Table 2.
Total sample (n=18 201) | All-cause deaths (n=1767) |
Model 1 |
Model 2 |
||||
---|---|---|---|---|---|---|---|
Incidence risk ratios (95% CI) | p value | Adjusted incidence risk ratios (95% CI) | p value | ||||
Area level indicators | |||||||
Index of Multiple Deprivation (per increase in fifths; from less to more deprived) | .. | .. | 0·98 (0·95–1·02) | 0·34* | 1·03 (0·99–1·07) | 0·19* | |
Urbanicity (per increase in fifths; from less to more urban) | .. | .. | 0·95 (0·92–0·99) | 0·01* | 0·97 (0·93–1·01) | 0·14* | |
Social fragmentation (per unit increase; from less to more fragmented) | .. | .. | 0·95 (0·92–1·00) | 0·03* | 0·98 (0·94–1·03) | 0·45* | |
Individual-level indicators | |||||||
Gender | |||||||
Male | 9610 (52·8%) | 908 (51·4%) | Ref | Ref | |||
Female | 8591 (47·2%) | 859 (48·6%) | 1·07 (0·98–1·18) | 0·13 | 0·86 (0·78–0·94) | 0·0015 | |
Diagnosis | |||||||
Non-affective | 13 160 (72·3%) | 1358 (76·9%) | Ref | Ref | |||
Affective | 5041 (27·7%) | 409 (23·1%) | 0·82 (0·73–0·91) | <0·0001 | 0·83 (0·74–0·93) | 0·0015 | |
Marital status | |||||||
Married or cohabiting | 2781 (15·3%) | 267 (15·1%) | Ref | Ref | |||
Divorced, separated, widowed, or single | 15 420 (84·7%) | 1500 (84·9%) | 1·01 (0·89–1·15) | 0·89 | 1·28 (1·12–1·46) | <0·0001 | |
Substance use disorder | |||||||
No substance disorder | 15 046 (82·7%) | 1519 (86·0%) | Ref | Ref | |||
Life-time substance use disorder | 3155 (17·3%) | 248 (14·0%) | 0·79 (0·69–0·90) | <0·001 | 1·17 (1·02–1·35) | 0·024 | |
Interaction of ethnicity with areal ethnic density | |||||||
Lowest ethnic density area (0% ethnic minorities) | |||||||
White British | .. | .. | Ref | Ref | |||
Ethnic minorities | .. | .. | 0·88 (0·65–1·17) | 0·38 | 0·96 (0·71–1·29) | 0·77 | |
Highest ethnic density area (95% ethnic minorities) | |||||||
White British | .. | .. | Ref | Ref | |||
Ethnic minorities | .. | .. | 0·31 (0·23–0·43) | <0·0001 | 0·52 (0·38–0·71) | <0·0001 | |
p value between ethnicity and ethnic density interaction | .. | .. | .. | <0·0001 | .. | 0·036 |
Data are n (%), unless otherwise specified. Model 1 is crude estimates. Model 2 is adjusted for age, interaction between area-level ethnic density and ethnicity, and all other displayed variables. p values are from Wald tests.
p value for linear trend.