Patterns in health service use during and up to 1 year after an acute episode of hospital-presenting self-harm: data linkage cohort study of over 1.3 million records.
Many people presenting to emergency departments after self-harm do not receive adequate care, even in well-resourced health systems.
To identify patterns of health care service use across two periods: (a) during and (b) up to 1 year after an index emergency department self-harm presentation.
A retrospective population-based cohort study including 4668 individuals aged ≥9 years who presented to the Royal Melbourne Hospital emergency department for self-harm between January 2012 and December 2019. Linked administrative data captured >1.3 million records across primary care, pharmacy, specialist mental and physical health services and emergency departments. Sequential pattern mining identified longitudinal service-use clusters. Multinomial regression explored associations with demographic, clinical, psychosocial and presentation characteristics. Cox proportional hazards models assessed associations between clusters and all-cause and suicide mortality.
Emergency department self-harm presentations triggered short-term increases in multi-sector contacts. However, most (68.7%) reverted to the same service-use cluster observed prior to their index presentation. Suicide risk was highest within 1 year, particularly among those in the specialist mental health services cluster (4.5% of the cohort).
A small subgroup engage intensively with specialist mental health services yet remain at high suicide risk, while one in ten remain disengaged altogether, suggesting that an acute presentation of self-harm often fails to alter patients' care trajectories long-term. Policy alignment with national recommendations for integrated, community-based care could improve sustained, evidence-based support beyond acute crises.
To identify patterns of health care service use across two periods: (a) during and (b) up to 1 year after an index emergency department self-harm presentation.
A retrospective population-based cohort study including 4668 individuals aged ≥9 years who presented to the Royal Melbourne Hospital emergency department for self-harm between January 2012 and December 2019. Linked administrative data captured >1.3 million records across primary care, pharmacy, specialist mental and physical health services and emergency departments. Sequential pattern mining identified longitudinal service-use clusters. Multinomial regression explored associations with demographic, clinical, psychosocial and presentation characteristics. Cox proportional hazards models assessed associations between clusters and all-cause and suicide mortality.
Emergency department self-harm presentations triggered short-term increases in multi-sector contacts. However, most (68.7%) reverted to the same service-use cluster observed prior to their index presentation. Suicide risk was highest within 1 year, particularly among those in the specialist mental health services cluster (4.5% of the cohort).
A small subgroup engage intensively with specialist mental health services yet remain at high suicide risk, while one in ten remain disengaged altogether, suggesting that an acute presentation of self-harm often fails to alter patients' care trajectories long-term. Policy alignment with national recommendations for integrated, community-based care could improve sustained, evidence-based support beyond acute crises.
Authors
Witt Witt, Gan Gan, Gao Gao, Knott Knott, Erlangsen Erlangsen, Madsen Madsen, Schmaal Schmaal
View on Pubmed