Factors Associated With Self-harm Among People In Long-term Care In Canada

Abstract

Objective: Self-harm is significantly associated with increased risk of suicide. People who are residing in long-term care facilities (LTCF) and receiving home care (HC) services are, conceivably, at highrisk of self-harm since their psychical and psychological needs for long-term care services are also risk factors of suicide. In this study, we aim to examine the incidence of self-harm and investigate factors associated with self-harm among people in long-term care in Ontario, Canada. Methods: Routinely collected administrative data from long-term care settings were linked to electronic health records from Ontario hospitals to identify risk factors of self-harm. Long-term care data were collected using interRAI-LTCF and interRAI–HC, assessment tools for evaluating the needs, strength, and preferences of people in long-term care settings. The initial sample included 3,628,261 assessment records from 499,122 LTCF residents and 2,459,191 records from 901,586 HC clients, between 2003 and 2015. Results: Self-harm records were identified in 1,833 LTCF residents and 4,517 HC clients, corresponding to 3.7 and 5.0 cases per 1000 persons, respectively. In LTCF, highly significant risk factors of self-harm included younger age (OR = 0.95; CI 0.94-0.96), separated or divorced (1.74; 1.28- 2.36), depression (2.28; 1.84-2.82), manic depression (2.61; 1.80-3.79), seizure disorder/epilepsy (1.67; 1.23-2.26), traumatic brain injury (3.32; 2.34-4.70), polypharmacy (2.85; 1.57-5.18), and pain (1.48; 1.18-1.85). In HC settings, additional risk factors were identified, including, male (1.31; 1.19-1.45), psychiatric disorder (2.36; 2.13-2.63), tobacco abuse (1.92; 1.73-2.13), alcohol abuse (2.00; 1.56-2.57), psychotropic medication (1.72; 1.53-1.95), higher score on the depression rating scale (1.65; 1.49-1.84), and loneliness (1.30; 1.17-1.46). In addition, people with a higher level of functional and cognitive impairment were associated with a lower risk of self-harm. Conclusion: Several risk and protective factors were found to be significantly associated with self-harm. The factors identified in this study can be used to develop a predictive algorithm for estimating the risk of self-harm for people in long-term care. The resulting algorithm will be used as a supporting tool for timely identification of at-risk individuals to facilitate effective intervention.