Factors affecting psychiatric bed utilisation by people with intellectual disabilities: A time series analysis using the English National Mental Health Services data set

Background: In 2015, the Building the Right Support programme was launched for England in an attempt to reduce the number of psychiatric inpatients with intellectual disabilities and/or autism by 35%–50%. This target, and subsequent targets, were missed, and for 2025–2026, the government further committed to reducing numbers by 10%. Considering these continued targets, we aimed to investigate psychiatric bed utilisation over time, and to further understand factors that may influence psychiatric admissions and discharges of people with intellectual disabilities and/or autism, by utilising time series modelling with national English data to explore the relationship between a set of chosen sociodemographic, clinical and service-related predictor variables and the following outcome variables: (1) total monthly number of hospital spells, (2) total monthly number of discharges, (3) total monthly number of admissions, (4) ratio of community to non-community discharges, (5) number of inpatients with a length of stay under 2 years, (6) number of patients with a length of stay over 2 years and (7) total number of distinct individuals who had been subjected to restraints.

Methods: Using data from the publicly available Mental Health Services Data set, we utilised linear regression (with moving average or auto-regressive errors) to examine the relationships between variables over time, from February 2013 to January 2024.

Results: Over time, the number of inpatients decreased by an average of 4.55 patients per month. The number of inpatients with a length of stay greater than 2 years reduced over time. Periods of time when the number of inpatients was greater were associated with more inpatients under the age of 18 years. Periods of time when hospital stays, admissions and discharges were higher were associated with fewer White inpatients relative to non-White inpatients. Periods of time with more patients detained under Part II of the Mental Health Act were associated with more admissions and the increased use of restraint.

Conclusions: Over the last 11 years, the planned closure of psychiatric inpatient beds has been unsuccessful. Our findings indicated that periods of increased psychiatric bed utilisation were associated with more admissions of younger people, non-White inpatients and those likely to be experiencing a crisis. Future research should explore how psychiatric beds can be utilised more effectively alongside community-based services and long-term trajectories using participant level data.