PT  - JOURNAL ARTICLE
AU  - M McCann
AU  - A Maguire
AU  - J Moriarty
AU  - D O’Reilly
TI  - OP86  Poor mental health after bereavement: analysis of population level linked administrative data
AID  - 10.1136/jech-2014-204726.88
DP  - 2014 Sep 01
TA  - Journal of Epidemiology and Community Health
PG  - A42--A43
VI  - 68
IP  - Suppl 1
4099  - http://jech.bmj.com/content/68/Suppl_1/A42.2.short
4100  - http://jech.bmj.com/content/68/Suppl_1/A42.2.full
SO  - J Epidemiol Community Health2014 Sep 01; 68
AB  - Background Circumstances of death, and relationship to the deceased have been highlighted as important risk factors for ‘complicated grief’ symptoms, but most research to date has been based on small, select samples. This project used administrative data to assess a range of risk factors for poor mental health outcomes for a representative population sample. The research questions were a.) Does bereavement affect mental health? b.) Does the likelihood of poor mental health following bereavement vary according to cause of death? And c.) Do individual, household and area characteristics mitigate or compound the risk of poor mental health following bereavement? Methods The Northern Ireland Longitudinal Study is based on a c.28% sample of the Northern Ireland population. Linkages between Health Card Registration information, 2001 Census records, and General Register Office death records were used to identify sociodemographic characteristics and living arrangements in 2001, co-resident deaths between April 2001 and December 2009, and cause of death. Further linkage to the Electronic Prescribing Database provided information on use of psychotropic medication in 2010. As retrospective prescribing information was not available, the study used a nested case-control design including all cohort members. Multilevel logistic regression models (with prescribing practice at level two) were fit with mental health medication use (Yes / No) as the outcome. Results There were 317,264 individuals in the study sample, 31,537 (10%) of whom received a mental health prescription. Around 7% of the cohort (22,353) were bereaved due to illness; 857 (0.3%) were bereaved due to sudden death, and 611 (0.2%) were bereaved through suicide. Within these respective groups, around 10%, 13%, 14% and 22% received a prescription. In fully adjusted models, bereavement was associated with poorer mental health outcome; and sudden death and suicide showed greater elevation in prescribing rates (illness OR 1.22 95% CI 1.14, 1.30; sudden death OR 1.71 95% CI 1.35, 2.16; suicide OR 1.87 95% CI 1.44, 2.44). There was a moderate degree of between-prescriber variation (median odds ratio: 1.28). Further analysis investigated variations with relationship to the deceased, age of the bereaved, and the effect of individual and household ‘coping resources’. Conclusion Bereavement makes a significant contribution to the burden of poor mental health in the population. By providing a more detailed evidence base for the ‘risk profile’ of bereaved individuals, bereavement support services will be better able to identify and provide support for those at greatest risk of poor outcomes.