How much centralization of critical care services in the era of telemedicine?

Abstract

Editorial The goal of modern health care is to improve outcomes and reduce costs. Centralisation, defined as the reorganisation of healthcare services into fewer specialised units, is one of the common strategies. The rationale is that increasing the volume and variety of cases promotes the development of highly specialised services, increases experience and efficiency, facilitates training, limits costs and reduces clinical variability [1–3]. The notion of focussing on volume to promote specialist expertise is well established in surgery. There is a clear association between volume of surgical cases and survival, even if workload increases [1]. Obvious examples are large cardiovascular units and trauma centres. The reasons for better outcomes are multifactorial, including expert teams, a high-level infrastructure with evidence-based protocols and standardised governance processes, state-of-the-art diagnostic tests and therapies, and cost-effective purchasing (Table 1). Critical care medicine is a complex, expensive and resource intensive specialty where centralisation has also received attention. A retrospective study of >20,000 mechanically ventilated, non-surgical adult patients concluded that ICU and hospital mortality were significantly lower in high-volume hospitals [4]. The “Conventional ventilatory support versus Extracorporeal membrane oxygenation (ECMO) for Severe Adult Respiratory failure” (CESAR) trial showed that outcomes were better in all patients transferred to the specialist unit regardless of whether they received ECMO or not [5]. Neurocritical care units have been shown to improve patient outcomes and reduce mortality, resource utilisation and costs compared to district hospitals [6]. Apart from specialist-led care, rapid access to neurosurgical intervention plays a role. Finally, a review of centralised paediatric critical care in Australia revealed that the odds ratio of mortality in the UK versus Australia was 2.09 [7]. The authors estimated that 453 deaths a year in the UK could be avoided if all children requiring mechanical ventilation for >12–24 h were transferred to specialist paediatric ICUs. However, the association between volume and outcomes is not consistently seen. Data from 2812 US hospitals showed that quality of care for elderly patients with pneumonia was lower among hospitals with the highest rates of ICU admission [8]. Similarly, an analysis of > 18,000 ECMO patients revealed that mortality was higher in high-volume compared to low-volume centres [9]. Whether this represents selection bias, differences in criteria for applying ECMO or any other variation in practice is unclear. Centralisation of limited resources has other unpredictable negative effects which can be broadly categorised into factors related to the geographical distance between centres, transport, the effects on staff in non-specialist centres, and the psychological impact on the patient and their relatives. Serious in-transit critical events may occur, including equipment failure and technical problems [10–12]. A review of 5144 urgent land transports revealed that critical events occurred in approximately 1 in 15 transports [12]. Hypotension was the most common incident. An observational study of > 10,000 patients with potentially life-threatening conditions showed an association between journey distance to hospital and mortality after adjustment for age, sex, clinical category and illness severity [10]. A 10-km increase in distance was associated with a 1% absolute increase in mortality. In contrast, a Canadian retrospective case-cohort study did not find an association between duration of transport and hospital mortality [13]. Instead, a longer time spent by paramedics at the sending hospital was associated with shorter length of stay in the referring hospital. At an institutional level, centralisation may lead to a reduction in available specialists in regional centres and the closure of specialty programmes, resulting in reduced job satisfaction and staff morale [11]. Another drawback is the impact on families and relatives, together with longer travel times and increased costs. Furthermore, patients are removed from their local networks which makes it more challenging to organise