Shared care-is it worth it for the patient?

Introduction

For any specialized medical service there is a fundamental tension between the demands for quality, access and sustainability, with affordability an ever more important factor. For cystic fibrosis (CF) services, patients ideally want their care from a high quality specialist CF centre, delivered on their door step with no limitation on the services provided. In practice this may be available only to those who live in close proximity to a CF centre, and the further a patient lives from a CF centre the more difficult it may be to offer optimal care. For patients who do not live close to a CF centre, compromises will have to be made between quality and access. The ideal should be to deliver specialized services as close as possible to the patient's home without compromising quality.

DECLARATIONS

Competing interests

None declared

Funding

None

Ethical approval

Not applicable

Guarantor

ID

Contributorship

Iolo Doull is the sole contributor

Underpinning the construction of guidelines for specialized medical services is the belief that specialized care is better than non-specialized care, and that there is a minimum number of patients who need to be cared for within that service to maintain expertise. Although there is considerable evidence of such effects in adult specialities, there is less evidence in children's specialities. It is arguable that this is a reflection of the difficulty with outcome measures for paediatric practice compared to adult practice – mortality is frequently used in adult studies, but in the 21st century is applicable to few paediatric diseases. This review will assess the evidence for specialized services for adults and children, and specifically the evidence for specialist and shared care for adults and children with CF.

Specialized adult services

A central axiom of most guidelines advocating specialized care is that there is a volume-outcome relationship and that a minimum number of patients must be cared for to gain expertise and thus justify specialized care. So specialized centres that care for large numbers of patients (increased volume) will develop expertise in that condition and this will result in improved outcomes for their patients. Such is the belief in the volume-outcome relationship that for some diseases without easily measurable outcomes, volume may be used as a surrogate for outcome.

It is unlikely that the volume-outcome relationship is linear – rather it probably rises steeply before flattening out (at a tipping point) to a gentle slope (Figure 1). It is possible that above a certain volume the V-O curve could go downwards. The ideal guideline would clearly identify the tipping point of the V-O curve, and set that volume as the minimum recommended number of patients to be cared for to offer specialized care. The current UK CF Trust guidelines ¹ for the care of patients with CF would suggest that the tipping point of the V-O curve is around 100 patients, while there is a general feeling that the V-O may decline when the volume is greater than 300 patients.

Figure 1

Relationship between volume and outcomes demonstrating tipping point

There is a large body of evidence in adults of improved outcomes for those receiving specialized care, with clear evidence of a volume-outcome relationship for many surgical procedures. Luft demonstrated that mortality due to open-heart surgery, vascular surgery, transurethral resection of the prostate, and coronary bypass decreased with increasing number of operations, with hospitals in which 200 or more of these procedures were performed annually having lower case mix adjusted death rates than hospitals with lower volumes. ² Adult patients with a myocardial infarct treated in dedicated coronary care units,^3,4 patients with a stroke cared for in dedicated stroke units,^5,6 and patients with intra-cerebral haemorrhage treated in neurosurgical intensive care units have improved outcomes compared to generalist care. ⁷ Mortality is the usual outcome measure for studies in adults.

Specialized paediatric services

Mortality is less useful as an outcome measure in studies in children as the mortality for most conditions (including CF) is mercifully low, and so softer outcome measures are often utilized. The exceptions are rarer conditions such as biliary atresia where a series of comprehensive national audits in the United Kingdom and France have demonstrated a clear relationship between center volume and mortality. ⁸ For congenital heart disease in children there is a consistent body of evidence for a surgical volume outcome relationship, particularly for high risk procedures.^9–14 There are older studies suggesting that children with rarer tumours have improved outcomes and survival if treated in specialized paediatric oncology centres,^15–17 and observational studies from the United States report lower mortality and morbidity rates when paediatric brain tumours are operated on by high-volume neuro-surgeons in high-volume centres.^18,19 There is evidence for improved survival and outcomes for both medical and surgical care of preterm infants in specialized centres,^20,21 and there is limited historical evidence for specialized asthma services in the US, ²² but it is noteworthy that CF care is often cited as a justification for specialized services in children.

Specialized versus non-specialized CF care

There is clear consensus that patients with CF should receive care from a specialist CF centre multidisciplinary team (MDT),^1,23,24 and there is historical literature to support these statements. It is however arguable that many of the cited supportive studies are historical, bearing little relevance to modern CF care.^25–29 Many of these studies reflect treatment practices and models of care from 20–30 years ago, and are not representative of the expectations and prognosis of patients with CF in the 21st century. Specifically they do not address the issue of shared care.

Although mortality is the hardest outcome measure in CF, with the improved life expectancy in CF, ³⁰ surrogate markers are frequently used. The major determinants of survival in CF are the forced expiratory volume in 1 second (FEV₁) expressed as a percentage predicted for age, sex and height, chronic Pseudomonas aeruginosa infection and poor nutritional status. ³¹ Increasingly FEV1% predicted is used as the most important outcome measure, although it is important to note that due to the large inherent variability in FEV1% predicted, smaller studies may have insufficient power to detect meaningful differences between models of care. ³²

It is now nearly 30 years since Phelan and Hey ²⁵ highlighted that mortality from CF in England and Wales was nearly double that in Victoria, Australia and speculated that this was due to the fact that 90% of patients in Victoria were seen in a specialist CF centre. There were similar conclusions from American ³³ and Danish reports ³⁴ – that specialist clinics improved outcomes. Since then, there have been a number of reports of specialist care offering advantages in important outcomes including mortality, ²⁹ weight gain ^{25 – 29} and symptom control. ²⁶ A complex assessment of a relatively small number of Dutch children undergoing a combination of newborn screening and centre based care, reported no significant effect on pulmonary function,^27,29 weight or height standard deviation scores(SDS), ²⁷ but the possibility of increased survival in those that received centre based care. ²⁹ An Australian observational study reported improved weight gain in adolescents who received specialist centre care compared to local care, but again no significant difference in pulmonary function; ^{25 – 29} while adults attending specialist cystic fibrosis clinics self-report better symptom control and greater service satisfaction compared to those attending general clinics. ²⁶

Amongst the strongest evidence for specialist CF services was the observational study reported by Mahadeva ³⁵ on adult CF patients who did or did not receive specialist paediatric and/or adult CF centre care. There was a significant dose response improvement in higher body mass index (BMI), FEV₁ and better CXR severity score in those who had received care from both a paediatric and adult CF centre, compared to those who received only specialist adult care and those who received neither paediatric nor adult specialist care. Of concern was an earlier age of pseudomonas aeruginosa infection in those who received specialist paediatric and adult care, although it was suggested that this might reflect more comprehensive microbiological surveillance in specialist CF centres. Despite this, a recent systematic review of specialized care concluded that for CF, outcomes were not superior in specialized centers compared with other care. ³⁶

Centre based versus shared care

With acceptance of the benefits of specialist care, alternative models of care have developed for children with CF geographically distant from CF centres. Models of shared care developed, where the patient receives the majority of their care from a local hospital, but has input from a specialist CF centre team either through travel to the CF centre, ³⁷ or review at their local hospital by the visiting CF centre multidisciplinary team (MDT).^38,39 There may be considerable variation in frequency of review by the CF centre MDT. Currently there is no evidence to support shared adult CF care, and national guidelines unanimously recommend CF centre based care for adults. ¹

There are a number of theoretical advantages to shared care. The patients are seen locally, usually in a smaller more communal hospital, with consequently much less disruption to the child and family's everyday living. The stress of regular CF clinic attendances should never be underestimated, but this is likely to be magnified if it is accompanied by an arduous journey before and after clinic. Many non-CF services (such as enuresis services) are locality based making it inappropriate or impossible to deliver from the CF centre, and for smaller clinics the risk of cross infection should be less than attending larger, more heterogeneous clinics. Finally, there is a perception that clinic attendance is improved if the journey is easy rather than a complex journey to attend a CF centre. The major disadvantages of shared care would be the possibility of inferior quality care leading to potentially worse outcomes.

Observational studies from The Netherlands ³⁷ and Australia ³⁸ have reported similar outcomes for children who received specialist care compared to those who received shared care. van Koolwijk compared outcomes of 105 Dutch children receiving either centralized care, close shared care with the CF centre or localized care (seen only once a year by the CF centre). They reported no significant differences in nutritional outcomes or rate of chronic pseudomonas aeruginosa between different models of care, with the only significant difference being a slightly lower FEV₁% predicted in those who received shared care. ³⁷ Similarly, Thomas compared outcomes for 273 Australian children receiving either centre based care or three models of outreach care, and reported similar outcomes apart from a higher likelihood for hospital admission for those receiving centre based care. ³⁸ Despite challenging distances, the vast majority of patients receiving outreach care in this report were reviewed at least twice per year by the CF centre specialist team.

In contrast an observational study of 199 children and young people with CF in South and Mid Wales reported significantly decreased pulmonary function in those who received shared care. ³⁹ In Wales, three models of care evolved – full centre care, where the child is seen regularly (including annual review) at the specialist CF centre; hybrid care, where the child is usually reviewed at least 3 times a year (including annual review) by the specialist CF centre MDT at the local hospital, with additional reviews by the local clinic team; and shared care where the child is seen regularly by the local clinic team and usually has only their annual review performed locally by the visiting consultant with or without the specialist CF MDT.

There was no significant difference in weight standard deviation scores, height SDS or BMI SDS between the three models of care or in the rate of chronic Pseudomonas aeruginosa infection. However, in the 123 subjects able to perform spirometry there was a significant difference between full, hybrid and shared care in mean FEV₁ (2.26 v 1.96 v 1.77 liters; P = 0.008), FEV₁% predicted (89.2 v 88.9% v 74.5%; P = 0.001) and forced vital capacity (FVC) (2.74 v 2.41 v 2.25 liters; P = 0.04). There were no significant differences in the use of long term treatments, suggesting that treatment burden could not explain the differences.

In the US, CF centre care frequency of review by the specialist CF team is associated with improved outcome measures, ⁴⁰ and the authors of the Welsh study speculated that the better outcomes seen in those subjects receiving hybrid care compared to shared care might reflect the more regular review by the CF centre MDT. If specialist care is indeed superior to local care, it is arguable that a single clinic appointment per year for an annual assessment with a specialist is unlikely to significantly influence a chronic life-long condition such as CF. Thus it is noteworthy that the new UK CF Trust guidelines on care mandate that all patients should be seen at least twice a year by the specialist centre MDT. ¹

Conclusions

Modern healthcare systems aim to deliver equitable care over wide geographical areas, and in CF shared care is increasingly offered to children and young people. There are many theoretical benefits in terms of access and convenience, but these must not be at the expense of quality. Recent reports suggest that appropriately delivered shared care can be as effective as centre based care, independent of the distance from the CF centre. However, this requires regular review by the CF centre MDT, and thus has significant financial implications.