Article Text

Download PDFPDF

Full, shared and hybrid paediatric care for cystic fibrosis in South and Mid Wales
  1. Iolo Doull1,
  2. Hazel Evans2,
  3. South and Mid Wales Paediatric Cystic Fibrosis Network
  1. 1Respiratory/Cystic Fibrosis Unit, Children's Hospital for Wales, Cardiff, UK
  2. 2Southampton General Hospital, Southampton, UK
  1. Correspondence to Dr Iolo Doull, Respiratory/Cystic Fibrosis Unit, Children's Hospital for Wales, Cardiff CF14 4XN, UK; doullij{at}


Background Although care for children with cystic fibrosis (CF) is increasingly shared between CF centres and local CF clinics, the optimal model is unclear.

Objectives The authors compared three models of care within a well established CF network: full centre care; local clinic based care with annual review by the CF centre; and hybrid care, where the child is usually reviewed at least three times a year by the specialist CF centre.

Results Of 199 children and young people with CF in South and Mid Wales, 77 were receiving full care, 102 shared care and 20 hybrid care. There were no significant differences in baseline characteristics, nutritional outcomes or use of chronic therapies. There was however a statistically significant difference between full, shared and hybrid care in mean forced expiratory volume in 1 s (FEV1) per cent predicted (89.2% vs 74.5% vs 88.9%; p=0.001).

Conclusions These differences in pulmonary function are likely to reflect the model of care received, and may affect long term outcomes.

Statistics from


Although there is consensus that all patients with cystic fibrosis (CF) should receive their care from a specialist CF centre multidisciplinary team (MDT),1,,3 there is debate over the optimal model for the delivery of care. Patients receiving all their care from a specialist CF centre (full care) have historically had significantly better pulmonary function, nutrition and survival compared to those receiving local hospital based care.4,,9 More recently, an alternative model of shared care has developed for children with CF, where the patient receives the majority of their CF care from a local hospital, but in addition has input from a specialist CF centre team through either travel to the CF centre10 or review at their local hospital by the visiting CF centre MDT.11

Over the last 12 years the paediatric CF unit in Cardiff has promoted a model of shared care where the patients received the majority of their care at their local hospital, with specialist input from the CF centre. Areas of South and Mid Wales are relatively isolated with poor road links, and our aim was to deliver optimal CF care independent of the distance from the CF centre. The objectives of this study were to compare measures of nutrition, pulmonary function and prevalence of chronic Pseudomonas aeruginosa infection between patients receiving full care and those receiving shared care in a well established CF network.

Subjects and methods

Wales has a paediatric population of approximately 600 000, of whom 80% live along the southern coastal corridor and 20% live along the northern coastal corridor (Mid Wales is rural and comparatively sparsely populated). The majority of patients with CF in South and Mid Wales receive local care from 10 paediatric units, and specialist CF care from the Children's Hospital for Wales in Cardiff. Because of differences for funding of specialist care between Wales and England, and our long established paediatric network of CF care, we are confident that we are aware of all children diagnosed with CF in South and Mid Wales. All patients under our care are offered a structured annual review as recommended by the UK CF Trust. Since 1996, 97% of patients with CF in Wales have been diagnosed through newborn screening.12

What is already known on this topic

  • It is recommended that all patients with cystic fibrosis (CF) receive care from a specialist CF centre multidisciplinary team (MDT).

  • Increasingly, care is shared between local hospital based care and specialist CF centres.

  • There is considerable debate over the optimal model for delivery of care and how frequently the CF centre should review patients.

What this study adds

  • Patients who received full care had better lung function than those who received primarily local hospital based care.

  • Frequency of review by the CF centre MDT is more important than distance from the CF centre.

Over the years three broad models of care have evolved: full centre care, where the child is seen every 6–8 weeks (including annual review) at the specialist CF centre; local clinic based care, where the child is seen regularly by the local clinic team and has only their annual review performed locally by the visiting consultant from the specialist CF centre; and hybrid care, where the child is usually reviewed at least three times a year (including annual review) by the specialist CF centre team including a physiotherapist at the local hospital, with additional reviews by the local clinic team. The model of care offered at the local hospital is unrelated to the distance from the CF centre, but in practice (and by chance) the hospitals offering hybrid care are those most geographically distant from the centre.

Height, weight, spirometry13 and use of chronic medication were recorded at each subject's annual assessment performed during 2008. For all children over 2 years of age, height, weight and body mass index (BMI) SD scores were calculated.14 For children over 7 years of age, forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) were recorded by either the specialist team or local team and the raw values expressed as the percentage predicted for age, sex and height.15 Chronic P aeruginosa infection was defined as having more than 50% of cough swab or sputum cultures positive during 2008.16

Statistical analysis was performed using SPSS. Continuous data were assessed by the Kolmogorov–Smirnov test for goodness of fit, and either analysis of variance or the Kruskal–Wallis test was used to compare differences in continuous variables between the groups. The χ2 test was used for categorical data. Data are presented as means and SD or medians and IQR.


A total of 199 children and young people (mean age 9.4 years, range 0.2–18.9 years) in South and Mid Wales were receiving care from the CF centre in Cardiff, of whom 77 received full care, 102 shared care and 20 hybrid care. Although those receiving hybrid care were slightly older, and those receiving full care were less likely to carry one or more copy of the dF508 mutation and to be pancreatic sufficient, there were no major differences in mean ages, sex distribution or distribution of dF508 mutations (table 1).

Table 1

Baseline characteristics and outcomes for patients receiving full, shared and hybrid care

The continuous variables had a normal distribution with the exception of the raw FEV1 and BMI values. There was no statistically significant difference in weight SDS, height SDS or BMI SDS between groups. There was no statistically significant difference in the mean number of airway cultures collected over the year, or in the rate of chronic P aeruginosa infection. There were no statistically significant differences in the use of long term treatments, including macrolide therapy, overnight feeds, nebulised antibiotics or nebulised DNAse.

Forty seven of 77 full care patients were aged 7 years or older and able to perform spirometry compared to 61 of 102 shared care patients, and 15 of 20 hybrid care patients (table 2). Of those aged 7 years or older, full care patients were slightly older and taller than shared care or hybrid care patients and will influence the raw lung function data. There was a statistically significant difference in mean FEV1 per cent predicted between the models of care, but no difference in mean FVC per cent predicted.

Table 2

Pulmonary function in patients aged 7 years or older receiving full, shared and hybrid care


Compared to children and young people who received full CF centre care or hybrid care, those who received shared care had statistically significantly lower measures of pulmonary function. The rate of chronic P aeruginosa infection was low in all three groups, and although the numbers are small, the rate was lowest in those receiving full care. There was no major difference in baseline characteristics, nutritional status or use of long term therapies between groups, suggesting that the differences observed are due to the model of care they received. In the longer term, this difference in pulmonary function may correspond to differences in potential life expectancy.

Although older studies have reported better outcomes for children receiving centre based specialist CF care, they often reflect upon care delivered 20–40 years ago.4,,9 In contrast, more recent reports suggest that outcomes for children receiving shared care are comparable to those receiving centre based care.10 11 van Koolwijk compared the outcomes of 105 children receiving either centralised care, shared care or localised care, and reported no statistically significant differences between the groups.10 Similarly, Thomas compared outcomes for 273 children receiving either centre based care or three models of outreach care, and reported similar outcomes apart from a higher likelihood of hospital admission for those receiving centre based care.11

It is difficult to make direct comparisons between our model of care and those of van Koolwijk and Thomas, but compared to those reported by van Koolwijk, our children are slightly younger, lighter and shorter with higher BMIs, while Thomas only reported SDS values compared to the Australasian CF population. Compared to UK national reference values, our children are lighter and shorter than the mean, while their BMIs are slightly above the mean. In contrast, the rate of P aeruginosa infection was higher in both van Koolwijk and Thomas' reports. Our rate of chronic P aeruginosa infection is markedly lower than in most previous reports17 despite regular sampling, and compares favourably with the lowest reported rates.18

It is perhaps unsurprising that the nutritional outcomes of the full care and shared care patients were so similar. It is arguable that patients with CF are constitutionally small compared to normal, and so height (and consequently weight) is a poor outcome measure. Infants with CF are smaller at birth,19,,25 and even with newborn screening and modern medical management their height SDS (and consequently their weight SDS) are below the mean. In contrast, the median BMI26 27 is often in the normal range during early childhood, and our mean BMI SDS at 0.08 has a normal distribution with an SD of 1.0. Although poor nutritional status is associated with decreased pulmonary function28 and poorer long term survival,29 30 its effect is more pronounced in older patients, and thus nutritional status may be an insensitive measure in our population.

A potential concern with centre based care is the possibility of earlier and higher rates of chronic P aeruginosa infection,8 although it is possible that this reflects better microbiological surveillance in CF centres, and that sicker patients tend to gravitate to CF centres. We undertake regular review of patients and regular airway cultures and the three groups of patients had very similar mean numbers of airway cultures per year, with all but two patients having four or more airway cultures (two patients had three airway cultures). In 2008 we did not segregate patients according to whether they isolated P aeruginosa, although we did segregate those with Burkholderia cepacia, methicillin-resistant Staphylococcus aureus or siblings or children of adults with CF. Contrary to previous reports, in our population the rate of chronic P aeruginosa infection was lower in the full care patients, and indeed might have been lower but for the transfer of a patient with chronic P aeruginosa infection from shared to full care in the preceding year.

Our policy is to treat first isolates of P aeruginosa with 3 weeks of high dose oral ciprofloxacin and 3–6 months of nebulised colistin. For second isolates, we treat patients with longer courses of ciprofloxacin and colistin, and once cultures are clear for 6 months, we often continue with long term once daily nebulised colistin. We do not perform regular bronchoalveolar lavage or P aeruginosa serology on our patients, and so we might underestimate the true prevalence of chronic P aeruginosa infection, and it is arguable that we merely suppress P aeruginosa infection rather than truly eradicate it. Nevertheless, current definitions of chronic P aeruginosa infection in CF are based on culture results rather than serology.16

It is difficult to ascertain why there is such a difference in pulmonary function between those who receive full or hybrid care and those who receive shared care. We have had the benefit of newborn screening since 1996,12 and advocate early and aggressive treatment early in life.31 This model of care is well established, the majority of patients having had the same model of care for 12 years. Although we have joint meetings, and have agreed management standards for the first 2 years of life, the majority of care is designed to comply with the UK CF Trust standards for clinical care.2 Although the numbers are small, those subjects who received a hybrid model of care with regular review from the centre had pulmonary function outcomes very similar to those who received full care as opposed to those who were only seen annually by the CF team. In particular, the hybrid model offers early and regular review of patients by the CF centre MDT in the first years of life, with regular communication between the centre MDT and local clinic team, while the shared care patients are often only reviewed annually from their first birthday onwards.

It is interesting to speculate on possible explanations for the observed differences. We review our full care patients very frequently (at least monthly in the first 2 years of life), while hybrid care patients are reviewed by the specialist CF team at least three times by their first birthday. In contrast, shared care patients may have their first specialist review at their first birthday. We do not have data on interventions in the first few years of life, but it is our perception that we are more aggressive with rescue oral and intravenous antibiotics. Thus we believe that the differences observed between models of care may reflect the frequency of specialist CF team review in the first few years of life.

Frequency of review by a specialist CF team is associated with improved outcome measures,32 but distance from the CF centre could be perceived to be a barrier to optimal care. Our findings suggest that frequency of review by the CF centre MDT is more important than distance from the CF centre, and that barriers to regular review, including distance from the CF centre, can be overcome with the anticipation that this will improve the quality of care for children with cystic fibrosis.



  • South and Mid Wales Paediatric Cystic Fibrosis Network: Jane Clarke, Kate Creese, Ian Hodges, Vishwa Narayan, Humphrey Okuonghae, Gwyneth Owen and Prem Pitchaikani.

  • Competing interest None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.