Centralisation of specialised healthcare services: a scoping review of definitions, types, and impact on outcomes.

Abstract

Background: Centralising specialised healthcare services into high-volume centres is proposed to improve patient outcomes and efficiency. Most reviews focus on relatively few conditions and a limited range of outcomes.

Objectives: To review the evidence on centralisation of a range of specialised acute services, to analyse (1) how centralisations are defined; (2) how centralisations are organised and delivered; and (3) the relationship between centralisation and several key outcomes.

Design: Scoping review, conducted in November 2020.

Setting: Specialised acute healthcare services.

Intervention: Centralisation of services into a reduced number of high-volume units.

Findings: We included 93 papers covering specialised emergency and elective acute healthcare services, published to November 2020. Definitions of centralisation commonly lacked detail, but, where available, covered centralisation's form, objectives, mechanisms and drivers. We proposed a typology of four forms of centralisation, reflecting the number and functions of specialist units (centralisation of whole pathway, centralisation of pathway components, hierarchy of specialist units, partial centralisation). For most outcomes, the majority of papers suggested a positive impact of centralisation: mortality (33/55 papers), survival (19/25), morbidity (17/27), quality of life (6/7), quality of care (22/30), length of stay (17/26), cost-effectiveness (3/3) and patient experience (3/3). Centralisation was associated with increased patient travel (9/12); 3/5 papers suggested no impact on inequalities.

Limitations: This review was conducted in November 2020 and did not include grey literature or studies that did not analyse outcomes, so more recent and further evidence - for example, on types of centralisation model and how centralisation was implemented - may exist. As this was a scoping review, we did not conduct a quality assessment, which may reduce the confidence with which we may view the presented impacts of centralisation.

Conclusions: Centralisation is commonly associated with improved care and outcomes. However, research seldom describes centralised services in sufficient detail, rarely compares different service models and tends to focus on a narrow range of outcomes. Therefore, understanding the extent and nature of centralisation's impact - and the mechanisms by which it is achieved - remains elusive. By addressing these gaps, future research may of greater use to all stakeholders with an interest in centralisation.

Future research: Should provide clearer descriptions of centralisations, compare different centralisation models and study a wider range of important outcomes, including patient experience and cost-effectiveness.

Funding: This article presents independent research funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme as award number NIHR133613.