Table 3.
Poisson regression with DUP as an outcome: univariate and multivariate analyses
| |
|
Univariate |
|
Multivariate |
|
|---|---|---|---|---|---|
| Total sample | Rate ratio (95% CI) | P | Rate ratio (95% CI) | P | |
| Age |
Per year |
1.02 (1–1.03) |
0.09 |
1.01 (1.00–1.04) |
0.14 |
| Gender |
Male |
1 |
|
1 |
|
| |
Female |
0.9 0.62–1.30) |
0.58 |
1.07 (0.77–1.48) |
0.69 |
| Single |
No |
1 |
|
1 |
|
| |
Yes |
1.39 (0.99–1.97) |
0.06 |
1.46 (0.99–2.14) |
0.06 |
| Qualifications |
None |
1 |
|
1 |
|
| |
Yes |
0.79 (0.56–1.13) |
0.20 |
0.99 (0.72–1.37) |
0.95 |
| Social class |
I,II,III |
1 |
|
1 |
|
| |
IV,V |
1.15 (0.77–1.71) |
0.51 |
1.07 (0.71–1.60) |
0.75 |
| |
Unclassified, not working |
1.16 (0.69–1.96) |
0.57 |
1.31 (0.75–2.32) |
0.34 |
| Number of close persons |
Per person |
0.9 (0.84–0.96) |
0.002 |
0.93 (0.89–0.99) |
0.03 |
| UK born |
No |
1 |
|
1 |
|
| |
Yes |
0.93 (0.65–1.32) |
0.68 |
0.88 (0.60–1.29) |
0.52 |
| Ethnicity |
White |
1 |
|
1 |
|
| |
Black |
0.57 (0.38–0.84) |
0.005 |
0.52 (0.34–0.82) |
0.004 |
| |
Indian sub-continent |
0.92 (0.58–1.45) |
0.72 |
0.88 (0.54–1.44) |
0.61 |
| |
Other |
0.96 (0.47–1.96) |
0.91 |
0.75 (0.39–1.47) |
0.41 |
| Schizophrenia |
|
1 |
|
1 |
|
| Delusional disorder |
|
1.64 (0.86–3.14) |
0.13 |
1.59 (0.90–2.82) |
0.11 |
| Brief psychotic disorder |
|
0.66 (0.27–1.64) |
0.37 |
0.74 (0.31–1.78) |
0.50 |
| Schizoaffective disorder |
0.6 (0.38–0.95) |
0.03 |
0.75 (0.44–1.26) |
0.28 |
|
| Non-affective psychosis: NOS |
0.67 (0.35–1.28) |
0.22 |
0.6 (0.26–1.38) |
0.23 |
|
| Psychotic depression |
0.84 (0.49–1.43) |
0.52 |
0.91 (0.53–1.55) |
0.72 |
|
| Psychotic mania |
0.19 (0.13–0.27) |
<0.001 |
0.22 (0.14–0.35) |
<0.001 |
|
| Affective psychosis-NOS |
0.16 (0.09–0.31) |
<0.001 |
0.18 (0.06–0.54) |
0.002 |
|
| Detained within a |
No |
1 |
|
1 |
|
| week of contact |
Yes |
0.68 (0.47–0.98) |
0.04 |
0.66 (0.45–0.96) |
0.03 |
| 1st Contact on Pathway |
Community health & social care |
1.08 (0.65–1.79) |
0.77 |
0.96 (0.54–1.69) |
0.88 |
| Criminal justice system |
1.28 (0.81–2.00) |
0.30 |
1.61 (1.00–2.58) |
0.05 |
|
| |
Hospital medicine |
2.44 (0.44–13.42) |
0.30 |
2.32 (0.88–6.16) |
0.09 |
| |
Psychiatric services |
0.85 (0.27–2.65) |
0.78 |
0.81 (0.29–2.27) |
0.68 |
| |
Accident & Emergency |
0.84 (0.52–1.33) |
0.45 |
0.84 (0.52–1.34) |
0.46 |
| Emergency clinics | 0.42 (0.24–0.72) | 0.002 | 0.40 (0.23–0.71) | 0.002 | |
Note. R2 = 22.3% of variance on 427 observations. Data point for detained within a week of contact complete for only 474. Data point for close persons has only 433 observations. All Poisson estimates used Poisson regression commands in STATA with the vce (robust) command to ensure robust confidence intervals and guarding against over-dispersion and violations of model assumption.