Changes in emergency healthcare use following intervention by Navigator, an emergency department social support programme: a multi-centre retrospective before-and-after study.

Background and importance: Patients living with social deprivation, and those with experiences of violence, substance misuse, mental ill-health and homelessness are known to use emergency departments (EDs) more often. It is not known whether a programme of social support initiated during ED attendance may lead to a reduction in healthcare use.

Objectives: The objective of this study is to determine the change in emergency, inpatient and outpatient healthcare use following a social support programme, Navigator, initiated during an ED attendance.

Design: Retrospective before-and-after study.

Settings and participants: Adult patients ≥16 years, attending EDs in the West of Scotland from 14th September 2016 to 10th March 2023, with a Navigator programme encounter.

Intervention or exposure if any: The Navigator social support programme, delivered by trained support workers, initiated during ED attendance, and targeting patients affected by issues including violence, substance misuse, mental ill-health, domestic abuse and homelessness.

Outcome measures and analysis: Healthcare use rates in the 365 days following intervention, as change compared to those in the 365 days prior to the intervention. The primary outcome was the number of ED attendances in the year following intervention compared with the year prior to intervention. Secondary outcomes included inpatient admissions, inpatient bed days, outpatient appointments and outpatient appointments where the patient did not attend. Changes in use rates were analysed with negative binomial regression and reported as incidence rate ratios for interpretation as percentage change. Analysis was repeated for a subgroup of frequent attenders to the ED.

Main results: Of 1421 Navigator programme encounters, 1056 were included for analysis. Median attendance in the year prior to intervention was 3 [interquartile range (IQR) 1-5], and in the year following intervention was 2 (IQR 0-4). Negative binomial regression demonstrated that in the year following Navigator intervention, there was a 29% (95% confidence interval: 24-33%) reduction in ED attendances.

Conclusion: The Navigator programme was associated with reduced emergency and acute healthcare use in the year following intervention, with increased scheduled outpatient care. There is the potential for a social support programme, delivered from the ED, to change patterns of healthcare use, and future work should consider prospectively assessing the impact of such an intervention.