Strategies to manage emergency ambulance telephone callers with sustained high needs: the STRETCHED mixed-methods evaluation with linked data.

Abstract

Background: Emergency ambulance services aim to respond to patients calling with urgent healthcare needs, prioritising the sickest. A small minority make high use of the service, which raises clinical and operational concerns. Multidisciplinary 'case management' approaches combining emergency, primary and social care have been introduced in some areas but evidence about effectiveness is lacking.

Aim: To evaluate effectiveness, safety and costs of case management for people frequently calling emergency ambulance services.

Design: A mixed-methods 'natural experiment', evaluating anonymised linked routine outcomes for intervention ('case management') and control ('usual care') patient cohorts within participating ambulance services, and qualitative data. Cohorts met criteria for 'Frequent Callers' designation; we assessed effects of case management within 6 months on processes, outcomes, safety and costs. The primary outcome combined indicators on mortality, emergency hospital admission, emergency department attendance and emergency ambulance call. Focus groups and interviews elicited views of service providers on acceptability, successes and challenges of case management; interviews with service users examined their experiences.

Setting: Four United Kingdom ambulance services each with one intervention and one control area.

Participants: Natural experiment: adults meeting criteria for 'frequent caller' classification by ambulance services during 2018. Service providers: service commissioners; emergency and non-acute health and social care providers. Service users: adults with experience of calling emergency ambulance services frequently.

Interventions: Usual care comprised within-service management, typically involving: patient and general practitioner letters; call centre flags invoking care plans; escalation to other services, including police. Intervention care comprised usual care with optional 'case management' referral to cross-service multidisciplinary team to review and plan care for selected patients.

Results: We found no differences in intervention (n = 550) and control (n = 633) patients in the primary outcome (adjusted odds ratio: 1.159; 95% confidence interval: 0.595 to 2.255) or its components. Nearly all patients recorded at least one outcome (95.6% intervention; 94.9% control). Mortality was high (10.5% intervention; 14.1% control). Less than 25% of calls resulted in conveyance (24.3% intervention; 22.3% control). The most common reasons for calling were 'fall' (6.5%), 'sick person' (5.2%) and acute coronary syndrome (4.7%). Case management models varied highly in provision, resourcing, leadership and implementation costs. We found no differences in costs per patient of healthcare resource utilisation (adjusted difference: £243.57; 95% confidence interval: -£1972.93 to £1485.79). Service providers (n = 31) recognised a range of drivers for frequent calling, with some categories of need more amenable to case management than others. Some service users (n = 15) reported deep-seated and complex needs for which appropriate support may not have been available when needed.

Conclusions: People who called frequently had a high risk of death and emergency healthcare utilisation at 6 months and were a heterogeneous group. Case management may work for some, but we did not find effects on emergency healthcare utilisation or mortality across the population.

Limitations: This retrospective study provided limited options in selecting control areas, or in meeting recruitment targets. Data quality was variable. Arranging patient interviews proved challenging.

Future research: This should prospectively evaluate different forms of case management; improve data collection; and include patients fully in qualitative components.

Study registration: This study is registered as research registry (www.researchregistry.com/) researchregistry7895.

Funding statement: This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: 18/03/02) and is published in full in Health and Social Care Delivery Research; Vol. 13, No. 37. See the NIHR Funding and Awards website for further award information.