Re-organisation of oesophago-gastric cancer care in England: progress and remaining challenges

Oesophageal and gastric cancer together represent the third most common cause of cancer death in the United Kingdom [1‐3]. The prognosis is often poor with overall 5-year survival rates in the United Kingdom being approximately 8% for oesophageal cancer and 14% for gastric cancer [1, 2]. The majority of patients present with advanced disease and many also have significant comorbidity and a poor level of overall fitness. As a result, the diseases' management, both curative and palliative, is complex and involves many different professional groups including surgeons, gastroenterologists, oncologists, palliative care physicians, radiologists, nurse specialists and dieticians [4‐7].

During the 1990s, it was recognised that oesophago-gastric cancer services in England were fragmented and poorly organised. To overcome these weaknesses, the National Health Services (NHS) Cancer Plan [8] and the Improving Outcomes Guidance in Upper Gastro-Intestinal Cancer [5] set out a reform strategy and a number of specific recommendations. Firstly, all hospitals caring for cancer patients should be integrated into regional cancer networks. Within each network, curative services should be centralised into specialist cancer centres and a system should be established to coordinate care between these centres and other network hospitals (designated local units). Secondly, it was recommended that clinicians within each hospital should work together as a multi-disciplinary team (MDT).

The implementation of this reform strategy has seen oesophago-gastric cancer services undergo an extensive reorganisation. In this paper, we describe the results of a survey of English cancer networks and NHS acute trusts carried out in September 2007 that aimed to investigate the extent to which the reorganisation of oesophago-gastric cancer services had been completed, and to examine variation in the provision of diagnostic, therapeutic and support services. We also discuss these findings with the results of the Cancer Peer Review Programme carried out by the UK Department of Health between 2004 and 2007. The survey was performed as part of the National Oesophago-Gastric Cancer Audit, details of which can be found in its first annual report [9].

Data on the organisation of oesophago-gastric cancer services were collected using two different questionnaires. The first contained questions that focused on the organisation of services within the cancer networks [see Additional file 1]. Networks are responsible for developing and planning all aspects of NHS cancer services within their region, including the process of centralisation. This requires coordinating the allocation of resources and establishing referral pathways between NHS acute trusts as well as other providers. The second questionnaire contained questions about the individual characteristics of services within the individual NHS trusts, focussing on the availability of key health professionals and the delivery of specific aspects of care [see Additional file 2]. Questions in both questionnaires were based on the recommendations in UK guidelines on the organisation of oesophago-gastric cancer services [5, 6, 10, 11] (Table 1) and covered issues previously identified as essential for high-quality care [9].

A database of English cancer networks and NHS acute trusts was prepared by combining information from various sources, including the Cancer Services Collaborative Improvement Partnership and the National Clinical Audit Support Program. Each network and NHS trust was then contacted to confirm the name and position of the lead clinician for oesophago-gastric cancer. The questionnaires were sent to these lead clinicians in September 2007 and non-responders were followed up by email and telephone.

Responses were received from all 30 cancer networks and from 38 of the 44 specialist cancer centres (84%) and from 88 of the 112 local units (79%). The structure of the networks are summarised in Table 2.

Since the publication of the Improving Outcomes Guidance [5], oesophago-gastric cancer services in England have made considerable progress in developing a centralised, specialist service. We found that, in September 2007, all NHS acute trusts were organised into cancer networks and patients in all networks had access to the core staging investigations as well as to a range of palliative treatments. Coinciding with this reorganisation, the overall survival of patients with oesophago-gastric cancer has improved, with 1-year survival increasing from 30% to 37% between 1998 and 2005 [9]. However, we found that some components of this reorganisation were not yet complete. The centralisation of curative surgery was still in progress in 11 of the 30 cancer networks. Fourteen local units were still performing curative surgery and in seven of these, this was being performed by a single surgeon. Also, in only 16 networks were all patients discussed at a specialist MDT meeting. Moreover, palliative care teams were not fully represented in about a quarter of the multidisciplinary teams and some local units did not have a clinical nurse specialist.

The response rate to our questionnaires was high which strengthens the representativeness of our findings. There are three limitations to the study. Firstly, this was a cross-sectional study so the results do not provide evidence of change over time. Secondly, respondents provided self-reported data and we cannot exclude "social desirability bias" given that the lead clinicians may have wanted to give favourable answers. However, the likely impact of this bias is small because most questions addressed facts rather than perceptions.

Thirdly, the study examined the facilities and policy of each network, not what actually happens to patients. For example, while the policy in each network recommended that all patients being considered for curative treatment should have a CT scan, the levels of compliance within networks were unknown. Routine activity data could be used to describe progress in centralisation, but more detailed evaluation of processes such as disease staging requires prospective data collection. The National Oesophago-Gastric Cancer Audit will assess the process and outcomes of the care received by patients with this cancer in England and Wales [9].

The results of this "snapshot" survey confirm and complement those of the National Cancer Peer Review Programme that included visits to all cancer networks in England between November 2004 and March 2007 [12]. The Peer Review programme reported that overall compliance with the upper gastro-intestinal performance measures was 70%; compliance with a selected number of measures is shown in Table 6. Our results are more specific because the Peer Review process examined upper-gastrointestinal (upper GI) cancer services, which grouped oesophago-gastric cancer together with liver, pancreatic and biliary tract cancer.

Peer Review found that the performance of upper GI cancer services was often worse than for other cancer sites. For example, the overall compliance rate of 70% was below that of breast, colorectal, lung and gynaecological cancer (77%, 77%, 73% and 75%, respectively). Similarly, attendance at the MDT meetings was the worst for any cancer site except lung with the core MDT membership present at less than half of the meetings (46% for centres and 33% for units). This compares to breast (77%), colorectal (58%) and gynaecology (83% for centres and 45% for local units). As in our survey, workforce issues were flagged up, especially with respect to the provision of clinical nurse specialists and palliative care teams. Overall upper GI cancer had the worst provision of nurse specialists of any cancer site. Compared to other services involved in the care of all cancer sites (such as oncology, radiology, etc.), specialist palliative care had the lowest overall level of compliance on the performance measures.

The Peer Review Programme was undertaken over a three year period while networks were reorganising their services. The reorganisation was supposed to be complete by the end of 2007 and this may be one reason why Peer Review found variation in how well the networks complied with the IOG recommendations. Our survey builds on Peer Review because it took place shortly before the deadline for full reorganisation. It is concerning that we found significant variation between networks in their provision of services, and that many had still not completed their reorganisation. It is also concerning that a minority were unable to give a date for completion.

The variation in service provision that we observed has implications for patient care and highlights several areas where improvements should be made. Firstly, there is good evidence that postoperative mortality is lower [13, 14] and long-term survival is higher [15‐17] if curative surgery is performed by high-volume specialist surgical teams. Over a third of the cancer networks had not completed the centralisation of curative surgery and just under half of NHS trusts were not providing 24-hour specialist consultant surgical cover. Of particular concern are the seven local units were still providing curative surgery with just one surgeon. There is a high-risk of complications after curative surgery and, for high quality postoperative care, surgeons with specialist knowledge and experience need to be continuously available.

Secondly, the provision of palliative care services, specialist nurses and nutritional support was variable and in a proportion of NHS trusts was insufficient. Studies have shown that patients with oesophago-gastric cancer require intensive support due to their poor quality of life and high level of symptoms [18, 19]. There is also evidence that shows that patients who are malnourished have reduced survival, a worse quality of life [20, 21], lower rates of completion of oncology treatment [22, 23] and a higher rate of postoperative complications [22, 23]. Nutritional support has been shown to reduce the impact of malnutrition and improve many of these outcomes [24]. A quarter of specialist centres do not formally assess their patients' nutritional status (either by dietician assessment or by a validated screening tool) before starting treatment. Simple tests such as the Nutritional Risk Index can be rapidly carried out by any doctor or nurse in the outpatients department and if performed on every patient could have a meaningful impact on their care.

Thirdly, although overall access to invasive palliative techniques was good with rapid access to stents and argon ablation therapy being available in all networks, access to other modalities was poor. Patients had to wait longer than 2 weeks for palliative chemotherapy in three of the thirty networks and for palliative radiotherapy in five networks. Brachytherapy was only available in 16 networks despite the fact that, compared to endoluminal stents, research has shown it to have fewer side-effects and provide better palliation in patients with an expected survival of greater than three months [25].

Finally, the Improving Outcomes Guidance recommended that all patients should be discussed by the specialist MDT [5]. The survey found that this occurs in 16 networks (53%). However, each network was also recommended to develop clinical guidelines on the selection of patients for referral to the specialist team [5]. The Peer Review Programme observed that these guidelines were in place in 71% of networks [11]. Consequently, not all patients are necessarily receiving the benefit of specialist experience. Processes need to be established so that each patient, irrespective of stage of disease or comorbidity, is evaluated by the network specialist teams, either by formal referral to the specialist MDT meeting or by discussion with a member of the specialist MDT.

Although considerable progress has been made in transforming oesophago-gastric cancer care into a specialist, multi-disciplinary service, further improvements are necessary. Action is required to rectify the deficits in the provision of surgery, palliative care, clinical nurse specialists and nutritional support. Some of this depends on the further allocation of scarce resources but some aspects, such as ensuring all patients are discussed with a member of the specialist multidisciplinary team and are screened for malnutrition, can be improved by relatively simple changes in practice.