Can service integration work for universal health coverage? Evidence from around the globe

Highlights

• Empirical impacts of integration experiments were explored through a global review.

• Positive outcomes identified for patients and clinicians without extra costs.

• Incremental improvements so integration should not be considered a ‘game-changer’.

Abstract

Universal health coverage (UHC) is at the heart of the new 2030 Agenda for Sustainable Development. Health service integration is seen by World Health Organization as an essential requirement to achieve UHC. However, to date the debate on service integration has focused on perceived benefits rather than empirical impact. We conducted a global review in a systematic manner searching for empirical outcomes of service integration experiments in UHC countries and those on the path to UHC. Sixty-seven articles and reports were found. We grouped results into a unique integration typology with six categories – medical staff from different disciplines; patients and medical staff; care package for one medical condition; care package for two or more medical conditions; specialist stand-alone services with GP services; community locations. We showed that it is possible to integrate services in different human development contexts delivering positive outcomes for patients and clinicians without incurring additional costs. However, the improved outcomes shown were incremental rather than radical and suggest that integration is likely to enhance already well established systems rather than fundamentally changing the outcomes of care.

Introduction

The new 2030 Agenda for Sustainable Development includes a target (3.8) to “achieve universal health coverage, including…access to quality, essential health-care services” [1]. The presence of universal health coverage (UHC) within the post 2015 international development agenda builds on recent focus on this objective by global agencies such as the World Health Organization (WHO) [2] and World Bank. Their focus had been on the financing and human resource arrangements needed to achieve UHC [3] and to understand which disease programmes were the most cost-effective [4]. More recently, WHO has shifted emphasis to ensuring high quality, integrated service delivery as “critical” such that “UHC and people centred integrated health services should be regarded as interdependent and mutually reinforcing if the goals of UHC are to be realized” [5, p. 14] Ensuring UHC presents unique challenges to the health care system. Demand for formal health care services increase [6] and government has a greater role in financing of health care through partial or full subsidy of health service costs for those who are unable to pay [7]. WHO therefore see integration as a strategy that can achieve a dual purpose: enable expanded and affordable UHC provision and ensure high quality and cost effective service delivery. In other words, align human and financial assets with provision of the right care at the right time in the right place to prevent waste and maximise scarce resources. “This principle is important in countries moving towards universal health coverage since scarce resources are likely to go to waste if governments do not also take action to transform service delivery” [5, p. 14].

The considerable debate on service integration has tended to focus on perceived benefits, such as health system efficiency, cost effectiveness, holistic patient management and better health outcomes for patients. As a result “very few studies report (…) empirically derived outcomes” [8], which only adds to the “urgency to evaluate and assess the efficacy, effectiveness, economics, and implementation” of such activities [9]. Ramsey, Fulop and Edwards agree that the evidence base is limited and focuses on processes of integration with less outcomes “especially regarding patient experience, clinical outcomes and costs” [10, p. 10]. Yaya and Danhoundo note “limited evidence of the impact…on either population health outcomes or the productivity and efficiency of health systems” [11, p. 7]. Since WHO consider integration an essential component to achieve UHC, it is timely to consider whether integration can indeed achieve key impacts associated with UHC such as: (a) positive change in population health and care process outcomes; (b) improved equity of access between different service users; (c) greater health effect at same or less cost of service provision (cost effectiveness); and (d) responsiveness to users and user satisfaction with healthcare services.

The focus of this review is on integration of service delivery, the critical locus of care where patients engage with the health system. This is closely related to the “service or clinical integration” type outlined by Kodner as the “coordination of services and the integration of care in a single process across time, place and discipline” [12, p. 11]. For the purpose of this review we consider, loosely following Atun [see 13, p. 5], that service delivery has four key components – medical conditions, medical professionals, patients, and tangible infrastructure. Our review therefore includes studies from different healthcare settings, delivered by different providers, and in different care locations such as family homes and the community for different patient groups. Our review is aligned with Curry and Ham's three levels of integration – macro, meso and micro – since the focus of our review is situated within the latter two. The meso level is where providers “seek to deliver integrated care for a particular group…with the same…conditions, through the redesign of care pathways and other approaches” and the micro level is where providers “seek to deliver integrated care for individual service users…through care coordination, care planning…and other approaches” [14, p. 7]. We recognise that the different levels and types of integration are inter-related and that broader, macro-level health system integration will impact upon service delivery and outcomes in the mid to long term. However, to manage the diversity and quantity of literature, we explicitly exclude studies examining broader health system integration such as integration between service delivery and education and training systems [e.g. 15], financing [e.g. 16] or integration of planning and regulatory structures. Service integration between formal healthcare and social care is also excluded.

The evidence base on service integration is heterogeneous and contentious, having developed from numerous disciplinary and professional perspectives [8]. Different conceptual definitions are used, with different methodologies in studies of varying sample size, across different contexts. We dealt with this heterogeneity by scanning for empirical impacts in UHC contexts reporting on service delivery changes and grouped our results thematically to create a unique service integration typology. Our typology contrasts and combines results returned from the different studies and identifies patterns in empirical experience of countries pursuing both UHC and service integration and has an advantage of broadly representing the evidence currently available in the public domain. Based on this typology, we generalise our results for UHC countries not returned in our search as well as countries yet to invest in UHC.

The greater part of previous literature reporting on service integration outcomes and integration typologies is based on evidence from high income countries (HICs) [10], [14], [17]. Our analysis examines integration in both high and low and middle income countries (LMICs). A global UHC agenda in which integrated care is fundamental must take into account the different contexts of healthcare delivery within and between countries so that all groups, particularly the vulnerable and marginalised, are included. High income countries are already moving to integrated service configurations and delivery but such care is not universally accessible to many groups, such as ethnic minorities, adolescents and chronic drug addicts, each of which is under-represented in care. Low and middle income countries have yet to significantly implement integrated care in their public health systems [18], not least because of the complexities of implementation in contexts where healthcare infrastructure and human resources for health are poor. In conflict and post conflict countries, such infrastructure is often non-existent.

The rest of this article is structured as follows. The next section describes the methods used to undertake the review. Based on our results, we present our typology cross-referencing integration type to empirically reported impacts, as well as to different human development contexts. We go on to discuss the evidence base and consider whether service integration can work for UHC as hoped for by international development policy makers.