Skip to main content
. 2023 Jun;10(6):414–425. doi: 10.1016/S2215-0366(23)00110-4

Table 2.

Effect of any stressful life events after onset of psychosis on number and length of subsequent relapses as indexed by hospital admission

Number of relapses (n=253)
Number of relapses (with medication adherence; n=240)
Length of relapses (n=253)
Length of relapses (with medication adherence; n=240)
IRR (95% CI) p value IRR (95% CI) p value IRR (95% CI) p value IRR (95% CI) p value
Any stressful life events before first relapse
No 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Yes 1·87 (1·24–2·80) 0·0026 1·81 (1·15–2·82) 0·010 2·53 (1·40–4·67) 0·0011 2·30 (1·26–4·32) 0·0045
Age of onset, years*
<22 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
22 to <25 0·94 (0·74–1·22) 0·66 0·94 (0·72–1·24) 0·66 0·93 (0·64–1·35) 0·69 1·03 (0·70–1·51) 0·86
25 to <30 1·18 (0·02–55·0) 0·93 0·76 (0·01–40·7) 0·89 0·87 (0·00–259) 0·96 0·24 (0·00–77·8) 0·60
30 to <39 0·79 (0·00–19 437) 0·96 3·20 (0·00–123 453) 0·83 1·58 (0·00–4 302 739) 0·95 40·3 (0·00–150 049 840) 0·59
≥39 1·10 (0·00–3565) 0·98 0·29 (0·00–1273) 0·77 0·78 (0·00–112 259) 0·96 0·06 (0·00–10 706) 0·62
Sex
Female 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Male 0·84 (0·57–1·25) 0·39 0·75 (0·50–1·12) 0·15 1·18 (0·67–2·07) 0·53 1·10 (0·62–1·93) 0·72
Ethnicity
White 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Asian 1·51 (0·57–3·55) 0·37 0·95 (0·31–2·48) 0·93 1·86 (0·55–6·34) 0·28 1·34 (0·38–4·72) 0·63
Black African or Caribbean 1·89 (1·17–3·16) 0·012 1·63 (0·99–2·75) 0·063 2·55 (1·37–4·83) 0·0031 2·37 (1·27–4·47) 0·0069
Mixed 2·05 (0·94–4·23) 0·060 1·78 (0·80–3·75) 0·14 2·28 (0·76–7·20) 0·12 2·03 (0·67–6·39) 0·18
Cannabis use
Former 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Never 1·35 (0·72–2·62) 0·36 1·58 (0·82–3·15) 0·18 1·45 (0·63–3·38) 0·38 1·94 (0·81–4·79) 0·13
Intermittent 1·05 (0·52–2·10) 0·90 1·11 (0·53–2·29) 0·78 2·31 (0·98–5·57) 0·048 2·69 (1·09–6·83) 0·024
Continued hash-type 1·19 (0·41–3·03) 0·73 1·18 (0·40–3·09) 0·75 0·84 (0·20–3·69) 0·81 0·81 (0·19–3·51) 0·77
Continued regular skunk-type 1·04 (0·46–2·29) 0·92 1·00 (0·44–2·21) >0·99 1·51 (0·52–4·52) 0·41 1·45 (0·50–4·34) 0·47
Continued heavy skunk-type 1·64 (0·86–3·19) 0·14 1·45 (0·73–2·93) 0·29 2·08 (0·82–5·38) 0·10 2·31 (0·89–6·17) 0·063
Other drug use
Non-use 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Experimental 0·74 (0·31–1·56) 0·46 0·57 (0·21–1·32) 0·23 1·48 (0·56–4·04) 0·42 1·43 (0·54–3·92) 0·47
Regular 1·91 (1·06–3·40) 0·029 1·82 (0·99–3·28) 0·050 2·29 (0·95–5·65) 0·051 2·28 (0·96–5·58) 0·052
Nicotine use
Non-use 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Intermittent 1·95 (0·85–4·16) 0·095 2·39 (1·04–5·15) 0·032 2·45 (0·93–6·66) 0·066 3·25 (1·23–8·85) 0·017
Continued 2·00 (1·21–3·31) 0·0069 2·33 (1·37–3·99) 0·0019 2·00 (1·05–3·84) 0·032 2·56 (1·34–4·98) 0·0045
Alcohol use
No use 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Continued 1·26 (0·72–2·12) 0·40 1·51 (0·83–2·65) 0·16 1·03 (0·48–2·23) 0·94 1·19 (0·56–2·57) 0·65
Care intensity at onset
Community mental health team 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Crisis team 0·99 (0·30–2·98) 0·99 0·87 (0·25–2·72) 0·82 0·92 (0·21–3·83) 0·90 0·82 (0·18–3·58) 0·79
Non-compulsory admission 2·15 (1·08–4·79) 0·042 1·88 (0·93–4·22) 0·10 2·29 (0·89–6·19) 0·071 2·20 (0·86–5·93) 0·091
Compulsory admission 2·26 (1·18–4·90) 0·024 2·09 (1·07–4·58) 0·045 4·30 (1·79–10·9) 0·0007 4·76 (1·98–12·2) 0·0003
Relationship status
In relationship 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Not in relationship 1·33 (0·87–2·09) 0·20 1·58 (1·00–2·58) 0·057 2·52 (1·36–4·74) 0·0025 3·29 (1·71–6·47) 0·0003
Psychosis diagnosis
Non-affective 1 (ref) .. 1 (ref) .. 1 (ref) .. 1 (ref) ..
Affective 0·74 (0·42–1·23) 0·27 0·91 (0·51–1·56) 0·75 0·48 (0·22–1·03) 0·044 0·55 (0·25–1·21) 0·11
Medication adherence
Regular compliance .. .. 1 (ref) .. .. .. 1 (ref) ..
Irregular compliance .. .. 1·84 (1·17–2·95) 0·010 .. .. 1·63 (0·91–2·94) 0·083
Poor compliance .. .. 3·30 (1·85–5·90) <0·0001 .. .. 2·66 (1·22–5·97) 0·0094

IRRs included adjustment for sociodemographic factors (sex, ethnicity, and relationship status) and clinical factors (psychosis diagnosis, intensity of care at onset, and alcohol, nicotine, and other drug use; with analysis stratified by cannabis use). Where indicated, the model also adjusted for medication adherence. IRR=incidence rate ratio.

*

Age of onset was modelled as a restricted cubic spline function with knots placed at ages of onset of 22 years, 25 years, 30 years, and 39 years.