.03–1.70) | 0.21 | (0.03–1.59) | ||
| Single with involuntary commitment | 1.06 | (0.61–1.85) | 0.97 | (0.55–1.69) | ||
| Married with involuntary commitment | 2.43 | (1.16–5.11)* | 2.21 | (1.04–4.67)* | ||
| Housing categories | ||||||
| Independent living | [1.00] | |||||
| Homeless | 2.47 | (0.79–7.70) | ||||
| Living with family | 1.10 | (0.50–2.42) | ||||
| Staffed residence | 1.05 | (0.58–1.90) | ||||
| Income above median (>$700) | 1.05 | (0.61–1.80) | ||||
| Social support | 1.04 | (0.56–1.95) | ||||
| Social support for treatment | 1.92 | (1.12–3.30)* | 1.88 | (1.08–3.25)* | ||
| Model significance | LR = 15.13, df = 4** | |||||
| Model 2: Clinical variable model | ||||||
| Psychotic disorder | 1.20 | (0.73–1.96) | ||||
| Substance abuse | 1.45 | (0.81–2.60) | ||||
| BPRS score above median (>30) | 1.48 | (0.89–2.47) | ||||
| GAF score above median (>47) | 0.63 | (0.36–1.10) | ||||
| Insight score above median (>18) | 1.93 | (1.12–3.34)* | 1.88 | (1.08–3.28)* | 2.06 | (1.17–3.66)* |
| Self harm | 1.29 | (0.77–2.16) | ||||
| Model significance | LR = 18.08, df = 2*** | |||||
| Model 3: Treatment and systems utilization | ||||||
| Past hospitalizations above median (>3) | 1.67 | (0.99–2.83)† | ||||
| Time in treatment above median (>20 y) | 1.25 | (0.75–2.08) | ||||
| Picked up by the police in past 6 mo (treatment) | 2.94 | (1.46–5.95)** | 3.05 | (1.50–6.21)** | 2.25 | (1.08–4.71)* |
| Picked up by the police in past 6 mo (arrest) | 2.29 | (0.92–5.69)† | ||||
| Outpatients visits above median (>2 monthly) | 0.84 | (0.48–1.49) | ||||
| Involuntary commitment | 1.14 | (0.65–1.99) | ||||
| Treatment satisfaction above median (>52) | 1.21 | (0.74–2.00) | ||||
| Drug Attitudes Inventory above median (>8) | 1.24 | (0.75–2.04) | ||||
| Medication compliance above median (>3) | 0.89 | (0.53–1.50) | ||||
| Appointment compliance above median (>3) | 1.10 | (0.62–1.95) | ||||
| Model significance | LR = 7.86, df = 1** | |||||
| Model 4: Leverage model | ||||||
| Outpatient commitment | 1.59 | (0.85–2.97) | ||||
| Criminal justice leverage | 1.48 | (0.85–2.58) | ||||
| Representative payee | 3.61 | (2.03–6.42)*** | 3.07 | (1.70–5.56)*** | 3.32 | (1.81–6.09)*** |
| Housing | 1.81 | (1.09–3.00)* | ||||
| Pressure to take medications above median (>1) | 1.54 | (0.93–2.53)† | ||||
| Pressure to keep appointments above median (>1) | 2.26 | (1.37–3.74)** | 1.97 | (1.16–3.32)* | 1.79 | (1.05–3.05)* |
| Perceived coercion above median (>7.23) | 1.58 | (0.96–2.60)† | ||||
| Autonomy above median (>24) | 0.75 | (0.43–1.28) | ||||
| Model significance | LR = 23.26, df = 2*** | LR = 34.38, df = 4**** | ||||
A common odds ratio with cluster-corrected confidence interval is given only if all five sites’ odds ratios were determined by Zelen’s test (p < 0.05) to represent a sampling distribution from a common population. Results were produced by logistic regression.
† p < 0.10 (trend);
* p < .05;
** p < .01;
*** p < .001;
**** p < .0001.