TY - JOUR
T1 - The complexities of suicide
T2 - a multilevel survival analysis examining individual, familial and neighbourhood determinants of suicide risk using Danish register-based data
AU - Davies, Megan
AU - Horsdal, Henriette Thisted
AU - Antonsen, Sussie
AU - Sigsgaard, Torben
AU - Fan, Chun Chieh
AU - Thompson, Wesley K.
AU - Pedersen, Carsten Bocker
AU - Sabel, Clive E.
N1 - Publisher Copyright:
Copyright © The Author(s), 2022. Published by Cambridge University Press.
PY - 2023/10/14
Y1 - 2023/10/14
N2 - Background Suicide risk is complex and nuanced, and how place impacts suicide risk when considered alongside detailed individual risk factors remains uncertain. We aimed to examine suicide risk in Denmark with both individual and neighbourhood level risk factors. Methods We used Danish register-based data to identify individuals born in Denmark from 1972, with full parental information and psychiatric diagnosis history. We fitted a two-level survival model to estimate individual and neighbourhood determinants on suicide risk. Results We identified 1723 cases of suicide in Denmark during the follow-up period from 1982 to 2015. Suicide risk was explained mainly by individual determinants. Parental comorbidities, particularly maternal schizophrenia [incidence rate ratio (IRR): 2.29, 95% CI 1.56-3.16] and paternal death (2.29, 95% CI 1.31-3.72) partly explained suicide risk when adjusted for all other determinants. The general contextual effect of suicide risk across neighbourhoods showed a median incidence rate ratio (MRR) of 1.13 (1.01-1.28), which was further reduced with full adjustment. Suicide risk increased in neighbourhoods with a higher proportion of manual workers (IRR: 1.08; 1.03-1.14), and decreased with a higher population density (IRR: 0.89; 0.83-0.96). Conclusion Suicide risk varies mainly between individuals, with parental comorbidities having the largest effect on suicide risk. Suicide risk was less impacted by neighbourhood, though, albeit to a lesser extent than individual determinants, some characteristics were associated with suicide risk. Suicide prevention policies might consider targeting interventions towards individuals more vulnerable due to particular parental comorbidities, whilst taking into account that some neighbourhood characteristics might exacerbate this risk further.
AB - Background Suicide risk is complex and nuanced, and how place impacts suicide risk when considered alongside detailed individual risk factors remains uncertain. We aimed to examine suicide risk in Denmark with both individual and neighbourhood level risk factors. Methods We used Danish register-based data to identify individuals born in Denmark from 1972, with full parental information and psychiatric diagnosis history. We fitted a two-level survival model to estimate individual and neighbourhood determinants on suicide risk. Results We identified 1723 cases of suicide in Denmark during the follow-up period from 1982 to 2015. Suicide risk was explained mainly by individual determinants. Parental comorbidities, particularly maternal schizophrenia [incidence rate ratio (IRR): 2.29, 95% CI 1.56-3.16] and paternal death (2.29, 95% CI 1.31-3.72) partly explained suicide risk when adjusted for all other determinants. The general contextual effect of suicide risk across neighbourhoods showed a median incidence rate ratio (MRR) of 1.13 (1.01-1.28), which was further reduced with full adjustment. Suicide risk increased in neighbourhoods with a higher proportion of manual workers (IRR: 1.08; 1.03-1.14), and decreased with a higher population density (IRR: 0.89; 0.83-0.96). Conclusion Suicide risk varies mainly between individuals, with parental comorbidities having the largest effect on suicide risk. Suicide risk was less impacted by neighbourhood, though, albeit to a lesser extent than individual determinants, some characteristics were associated with suicide risk. Suicide prevention policies might consider targeting interventions towards individuals more vulnerable due to particular parental comorbidities, whilst taking into account that some neighbourhood characteristics might exacerbate this risk further.
KW - Health registers
KW - Multilevel modelling
KW - Suicide
KW - Survival analysis
UR - http://www.scopus.com/inward/record.url?scp=85172425666&partnerID=8YFLogxK
U2 - 10.1017/S0033291722003701
DO - 10.1017/S0033291722003701
M3 - Article
C2 - 36515183
AN - SCOPUS:85172425666
SN - 0033-2917
VL - 53
SP - 6356
EP - 6365
JO - Psychological Medicine
JF - Psychological Medicine
IS - 13
ER -