Investigating the use and impact of community Care (Education) and Treatment Reviews (C(E)TRs) in people with intellectual disability and autistic people: protocol for a cohort study using electronic health records.

Abstract

Introduction: Care (Education) and Treatment Reviews (C(E)TRs) are intended to reduce unnecessary psychiatric hospital admission and length of stay for people with intellectual disability and autistic people. The use and impact of C(E)TRs have not been systematically evaluated since their introduction in England in 2015. The aims of this study are to describe the demographic and clinical profiles of people who receive a community C(E)TR and to investigate their effects on admission, length of hospital stay and clinical and functional change.

Methods and analysis: We will conduct a retrospective cohort study using de-identified data from electronic health records derived from two large National Health Service mental health providers in London, England, including one replication site. Data will be extracted using the Clinical Record Interactive Search (CRIS) tool for all people with recorded intellectual disability and/or autism who received mental healthcare from 2015. We will identify community C(E)TR events using keyword searches. Community C(E)TRs will be examined in two ways: (1) In a community cohort, we will capture data in the 6-month periods before and after a community C(E)TR and compare this to a matched control group and (2) In a hospital cohort, we will compare groups who did and did not receive a community C(E)TR prior to their admission. We will describe the socio-demographic and clinical profiles of each group and their health service use, and compare C(E)TR and no C(E)TR groups using t-tests (or a non-parametric equivalent). The primary outcomes are admission to a psychiatric hospital (community cohort) and length of psychiatric hospital admission and clinical change (hospital cohort). Admission to psychiatric hospital will be estimated using propensity score weighting and difference-in-differences methods. Cox's proportional hazard model will be used for length of hospital admission and repeated-measures analysis of variance (ANOVA) will be used to assess clinical change.

Ethics and dissemination: Use of CRIS to examine de-identified clinical data for research purposes has overarching ethical approval. This study has been granted local approval by the South London and Maudsley CRIS Oversight Committee. Findings will be disseminated in an open-access peer-reviewed academic publication, at conference presentations, and to service users and carers in accessible formats.