Determinants of acute psychiatric inpatient length of stay in Ireland

This Bulletin summaries the findings from Brick, A., B. Walsh, L. Hill, C. Behan, A. Daly and S. Craig (2024). 'Determinants of acute psychiatric inpatient length of stay in Ireland', Irish Journal of Medical Science. Available at: https://doi.org/10.1007/s11845-024-03819-x

Introduction

Ireland has historically had an over-reliance on the provision of mental health services in inpatient settings (e.g. a psychiatric hospital). This has been accompanied by poorly resourced community services. In recent years there has been a substantive change in policy direction in Ireland, and internationally, to providing mental health services in the community when appropriate. While a decrease in the number of people admitted has been observed, the current inpatient system is under pressure, with high occupancy rates and significant increases in service requirements projected. A clear understanding of the determinants of inpatient length of stay (LOS), including geographic variation, would allow service providers to target community services where they would be most effective in diverting care from in-patient settings entirely or achieving earlier discharge.

Data and methods 

This study focuses on patients discharged from the 29 publicly financed adult inpatient facilities classified as acute services over the 2015–2019 period. We use anonymised patient-level information from the National Psychiatric Inpatient Reporting System which includes sociodemographic characteristics such as age, sex, marital status, employment status, and ethnicity. We also include information on the clinical profile of patients such as their diagnosis, legal status, whether this was a first-time admission or a readmission, and the patient area of residence. As well as this data about individual patients, we use data on the acute and community service provision in each area of residence: the number of acute beds per 100,000 population, staffing of community services compared to policy recommendations and waiting list pressures. Our final sample included 60,607 discharges. Statistical methods which include all the above-mentioned information were then used to examine the determinants of inpatient LOS.

Results

We find large geographic variation in the use of psychiatric inpatient services and community provision. The statistical analysis finds that longer LOS is associated with older age, being female, and having an involuntary admission. Clinical diagnosis significantly impacts LOS, with diagnoses such as schizophrenia being associated with significantly longer stays than for example depressive episodes. Average LOS was 20.9 days but there is also significant variation across area of residence, with people discharged in some regions staying up to 5 days longer than similar cases in other regions. 

Interestingly we find that our included community provision metrics do not impact on LOS. This result is not surprising however, as the current lack of data collected on services provided in the community is a limitation of the study. There are many specialist mental health services provided in a community setting (e.g. day hospitals, home care teams, and assertive outreach teams) that may facilitate earlier discharge from acute care but on which no data currently exists.

Conclusions

The regional variation we find in psychiatric inpatient LOS suggests a relationship between the provision of community services and the amount of care required in inpatient settings. However, substantive improvements in the collection and reporting of data about community services are needed to improve future planning of mental health services, both community and inpatient.