Table 3.
Sub-hazard ratio estimates for natural-cause mortality in people with severe mental illness
| Total sample (n=18 201) | Natural-cause deaths (n=1417) |
Model 1 |
Model 2 |
||||
|---|---|---|---|---|---|---|---|
| Sub-hazard ratios (95% CI) | p value | Sub-hazard ratios (95% CI) | p value | ||||
| Area-level indicators | |||||||
| Index of Multiple Deprivation (per increase in fifths; from less to more deprived) | .. | .. | 0·95 (0·91–1·00) | 0·05* | 1·02 (0·97–1·06) | 0·49* | |
| Urbanicity (per increase in fifths; from less to more urban) | .. | .. | 0·94 (0·90–0·98) | 0·005* | 0·99 (0·94–1·03) | 0·55* | |
| Social fragmentation (per unit increase; from less to more fragmented) | .. | .. | 0·93 (0·88–0·97) | 0·002* | 0·98 (0·93–1·03) | 0·41* | |
| Individual-level indicators | |||||||
| Gender | |||||||
| Male | 9610 (52·8%) | 707 (49·9%) | Ref | Ref | |||
| Female | 8591 (47·2%) | 710 (50·1%) | 1·25 (1·11–1·39) | <0·0001 | 0·92 (0·82–1·02) | 0·11 | |
| Diagnosis | |||||||
| Non-affective | 13 160 (72·3%) | 1090 (76·9%) | Ref | Ref | |||
| Affective | 5041 (27·7%) | 327 (23·1%) | 0·90 (0·80–1·02) | 0·11 | 0·82 (0·73–0·93) | 0·002 | |
| Marital status | |||||||
| Married or cohabiting | 2781 (15·3%) | 220 (15·5%) | Ref | Ref | |||
| Divorced, separated, widowed, or single | 15 420 (84·7%) | 1197 (84·5%) | 0·87 (0·74–1·01) | 0·07 | 1·16 (1·00–1·35) | 0·05 | |
| Substance use disorder | |||||||
| No substance use disorder | 15 046 (82·7%) | 1251 (88·3%) | Ref | Ref | |||
| Life-time substance use disorder | 3155 (17·3%) | 166 (11·7%) | 0·61 (0·52–0·71) | <0·0001 | 0·91 (0·77–1·06) | 0·22 | |
| Interaction of ethnicity with areal ethnic density | |||||||
| Lowest ethnic density area (0% ethnic minorities) | |||||||
| White British | .. | .. | Ref | Ref | |||
| Ethnic minorities | .. | .. | 0·69 (0·51–0·94) | 0·019 | 0·78 (0·57–1·05) | 0·11 | |
| Highest ethnic density area (95% ethnic minorities) | |||||||
| White British | .. | .. | Ref | Ref | |||
| Ethnic minorities | .. | .. | 0·31 (0·22–0·44) | <0·0001 | 0·44 (0·32–0·62) | <0·0001 | |
| p value between ethnicity and ethnic density interaction | .. | .. | .. | 0·013 | .. | 0·071 | |
Data are n (%), unless otherwise specified. Competing risks regression models with robust standard errors to adjust for clustering at Lower Super Output Area level were used to estimate sub-hazard ratios. Model 1 is crude estimates. Model 2 is adjusted for age, an interaction between area level own ethnic density and ethnicity, and all other displayed variables. p values are from Wald tests.
*
p value for linear trend.