The handle http://hdl.handle.net/1887/55893 holds various files of this Leiden University dissertation.
Author: Busweiler, L.A.D.
Title: Evaluating quality of care and setting future goals in oesophagogastric cancer treatment
Issue Date: 2017-11-29
Evaluating Quality of Care and Setting Future Goals
in Oesophagogastric Cancer Treatment
Care and Setting Future Goals in Oesophagogastric Cancer TreatmentLinde A.d. BusweiLer
in Oesophagogastric Cancer Treatment
Linde A.D. Busweiler
Copyright: Linde. A.D. Busweiler, 2017 © ISBN: 978-94-6361-028-5
Cover design: Neil Webb (published in de Volkskrant)
Layout and printing: Optima Grafische Communicatie, Rotterdam, the Netherlands (www.ogc.nl)
in Oesophagogastric Cancer Treatment
PROEFSCHRIFT
ter verkrijging van
de graad van Doctor aan de Universiteit Leiden, op gezag van Rector Magnificus prof.mr. C.J.J.M. Stolker,
volgens besluit van het College voor Promoties te verdedigen op woensdag 29 november 2017
klokke 15.00 uur
door
Linde Anna Drieske Busweiler Geboren te Hilversum in 1985
Copromotores
Dr. J.W. van Sandick (NKI-AVL) Dr. M.W.J.M. Wouters (NKI-AVL)
Leden promotiecommissie
Prof. dr. M.I. van Berge Henegouwen (AMC) Prof. dr. C. Rosman (Radboud MC)
Dr. M. Slingerland
Prof. dr. E.W. Steyerberg (LUMC, Erasmus MC)
Chapter 1 Introduction and outline of this thesis 7
Chapter 2 Early outcomes from the Dutch Upper Gastrointestinal Cancer Audit
19 British Journal of Surgery 2016
Chapter 3 Alternative methods displaying variation in hospital performance in oesophageal cancer surgery using funnel plots
41
Submitted
Chapter 4 International benchmarking in oesophageal and gastric cancer surgery using Swedish and Dutch quality registries
63
Submitted
Chapter 5 Textbook outcome as a composite measure in oesophagogastric cancer surgery
89 British Journal of Surgery 2017
Chapter 6 Failure-to-Rescue in patients undergoing surgery for oesophageal or gastric cancer
109 European Journal of Surgical Oncology 2017
Chapter 7 The influence of a composite hospital volume on outcomes for gastric cancer surgery: a Dutch population-based study
129
Journal of Surgical Oncology 2017
Chapter 9 Dutch Summary 167
Appendices Dankwoord 181
Curriculum Vitae 185
List of Publications 187
Ch ap te r
1
Introduction and outline of this thesis
INTRODuCTION AND OuTLINE OF THIS THESIS
Cancer care is increasingly recognized as a major worldwide challenge given its global, financial, social, and health implications1. The prevalence of cancer, primarily a disease of aging, is increasing. Together with rapid innovations in technology and treatments strategies, this puts a heavy burden on current health care. Moreover, the need to satisfy patients’ ex- pectations and the use of pay-for-performance models for the reimburse- ment of health care providers require a critical review of current practice.
This demand for performance has become a major issue for today’s healthcare system and has spurred the development of many national and international quality improvement initiatives2,3.
Evaluating quality in cancer care
In 2007, the Signalling Committee Cancer of the Dutch Cancer Society evaluated quality of cancer care in the Netherlands4. In their report ‘Qual- ity of Cancer Care’ the committee concluded that the overall quality of care for cancer patients in the Netherlands was high, though reducing variation between hospitals could lead to further improvement. Intro- ducing quality standards for organisation of care on the one hand, and monitoring and benchmarking patient outcomes between providers on the other hand, were suggested as strategies to reduce variation and to improve quality of care.
Where benchmarking is common in most branches, this is not the case for health care. Benchmarking made its first appearance in the healthcare system in the mid 1990’s and involves a sustained effort to measure outcomes, compare these outcomes against those of other organizations, learn how those outcomes were achieved, and apply the lessons learned in order to improve3. To do so, useful, reliable and up-to-date information is essential.
Following international examples of successful clinical audits, in 2009 the Dutch Surgical Colorectal Audit (DSCA) was initiated5. With this audit colorectal cancer surgeons were among the first to embrace a nationwide clinical registry and implement clinical benchmarking in the Netherlands.
Shortly after, a second national audit on the quality of surgery for upper gastrointestinal (GI) cancer was initiated6.
Cancer of the upper GI tract
Cancer of the upper GI tract includes both oesophageal cancer and gastric cancer. In the Netherlands, the incidence of oesophageal cancer increased during the last two decades7, whilst the incidence of gastric cancer de- creased8. Approximately 2,200 patients are diagnosed with oesophageal cancer, and 1,200 patients are diagnosed with gastric cancer, each year.
Despite significant improvements in diagnostic modalities and treatment, prognosis remains dismal with 5-year survival rates around 20 per cent7,8.
Multimodality treatment
In Western countries, surgery combined with multimodality therapy has become standard of care for patients with advanced oesophagogastric cancer. In several studies performed in Europe and the United States, the benefits of multimodal therapy on both disease free and overall survival have been shown9-13. However, international consensus on the best multimodal approach has not yet been reached. According to Dutch treatment guidelines, all patients with potentially curable oesophageal or oesophagogastric-junction cancer (above T1) should be treated with preoperative chemoradiotherapy10,14. For patients with non-metastasized resectable gastric cancer (excluding stage I) perioperative chemotherapy is recommended9,15.
High-risk surgery and centralization
Although the treatment of both oesophageal cancer and gastric cancer requires a multidisciplinary approach, surgical resection remains the cornerstone of curative treatment. Oesophagectomies and gastrectomies for cancer are complex surgical procedures with considerable morbidity and mortality16-20. Next to technical skills needed to perform the opera- tion, careful patient selection with accurate staging and risk assessment are essential21. Moreover, early recognition and timely management of serious complications once they occur, are needed to prevent the patient from dying postoperatively. This requires the presence of an experienced surgical team working in the context of a well-structured multidisciplinary environment with excellent hospital facilities, appropriate consultative and critical care staff, experienced nursing staff and perioperative clinical pathways22-25. To increase the experience and expertise of surgical teams, minimum volume standards were introduced by the professional society, the Association of Surgeons in the Netherlands (ASN). For oesophageal cancer surgery, a minimum volume standard was set at 20 resections per hospital per year in 2011. For gastric cancer surgery, a minimum volume of ten resections per hospital per year in 2012, and 20 resections per hospital per year as of 2013, was required.
The Dutch upper GI Cancer Audit
After the initiation of the DSCA in 2009, the Dutch Institute for Clinical Auditing (DICA) was founded in 2011 with the objective to facilitate and organize the initiation of nationwide audits in a uniform format. In the same year, the Dutch Upper GI Cancer Audit (DUCA) group began a nationwide registration of all patients with oesophageal or gastric can- cer who underwent surgery with the intent of a surgical resection with or without preoperative therapy. The main purpose of the DUCA was to improve the quality of care for these patients by providing (surgical) teams with reliable, benchmarked information on process and (case mix-
adjusted) outcome parameters regarding their patients. The initiation and implementation of the DUCA, together with an overview of the first results, are described in Chapter 26.
Hospital comparisons
Reducing hospital variation was recommended by the Signalling Com- mittee Cancer for further quality improvement. In order to display such variation in hospital performance, a widely used graphical aid is the funnel plot26. This plot shows the outcome of interest (vertical axis) per hospital using 95.0 per cent or 99.8 per cent control limits. Those hos- pitals that fall outside these control limits can be identified as potential outliers and could be subject for an in-depth investigation. Identifying such outliers becomes a challenge when evaluating high complex and low volume surgery, like resections for oesophageal and gastric cancer monitored in the DUCA. Minimum volume standards are set, but annual volumes vary considerably between hospitals6. In Chapter 3, alternative statistical approaches using funnel plots that can be used for displaying variation in hospital performance within the context of clinical auditing are described27.
International benchmarking
In several European countries, registries for collecting detailed clinical information about patients treated for oesophagogastric cancer have been started. For instance, the Swedish National Register for Oesophageal and Gastric Cancer (NREV) was launched in 2006 and the British National Oesophago-Gastric Cancer Audit (NOGCA) started collecting detailed information about this group of patients in 201128,29. These clinical audit systems on oesophagogastric cancer care provide opportunities for benchmarking on a European level30,31. In a previous study, postoperative mortality for patients with gastric cancer in the Netherlands was high com- pared with that in other European countries32. In Chapter 4, a comparison
is made between the results from the DUCA and those from the NREV33. Are there any differences between patient and treatment characteristics and how do the surgical outcomes as registered in the DUCA compare to those in the Swedish registry?
Textbook outcome
In previous years, quality measurements for outcome of complex surgical procedures have focussed on readily available and easily understand- able parameters, such as procedural volume, in-hospital mortality and duration of postoperative hospital stay34-37. This provides information on a single parameter and the multidimensional aspect of the whole surgical process is not valued. As proposed by the Institute of Medicine, health care should be safe, effective, patient-centred, timely, efficient and equitable38. For patients undergoing potentially curative oesophago- gastric cancer surgery, the surgical process can be considered safe if no adverse outcomes (mortality and morbidity) have occurred, and effective if complete tumour removal and adequate lymphadenectomy have been achieved. This information is gathered in the composite measure textbook outcome (Chapter 5)39.
Failure-to-rescue
Failure-to-rescue (FTR) is a relatively new quality parameter and was first described in patients undergoing coronary artery bypass graft (CABG) sur- gery by Silber et al40. FTR is defined as the mortality rate among patients with post interventional complications. In general, timely recognition and effective management of postoperative complications may have a great positive effect on prevention of postoperative mortality. This might be even more true for postoperative management of oesophagogastric can- cer patients, as it concerns a relatively frail group of patients, after major surgery, who are prone to quick clinical deterioration once a complication has occurred. In Chapter 6 trends in postoperative outcomes are evalu-
ated and a comparison is made between patients with oesophageal and those with gastric cancer41. Can differences in FTR rates after oesophageal cancer resections versus those after gastric cancer resections be explained by differences in patient, tumour and treatment characteristics?
Composite hospital volume
In the Netherlands, both oesophageal and gastric cancer resections are performed by certified gastro-intestinal surgeons. One could infer a mutual benefit of combining these procedures in specialized centers. In Dutch treatment guidelines, it is described that gastric cancer surgery is preferrably centralized towards hospitals with broad experience in oesophageal cancer surgery. A dedicated multidisciplinary team with well-trained surgeons, operative teams familiar with these procedures, dedicated anaesthesiologists, a specialized Intensive Care Unit and an experienced nursing staff at the hospital ward might result in better out- comes. If placed under a broader scope, the quality of gastric cancer sur- gery might also benefit from shared experience with other high complex upper GI cancer resections such as pancreatectomies. Is such a composite hospital volume associated with improved postoperative outcomes for patients undergoing surgery for gastric cancer? (Chapter 7)42
REFERENCES
1. Lyerly HK, Abernethy AP, Stockler MR, et al. Need for global partnership in cancer care: perceptions of cancer care researchers attending the 2010 australia and Asia pacific clinical oncology research development workshop. J Oncol Pract 2011;7(5):
324-329.
2. Berwick DM. Era 3 for Medicine and Health Care. Jama 2016;315(13): 1329-1330.
3. Ettorchi-Tardy A, Levif M, Michel P. Benchmarking: a method for continuous qual- ity improvement in health. Healthc Policy 2012;7(4): e101-119.
4. Wouters MW, Jansen-Landheer ML, van de Velde CJ. The Quality of Cancer Care initiative in the Netherlands. Eur J Surg Oncol 2010;36 Suppl 1: S3-s13.
5. Van Leersum NJ, Snijders HS, Henneman D, et al. The Dutch surgical colorectal audit. Eur J Surg Oncol 2013;39(10): 1063-1070.
6. Busweiler LA, Wijnhoven BP, van Berge Henegouwen MI, et al. Early outcomes from the Dutch Upper Gastrointestinal Cancer Audit. Br J Surg 2016;103(13): 1855-1863.
7. Dikken JL, Lemmens VE, Wouters MW, et al. Increased incidence and survival for oesophageal cancer but not for gastric cardia cancer in the Netherlands. Eur J Cancer 2012;48(11): 1624-1632.
8. Dassen AE, Dikken JL, Bosscha K, et al. Gastric cancer: decreasing incidence but stable survival in the Netherlands. Acta Oncol 2014;53(1): 138-142.
9. Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006;355(1):
11-20.
10. van Hagen P, Hulshof MC, van Lanschot JJ, et al. Preoperative chemoradiotherapy for esophageal or junctional cancer. N Engl J Med 2012;366(22): 2074-2084.
11. Ychou M, Boige V, Pignon JP, et al. Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin Oncol 2011;29(13): 1715-1721.
12. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastro- esophageal junction. N Engl J Med 2001;345(10): 725-730.
13. Sjoquist KM, Burmeister BH, Smithers BM, et al. Survival after neoadjuvant che- motherapy or chemoradiotherapy for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol 2011;12(7): 681-692.
14. Oncoline. Guideline Treatment Gastric Carcinoma. http://www.oncoline.nl/
maagcarcinoom [accessed 21 June 2017].
15. Oncoline. Guideline Treatment Oesophageal Carcinoma. http://www.oncoline.nl/
oesofaguscarcinoom [accessed 21 June 2017].
16. Connors RC, Reuben BC, Neumayer LA, et al. Comparing outcomes after trans- thoracic and transhiatal esophagectomy: a 5-year prospective cohort of 17,395 patients. J Am Coll Surg 2007;205(6): 735-740.
17. Bailey SH, Bull DA, Harpole DH, et al. Outcomes after esophagectomy: a ten-year prospective cohort. Ann Thorac Surg 2003;75(1): 217-222.
18. Damhuis RA, Meurs CJ, Dijkhuis CM, et al. Hospital volume and post-operative mortality after resection for gastric cancer. Eur J Surg Oncol 2002;28(4): 401-405.
19. Ikeguchi M, Oka S, Gomyo Y, et al. Postoperative morbidity and mortality after gas- trectomy for gastric carcinoma. Hepatogastroenterology 2001;48(41): 1517-1520.
20. Lepage C, Sant M, Verdecchia A, et al. Operative mortality after gastric cancer resection and long-term survival differences across Europe. Br J Surg 2010;97(2):
235-239.
21. Grotenhuis BA, van Heijl M, Zehetner J, et al. Surgical management of submu- cosal esophageal cancer: extended or regional lymphadenectomy? Ann Surg 2010;252(5): 823-830.
22. Ghaferi AA, Dimick JB. Importance of teamwork, communication and culture on failure-to-rescue in the elderly. Br J Surg 2016;103(2): e47-51.
23. Sheetz KH, Dimick JB, Ghaferi AA. Impact of Hospital Characteristics on Failure to Rescue Following Major Surgery. Ann Surg 2016 Apr;263(4):692-7.
24. Ghaferi AA, Birkmeyer JD, Dimick JB. Hospital volume and failure to rescue with high-risk surgery. Med Care 2011;49(12): 1076-1081.
25. Nguyen NT, Paya M, Stevens CM, et al. The relationship between hospital vol- ume and outcome in bariatric surgery at academic medical centers. Ann Surg 2004;240(4): 586-593; discussion 593-584.
26. Spiegelhalter DJ. Funnel plots for comparing institutional performance. Stat Med 2005;24(8): 1185-1202.
27. Busweiler LA, Schouwenburg MG, Amodio S, et al. Alternative methods displaying variation in hospital performance in oesophageal cancer surgery using funnel plots. Submitted.
28. Health and Social Care Information Centre. National Oesophago-Gastric Cancer Audit (NOGCA). http://www.hscic.gov.uk/org [accessed 21 October 2015].
29. Linder G, Lindblad M, Djerf P, et al. Validation of data quality in the Swedish National Register for Oesophageal and Gastric Cancer. Br J Surg 2016;103(10):
1326-1335.
30. de Steur WO, Henneman D, Allum WH, et al. Common data items in seven Euro- pean oesophagogastric cancer surgery registries: towards a European upper GI cancer audit (EURECCA Upper GI). Eur J Surg Oncol 2014;40(3): 325-329.
31. Messager M, de Steur WO, van Sandick JW, et al. Variations among 5 European countries for curative treatment of resectable oesophageal and gastric cancer:
A survey from the EURECCA Upper GI Group (EUropean REgistration of Cancer CAre). Eur J Surg Oncol 2016;42(1): 116-122.
32. Dikken JL, van Sandick JW, Allum WH, et al. Differences in outcomes of oesopha- geal and gastric cancer surgery across Europe. Br J Surg 2013;100(1): 83-94.
33. Busweiler LA, Jeremiasen M, Wijnhoven BP, et al. International benchmarking in oesophageal and gastric cancer surgery using Swedish and Dutch quality regis- tries. Submitted.
34. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortal- ity in the United States. N Engl J Med 2002;346(15): 1128-1137.
35. Chowdhury MM, Dagash H, Pierro A. A systematic review of the impact of volume of surgery and specialization on patient outcome. Br J Surg 2007;94(2): 145-161.
36. Dikken JL, Dassen AE, Lemmens VE, et al. Effect of hospital volume on postop- erative mortality and survival after oesophageal and gastric cancer surgery in the Netherlands between 1989 and 2009. Eur J Cancer 2012;48(7): 1004-1013.
37. Wouters MW, Gooiker GA, van Sandick JW, et al. The volume-outcome relation in the surgical treatment of esophageal cancer: a systematic review and meta- analysis. Cancer 2012;118(7): 1754-1763.
38. Institute of Medicine Committee on Quality of Health Care in America. Crossing the Quality Chasm: a New Health System for the 21st Century. National Academies Press, National Academy of Sciences:Washington, 2001.
39. Busweiler LA, Schouwenburg MG van Berge Henegouwen MI, et al. Textbook outcome as a composite measure in oesophagogastric cancer surgery. Br J Surg 2017;104(6): 742-750.
40. Silber JH, Rosenbaum PR, Schwartz JS, et al. Evaluation of the complication rate as a measure of quality of care in coronary artery bypass graft surgery. JAMA 1995;274(4): 317-323.
41. Busweiler LA, Henneman D, Dikken JL, et al. Failure-to-rescue in patients undergo- ing surgery for esophageal or gastric cancer. Eur J Surg Oncol 2017 Oct;43(10):1962- 1969.
42. Busweiler LAD, Dikken JL, Henneman D, et al. The influence of a composite hos- pital volume on outcomes for gastric cancer surgery: A Dutch population-based study. J Surg Oncol. 2017 May;115(6):738-745.
Ch ap te r
2
Early outcomes from the Dutch Upper Gastrointestinal Cancer Audit
L.A.D. Busweiler1,2, B.P.L. Wijnhoven3, M.I. van Berge Henegouwen4, D. Henneman1,2, N.C.T.
van Grieken5, M.W.J.M. Wouters1,6, R. van Hillegersberg7 and J.W. van Sandick6, on behalf of the Dutch Upper Gastrointestinal Cancer Audit (DUCA) group.
1Dutch Institute for Clinical Auditing, Leiden
2Department of Surgery, Leiden University Medical Centre, Leiden
3Department of Surgery, Erasmus Medical Centre, Rotterdam
4Department of Surgery, Academic Medical Centre, Amsterdam
5Department of Pathology, VU University Medical Centre, Amsterdam
6Department of Surgical Oncology, Netherlands Cancer Institute / Antoni van Leeuwenhoek Hospital, Amsterdam
7Department of Surgery, University Medical Centre Utrecht, Utrecht
Br J Surg. 2016 Dec;103(13):1855-1863.
ABSTRACT
Background: In 2011, the Dutch Upper Gastrointestinal Cancer Audit (DUCA) group began nationwide registration of all patients undergoing surgery with the intention of resection for oesophageal or gastric cancer.
The aim of this study was to describe the initiation and implementation of this process along with an overview of the results.
Methods: The DUCA is part of the Dutch Institute for Clinical Audit- ing. The audit provides (surgical) teams with reliable, weekly updated, benchmarked information on process and (case mix-adjusted) outcome measures. To accomplish this, a web-based registration was designed, based on a set of predefined quality measures.
Results: Between 2011 and 2014, a total of 2786 patients with oesopha- geal cancer and 1887 with gastric cancer were registered. Case ascertain- ment approached 100 per cent for patients registered in 2013. The per- centage of patients with oesophageal cancer starting treatment within 5 weeks of diagnosis increased significantly over time from 32.5 per cent in 2011 to 41.0 per cent in 2014 (P<0.001). The percentage of patients with a minimum of 15 examined lymph nodes in the resected specimen also increased significantly for both oesophageal cancer (from 50.3 per cent in 2011 to 73.0 per cent in 2014; P<0.001) and gastric cancer (from 47.5 per cent in 2011 to 73.6 per cent in 2014; P<0.001). Postoperative mortal- ity remained stable (around 4.0 per cent) for patients with oesophageal cancer, and decreased for patients with gastric cancer (from 8.0 per cent in 2011 to 4.0 per cent in 2014; P=0.031).
Conclusion: Nationwide implementation of the DUCA has been success- ful. The results indicate a positive trend for various process and outcome measures.
INTRODuCTION
In the Netherlands, the incidence of oesophageal cancer has increased over the past two decades1. At the same time, the incidence of gastric cancer has decreased2. As a consequence, the annual number of oesopha- geal resections for cancer doubled whereas the annual number of gastric resections for cancer decreased3. Together with the introduction of a minimum volume standard (per year per hospital) for oesophageal cancer surgery in 2006, this led to centralization for oesophageal, but not gas- tric, cancer surgery4. Within the past 20 years, 5-year survival rates have doubled for patients with oesophageal cancer, but with no improvement for those with gastric cancer1,2. The postoperative mortality rate after gas- tric resection for cancer has remained high compared with that in other European countries5. In 2007, the Quality of Cancer Care taskforce of the Dutch Cancer Society evaluated quality of cancer care in the Netherlands6. This taskforce concluded that the overall quality of cancer care was high, but reducing variation between providers could lead to further improve- ment. Recommendations involved the introduction of multidisciplinary quality standards, and monitoring and benchmarking patient outcomes between providers (Supplemental table 1). The minimum volume stan- dard for oesophageal cancer surgery was set at ten resections per year per hospital in 2006, and 20 resections per year per hospital from 2011. For gastric cancer surgery, a minimum volume of ten resections per year per hospital in 2012, and 20 resections per year per hospital as of 2013, was required. In 2011, the Dutch Upper Gastrointestinal Cancer Audit (DUCA) group began nationwide registration of all patients undergoing surgery with the intention of resection for oesophageal or gastric cancer. The goal was to improve the quality of care by providing (surgical) teams with reli- able and benchmarked information on process and (case mix-adjusted) outcome parameters regarding their patients. This information was used to monitor national guideline adherence and patient outcomes.
The aim of this study was to describe the initiation and implementation of the DUCA, and provide an overview of the first results after 4 years of auditing.
METHODS
Organization, implementation and funding
The Dutch Institute for Clinical Auditing (DICA) was founded in 2011 with the objective to facilitate and organize the initiation of nationwide audits in a uniform format7. In the same year, the DUCA group began nationwide registration of all patients undergoing surgery with the intention of resec- tion for oesophageal or gastric cancer (Table 1). The DUCA was initiated by the former Dutch Oesophageal Cancer Group and the Dutch Gastric Cancer Group, nowadays fused into the Dutch Upper Gastrointestinal Cancer Group, and was mandated by the Dutch Association of Surgical Oncologists and the Association of Gastrointestinal Surgeons of the Neth- erlands. The audit is a surgical registration, initially funded and executed via quality improvement grants donated by the Dutch Associated Health Insurance Companies, Zorgverzekeraars Nederland. Structural funding was later achieved by hospitals paying a subscription fee for participa- tion. These subscription costs were returned to the hospitals as they were enclosed in the reimbursement schemes for treating patients with oesophageal and gastric cancer. All participating hospitals were respon- sible for data registration and the costs of data registration themselves.
Nationwide coverage of the audit was stimulated via the Association of Surgeons of the Netherlands (ASN) and the Health Care Inspectorate, as participation in the DUCA was defined as a mandatory quality standard as of 2012. Participation in the DUCA was also incorporated as a multidisci- plinary quality standard defined by the Dutch Federation for Oncological Societies (Supplemental table 1). As healthcare in the Netherlands is based on a public healthcare system, no private institutions are involved
in oesophagogastric cancer surgery. Both the ASN and the Health Care Inspectorate ensure that all hospitals participate in this audit.
Table 1. In- and exclusion criteria DUCA.
Inclusion criteria: All patients who are operated with the intent of a resection, with or without preoperative treatment, for:
Primary tumours of the oesophagus, stomach or gastro oesophageal junction Recurrent tumours of the oesophagus, stomach or gastro oesophageal junction
Prophylactic resection of the oesophagus or stomach (because of high grade dysplasia or CDH1 mutation, respectively)
Exclusion criteria: Non-epithelial tumours (e.g. gastrointestinal stromal cell tumours) No cancer resection planned
Quality measures and data set
A directional board and a scientific committee were formed comprising medical professionals in upper gastrointestinal cancer care (surgeons, medical oncologists, gastroenterologists, pathologists and radiation oncologists). Dutch evidence-based guidelines, and information from the Swedish national quality registry and the British National Oesophago- Gastric Cancer Audit (NOGCA) were used to formulate a set of quality measures to compose a data set8–11.
The items in the registry were grouped in three categories. The first category included items necessary to enable sound and reliable data comparisons between hospitals (case mix variables). This concerned information regard- ing patient and tumour characteristics (age, BMI, co-morbidities, clinical stage of disease). The second category included items regarding processes of care (different modalities used during the diagnostic process, time interval between diagnosis and start of treatment, evaluation of patients’
treatment plans in a multidisciplinary team (MDT) meeting). Clinical and pathological outcomes of surgery were registered in the third category. The content of the data set has been evaluated on a yearly basis and items were removed or replaced by others as deemed necessary.
Data entry, storage and privacy
A generic, internet-based program was used to enable data entry to a secure online environment. Hospitals were free to decide who carried out the data registration (for example data managers, nurse practitioners or medical specialists). In all participating hospitals the final responsibility for data entry remained with the surgeon. Detailed descriptions of definitions used in the registry are provided via an information button in the online registry program. DICA has an advisory board to stimulate uniform data registration, and offers support via a help desk and a website (frequently asked questions) regarding data entry and for feedback regarding the registry. This information is also used for yearly evaluation of the registry.
The help desk also provides yearly updated case report forms and data dictionaries. Registered data can be updated whenever necessary.
A third trusted party (TTP) is responsible for processing the data, and in- formation regarding patient identification is encrypted directly after data entry. The TTP uses multiple secure and certified data centres throughout the Netherlands for data storage, and adheres to strict security standards required by the medical field and the Dutch law. Anonymized data are provided for quality assurance and research purposes. Participating hos- pitals maintain ownership of their own data.
Feedback and auditing
A secure online website is used for hospital feedback. The information presented via this website is updated once a week, and all participating hospitals can use it to monitor their own results in relation to the national average. Besides information about patient population, results for all process and (case mix-adjusted) outcome measures are presented using funnel plots with 95 per cent confidence limits that vary in relation to hospital volume12. The funnel plots provide information regarding specific
process or outcome measures for individual hospitals in relation to the national average and in relation to results of other anonymized hospitals.
Both the Health Care Inspectorate and the ASN use this information on a yearly basis to monitor the quality of surgical care in all participating hos- pitals, and intervene when results show negative outliers (below or above the 95.0 per cent confidence interval, depending on the quality measure).
A special audit taskforce, appointed by the ASN, visits underperforming hospitals and serves as an advisory board.
Verification and validation
For all DICA registries, validation of data entry takes place at two levels.
During the registration process, various data fields report a warning or error whenever data are missing or if data seem implausible compared with previously registered information. Each hospital has access to an electronic report, summarizing missing variables and those that are po- tentially erroneous.
To improve further the reliability of the registered data, an independent team of data managers evaluated the total number of patients registered in 2013, and an in-depth quality investigation was performed on a ran- dom data sample.
Completeness of the total number of patients included in the DUCA was also evaluated in a comparison with an external data registration (Nether- lands Cancer Registry, NCR), which contains data on all newly diagnosed malignancies in the Netherlands.
Analysis of registered data
A minimum number of items per patient was required in order to consider a patient eligible for analysis. These were: information on tumour location,
date of birth, date of surgery, intent of surgery as defined at the end of the operation (potentially curative resection, palliative resection or no resec- tion) and vital status 30 days after surgery and/or at time of discharge.
Patient, tumour and treatment characteristics were described using frequency tables. The number of participating hospitals was analysed for different volume categories. National results for various quality measures are shown for different years of registration, and results are compared using χ2 tests for trend. For this study, no ethical approval or informed consent was required under Dutch law.
Reporting
Both the directional board and the scientific committee of the DUCA were responsible for presenting results and defining targets for quality im- provement. Annual overall results of the audit were published, facilitated by the scientific bureau of DICA4,13. DICA has a methodological board con- sisting of statisticians and epidemiologists. This advisory board supervises methodology used by DICA registries7. As of 2016, clinicians participating in the DUCA can use registered data for research purposes by submitting a research application. Two members of the scientific committee of the DUCA and the scientific bureau of DICA review each application and monitor statistical methods described in these applications.
Transparency
One of the objectives of the DUCA was to improve transparency regarding the quality of (surgical) care for patients with oesophageal or gastric cancer.
To accomplish this, individual hospital results for all quality measures were made available for third parties after authorization by the hospital board.
For instance, healthcare insurance companies are able use this information in their negotiations with hospitals, and patients receive this informa- tion via patient advocacy groups. Transparency is achieved according to
a stepwise model, as defined by the ASN: participating in the audit (first year), transparency of process measures (second year) and transparency of outcome measures (third year).
RESuLTS
Between January 2011 and December 2014, a total of 2786 patients with oesophageal cancer and 1887 patients with gastric cancer were regis- tered. Case ascertainment for the DUCA in 2013 was estimated at 97.8 per cent of all primary oesophageal cancer resections and 96.2 per cent of all primary gastric cancer resections, as registered in the NCR. Patient and tumour characteristics are shown in Table 2.
Table 2. Patient and tumour characteristics for patients with oesophageal cancer or gastric cancer included in the Dutch Upper Gastrointestinal Cancer Audit (2011–2014)
Oesophageal cancer (n = 2786) Gastric cancer (n = 1887) Age (years)
≤ 70 2009 (72·1) 971 (51·5)
> 70 765 (27·5) 913 (48·4)
Unknown 12 (0·4) 3 (0·2)
Sex ratio (M : F) 2156 : 630 1179 : 708
ASA fitness grade
I–II 2109 (75·7) 1293 (68·5)
≥ III 643 (23·1) 571 (30·3)
Unknown 34 (1·2) 23 (1·2)
BMI (kg/m2)
< 20 180 (6·5) 169 (9·0)
20–24 1061 (38·1) 759 (40·2)
25–29 1047 (37·6) 634 (33·6)
≥ 30 452 (16·2) 240 (12·7)
Unknown 46 (1·7) 85 (4·5)
Table 2. (continued)
Oesophageal cancer (n = 2786) Gastric cancer (n = 1887) Charlson Co-Morbidity Index score
0 1399 (50·2) 860 (45·6)
1 706 (25·3) 426 (22·6)
≥ 2 681 (24·4) 601 (31·8)
Clinical tumour stage*
0 (including T0 N0–2 M0) 12 (0·4) 29 (1·5)
I 380 (13·6) 338 (17·9)
II 661 (23·7) 601 (31·8)
III 1422 (51·0) 162 (8·6)
IV 21 (0·8) 53 (2·8)
Unknown 290 (10·4) 704 (37·3)
Histological type
Adenocarcinoma 2186 (78·5) 1765 (93·5)
Squamous cell carcinoma 523 (18·8) 2 (0·1)
Other/unknown 77 (2·8) 120 (6·0)
Tumour location†
Oesophagus, cervical part 5 (0·2) –
Oesophagus, upper third 27 (1·0) –
Oesophagus, middle third 313 (11·2) –
Oesophagus, lower third 1629 (58·5) –
Gastro-oesophageal junction 787 (28·2) –
Stomach, fundus – 162 (8·6)
Stomach, corpus – 560 (29·7)
Stomach, antrum – 728 (38·6)
Stomach, pylorus – 145 (7·7)
Stomach, overlapping lesions – 108 (5·7)
Gastric stump/anastomosis – 74 (3·9)
Unknown 25 (0·9) 110 (5·8)
Values in parentheses are percentages.
*TNM system (7th edition).
†ICD-0 codes.
Patients with gastric cancer were older, and had more co-morbidities as reflected by higher Charlson scores and a higher ASA grade, compared with patients with oesophageal cancer. The clinical tumour stage was not regis- tered completely in 37.3 per cent of the patients with gastric cancer. Patient and tumour characteristics did not change over the years (data not shown).
Treatment
Treatment characteristics are shown in Table 3 according to the year of registration. The percentage of minimally invasive procedures (at the start of the procedure, including conversions, and hybrid procedures for oesophageal cancer) increased over time for both patients with oesopha- geal cancer (from 31.0 per cent in 2011 to 64.8 per cent in 2014; P <0.001) and those with gastric cancer (from 4.1 per cent in 2011 to 43.4 per cent in 2014; P <0.001). The proportion of patients with oesophageal cancer who underwent resection via a transthoracic approach increased compared with the transhiatal approach (from 43.0 per cent in 2011 to 64.1 per cent in 2014; P <0.001), with more frequent intrathoracic anastomoses (from 11.2 per cent in 2011 to 35.9 per cent in 2014; P <0.001). No clinically relevant trends were seen with regard to perioperative treatment.
Quality measures
Improvements were seen in various predefined process and outcome measures (Table 4). The percentage of patients discussed in a pre- and post-treatment MDT meeting remained high during the study period (95.2–99⋅7 and 91.3–98.0 per cent respectively). The percentage of pa- tients with oesophageal cancer starting treatment within 5 weeks after diagnosis significantly increased over time (from 32.5 per cent in 2011 to 41.0 per cent in 2014; P <0.001).
The percentage of patients with a minimum of 15 examined lymph nodes in the resection specimen increased for both patients with oesophageal
Table 3. Treatment characteristics for patients with oesophageal cancer or gastric cancer included in the Dutch Upper Gastrointestinal Cancer Audit
Oesophageal cancer Gastric cancer
2011 2012 2013 2014 2011 2012 2013 2014
Preoperative therapy* n = 550 n = 716 n = 726 n = 781 n = 290 n = 387 n = 526 n = 562 None 40 (7·3) 76 (10·6) 77 (10·6) 88 (11·3) 132 (45·5) 165 (42·6) 254 (48·3) 245 (43·6) Chemotherapy 63 (11·5) 84 (11·7) 57 (7·9) 55 (7·0) 150 (51·7) 204 (52·7) 264 (50·2) 307 (54·6) Chemoradiotherapy 439 (79·8) 549 (76·7) 586 (80·7) 635 (81·3) 4 (1·4) 9 (2·3) 8 (1·5) 8 (1·4) Radiotherapy 2 (0·4) 0 (0) 4 (0·6) 1 (0·1) 1 (0·3) 0 (0) 0 (0) 0 (0) Unknown therapy 4 (0·7) 3 (0·4) 2 (0·3) 2 (0·3) 3 (1·0) 7 (1·8) 0 (0) 1 (0·2)
Unknown 2 (0·4) 4 (0·6) 0 (0) 0 (0) 0 (0) 2 (0·5) 0 (0) 1 (0·2)
Surgery n = 551 n = 725 n = 728 n = 782 n = 314 n = 420 n = 565 n = 588
Type of surgery Transhiatal oesophagectomy
270 (49·0) 272 (37·5) 249 (34·2) 225 (28·8) 4 (1·3) 3 (0·7) 6 (1·1) 2 (0·3)
Transthoracic oesophagectomy
237 (43·0) 408 (56·3) 428 (58·8) 501 (64·1) 1 (0·3) 5 (1·2) 1 (0·2) 1 (0·2)
Total gastrectomy 7 (1·3) 9 (1·2) 18 (2·5) 21 (2·7) 109 (34·7) 147 (35·0) 207 (36·6) 223 (37·9) Partial gastrectomy 2 (0·4) 2 (0·3) 1 (0·1) 2 (0·3) 177 (56·4) 211 (50·2) 272 (48·1) 293 (49·8) No resection 30 (5·4) 29 (4·0) 31 (4·3) 31 (4·0) 20 (6·4) 50 (11·9) 75 (13·3) 59 (10·0) Other/unknown 5 (0·9) 5 (0·7) 1 (0·1) 2 (0·3) 3 (1·0) 4 (1·0) 4 (0·7) 10 (1·7) Approach
Open 378 (68·6) 403 (55·6) 366 (50·3) 275 (35·2) 301 (95·9) 384 (91·4) 416 (73·6) 332 (56·5) Minimally invasive† 171 (31·0) 320 (44·1) 362 (49·7) 507 (64·8) 13 (4·1) 32 (7·6) 149 (26·4) 255 (43·4)
Unknown 2 (0·4) 2 (0·3) 0 (0) 0 (0) 0 (0) 4 (1·0) 0 (0) 1 (0·2)
Location of anastomosis‡ n = 520 n = 662 n = 695 n = 752 n = 291 n = 370 n = 486 n = 528 No anastomosis 1 (0·2) 0 (0) 2 (0·3) 5 (0·7) 4 (1·4) 5 (1·4) 4 (0·8) 6 (1·1) Neck 445 (85·6) 532 (80·4) 461 (66·3) 459 (61·0) 2 (0·7) 4 (1·1) 5 (1·0) 3 (0·6) Intrathoracic 58 (11·2) 115 (17·4) 201 (28·9) 270 (35·9) 8 (2·7) 25 (6·8) 46 (9·5) 40 (7·6) Intra-abdominal 7 (1·3) 8 (1·2) 13 (1·9) 7 (0·9) 266 (91·4) 314 (84·9) 412 (84·8) 463 (87·7) Other/unknown 9 (1·7) 7 (1·1) 18 (2·6) 11 (1·5) 11 (3·8) 22 (5·9) 19 (3·9) 16 (3·0)
cancer (from 50.3 per cent in 2011 to 73.0 per cent in 2014; P <0.001) and those with gastric cancer (from 47.5 per cent in 2011 to 73.6 per cent in 2014; P <0.001). A higher percentage of tumour-negative resection mar- gins was found for both patients with oesophageal cancer (from 92.0 per cent in 2011 to 95.0 per cent in 2014; P =0.023) and those with gastric cancer (from 82.8 per cent in 2011 to 87.9 per cent in 2014; P =0.157).
For patients with oesophageal cancer, in-hospital/30-day mortality rates remained stable at around 4.0 per cent, whereas the rate decreased sig- nificantly from 8.0 per cent in 2011 to 4.0 per cent in 2014 among patients with gastric cancer (P =0.031).
Hospitals
In the first year of registration, a total of 39 hospitals participated in the DUCA. Since the start of registration, the number of hospitals performing oesophageal cancer surgery and/or gastric cancer surgery decreased, while the annual caseload per hospital increased over time (Figure 1).
Table 3. (continued)
Oesophageal cancer Gastric cancer
2011 2012 2013 2014 2011 2012 2013 2014
Postoperative therapy§ n = 520 n = 656 n = 691 n = 749 n = 265 n = 337 n = 463 n = 502 None 490 (94·2) 614 (93·6) 653 (94·5) 713 (95·2) 165 (62·3) 201 (59·6) 299 (64·6) 293 (58·4) Chemotherapy 27 (5·2) 30 (4·6) 20 (2·9) 25 (3·3) 77 (29·1) 98 (29·1) 119 (25·7) 157 (31·3) Chemoradiotherapy 0 (0·0) 4 (0·6) 11 (1·6) 3 (0·4) 15 (5·7) 28 (8·3) 37 (8·0) 37 (7·4)
Radiotherapy 1 (0·2) 0 (0) 2 (0·3) 1 (0·1) 0 (0) 0 (0) 1 (0·2) 2 (0·4)
Unknown therapy 0 (0) 4 (0·6) 2 (0·3) 3 (0·4) 5 (1·9) 9 (2·7) 6 (1·3) 9 (1·8)
Unknown 2 (0·4) 4 (0·6) 3 (0·4) 4 (0·5) 3 (1·1) 1 (0·3) 1 (0·2) 4 (0·8)
Values in parentheses are percentages.
*Patients scheduled for curative resection.
†Determined at start of surgical procedure, including conversions (and hybrid procedures for oesophageal cancer).
‡Patients who underwent resection.
§Patients who underwent curative resection, determined at time of surgery.
Table 4. Results of individual performance indicators for patients with oesophageal cancer or gastric cancer included in the Dutch Upper Gastrointestinal Cancer Audit
2011 2012 2013 2014 P§
Oesophageal cancer
Process n = 551 n = 725 n = 728 n = 782
Preoperative MDT meeting 546 (99·1) 720 (99·3) 724 (99·5) 780 (99·7) 0·108
Time from diagnosis to treatment < 5 weeks* 179 (32·5) 226 (31·2) 248 (34·1) 321 (41·0) < 0·001 Postoperative MDT meeting 503 (91·3) 682 (94·1) 693 (95·2) 762 (97·4) < 0·001 Preoperative treatment† 508 of 550 (92·4) 636 of 716 (88·8) 649 of 726 (89·4) 693 (88·6) 0·037
Outcome‡ n = 511 n = 644 n = 678 n = 738
≥ 15 lymph nodes in resection specimen 257 (50·3) 390 (60·6) 431 (63·6) 539 (73·0) < 0·001 Tumour-negative resection margins 470 (92·0) 586 (91·0) 621 (91·6) 701 (95·0) 0·023 Complicated postoperative course 162 (31·7) 180 (27·9) 195 (28·8) 253 (34·3) 0·183
In-hospital/30-day mortality 21 (4·1) 26 (4·0) 31 (4·6) 26 (3·5) 0·684
Gastric cancer
Process n = 314 n = 420 n = 565 n = 588
Preoperative MDT meeting 299 (95·2) 411 (97·9) 552 (97·7) 579 (98·5) 0·009
Time from diagnosis to treatment < 5 weeks* 149 (47·4) 186 (44·3) 283 (50·1) 270 (45·9) 0·915 Postoperative MDT meeting 293 (93·3) 397 (94·5) 542 (95·9) 576 (98·0) < 0·001 Preoperative treatment† 158 of 290 (54·5) 220 of 387 (56·8) 272 of 526 (51·7) 316 of 562 (56·2) 0·917
Outcome‡ n = 261 n = 334 n = 457 n = 497
> 15 lymph nodes in resected specimen 124 (47·5) 197 (59·0) 295 (64·6) 366 (73·6) < 0·001 Tumour-negative resection margins 216 (82·8) 297 (88·9) 397 (86·9) 437 (87·9) 0·157 Complicated postoperative course 51 (19·5) 71 (21·3) 94 (20·6) 90 (18·1) 0·474
In-hospital/30-day mortality 21 (8·0) 16 (4·8) 21 (4·6) 20 (4·0) 0·031
Values in parentheses are percentages.
*Interval between diagnosis and start of neoadjuvant therapy or day of surgery.
†Patients scheduled for curative resection.
‡Patients with a primary tumour who underwent potentially curative surgery, as decided by the surgeon at the end of surgery. MDT, multidisciplinary team.
§χ2 test for trend.
11
5
3 4
34 40 26
16 22 23
10 6 4
9 15 7
1 2
5
7 8
4
3 3 1
4
6
1 2
5 5
8 4
4 5 5 5
1 2 4
10 13
17
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2011 2012 2013 2014 2011 2012 2013 2014 2011 2012 2013 2014 Hospitals performing
oesophagectomies for oesophageal cancer
Hospitals performing gastrectomies for
gastric cancer
Hospitals performing resections for oesophagogastric
cancer
% Hospitals
≥ 40 resections / hospital 30 - 39 resections / hospital 20 - 29 resections / hospital < 20 resections / hospital
Figure 1. Process of centralization for hospitals involved in the surgical treatment of patients with oesophageal or gastric cancer
DISCuSSION
This study provides an overview of the implementation of a nationwide audit evaluating the quality of care for patients who had surgery for oesophageal or gastric cancer in the Netherlands. The main purpose of the audit was to improve the quality of care for these patients by provid-
ing (surgical) teams with reliable, benchmarked information on process and (case mix-adjusted) outcome parameters regarding their patients.
Between 2011 and 2014, 2786 patients with oesophageal cancer and 1887 with gastric cancer were registered. Case ascertainment approached 100 per cent for patients registered from 2013. During the study period a trend towards better results for various process and outcome measures was observed. Both auditing and the process of centralization might have contributed to this14.
In 2014, approximately 2500 patients were diagnosed with invasive oe- sophageal cancer (including gastro oesophageal junction tumours) and approximately 1400 patients were diagnosed with invasive gastric cancer in the Netherlands15. Oesophagogastric cancer surgery is considered a high-risk and low-volume surgical procedure, with substantial morbidity and mortality and poor survival16, making it an appropriate subject for clinical audit.
Clinical auditing aims to improve standards of care using benchmarked information, and by initiating improvement programmes or in depth re- search to clarify underlying causes and mechanisms. Additionally, clinical auditing has the potential to provide patients and third parties with valid case mix-adjusted quality information. Transparency of treatment results could lead to outcome-based referral instead of volume-based referral17. Clinical audit can also act as a source of information for research by pro- viding data on a non-selected patient cohort including patients who do not enter clinical trials.
Several European countries have started registering detailed informa- tion about patients treated for oesophagogastric cancer. For instance, in Denmark and Sweden a national oesophagogastric cancer registry was
initiated in 2003 and 2006 respectively, and the NOGCA in the UK started collecting detailed information about this group of patients in 2011.
The DUCA differs from these other audits in a number of ways. Weekly updated and benchmarked feedback of individual hospital results is provided, allowing hospitals and their medical teams to act on results at relatively short notice. Participation in the DUCA has been incorporated as a quality standard defined by medical specialists, supported by the Health Care Inspectorate. This stimulated successful nationwide implementation of the audit, in contrast with the voluntary nature of some audits and registries that could lead to participation mainly from dedicated hospitals and underrepresentation of underperforming hospitals (as in the US National Surgical Quality Improvement Program and the initial years of the UK NOGCA).
The present audit has limitations. As a surgical audit, there is no informa- tion about patients with oesophageal or gastric cancer undergoing non- surgical treatment or no treatment. The audit cannot, therefore, provide information about resection rates, toxicity of perioperative treatment and non-surgical treatment measures. As a result, no comparisons can be made between different surgical and non-surgical treatment strategies.
The board and the scientific committee of the DUCA are working on the development of a nationwide multidisciplinary quality registration. The audit does not contain information about long-term follow-up. Links be- tween other databases including the NCR could resolve this issue, making both registries more meaningful. There is no information about patient- reported outcome measures. Insight into patient views on outcomes will likely play an increasingly important role in improving the quality of care.
Finally, registration of all data is time-consuming and burdensome for individual surgeons and hospitals. This issue could be resolved by linking the DUCA database with existing data systems and by computerizing as
much as possible. Currently, a link between the DUCA and PALGA, the nationwide network and registry of histopathology and cytopathology in the Netherlands, is being established to reduce some of the registration burden.
Existing clinical audits focusing on oesophagogastric cancer care offer great opportunities. Benchmarking of outcomes might be expanded to a European level and treatment results can then be compared with those from a wider range of centres in different settings18. A comparison between four different European cancer registries and audits performed by Dikken and colleagues5 in 2012 showed a significantly higher 30-day mortality rate after gastric resections for cancer in the Netherlands but, after 4 years of auditing, postoperative mortality rates in the Netherlands have declined and become more comparable to results in other European countries. Even so, it must be acknowledged that recorded data vary by country and there are differences in data interpretation. A common data item list, as presented by the European Registration of Cancer Care in 2014, is likely to be of great value18. This list may prove beneficial for existing audits and could serve as an example for new audits. Additionally, international consensus on standardization of data collection is essential to assess and compare results from different countries19.
Nationwide implementation of a surgical audit in the Netherlands has been successful, and included nearly 100 per cent of the patients with oesophageal cancer or gastric cancer who underwent surgery in 2013.
The first results give a valuable insight into the quality of surgical care for patients with oesophageal or gastric cancer, with a positive trend for various process and outcome measures.
REFERENCES
1. Dikken JL, Lemmens VE, Wouters MW, et al. Increased incidence and survival for oesophageal cancer but not for gastric cardia cancer in the Netherlands. Eur J Cancer 2012; 48: 1624–1632.
2. Dassen AE, Dikken JL, Bosscha K, et al. Gastric cancer: decreasing incidence but stable survival in the Netherlands. Acta Oncol 2014; 53: 138–142.
3. Dikken JL, Dassen AE, Lemmens VE, et al. Effect of hospital volume on postop- erative mortality and survival after oesophageal and gastric cancer surgery in the Netherlands between 1989 and 2009. Eur J Cancer 2012; 48: 1004–1013.
4. Dutch Institute for Clinical Auditing (DICA). Annual Report 2014. http://www.
clinicalaudit.nl/jaarrapportage/2014/duca.html [accessed 21 October 2015].
5. Dikken JL, van Sandick JW, Allum WH, et al. Differences in outcomes of oesopha- geal and gastric cancer surgery across Europe. Br J Surg 2013; 100: 83–94.
6. Wouters MW, Jansen-Landheer ML, van de Velde CJ. The Quality of Cancer Care initiative in the Netherlands. Eur J Surg Oncol 2010; 36(Suppl 1): S3–S13.
7. Van Leersum NJ, Snijders HS, Henneman D, et al. The Dutch surgical colorectal audit. Eur J Surg Oncol 2013; 39: 1063–1070.
8. Boulkedid R, Abdoul H, Loustau M, et al. Using and reporting the Delphi method for selecting healthcare quality indicators: a systematic review. PLoS One 2011; 6:
e20476.
9. Health and Social Care Information Centre. National Oesophago-Gastric Cancer Audit (NOGCA). http://www.hscic.gov.uk/org [accessed 21 October 2015].
10. Oncoline. Guideline Treatment Gastric Carcinoma. http://www.oncoline.nl/
maagcarcinoom [accessed 7 August 2015].
11. Oncoline. Guideline Treatment Oesophageal Carcinoma. http://www.oncoline.nl/
oesofaguscarcinoom [accessed 21 August 2015].
12. Spiegelhalter DJ. Funnel plots for comparing institutional performance. Stat Med 2005; 24: 1185–1202.
13. Dutch Institute for Clinical Auditing (DICA). Annual Report 2013. https://www.dica.
nl/jaarrapportage-2013/ [accessed 21 October 2015].
14. van Lanschot JJ, Hulscher JB, Buskens CJ, et al. Hospital volume and hospital mortality for esophagectomy. Cancer 2001; 91: 1574–1578.
15. Netherlands Cancer Registry. Cijfers over kanker. http://www.cijfersoverkanker.nl [accessed 1 October 2015].
16. Anderson O, Ni Z, Moller H, et al. Hospital volume and survival in oesophagectomy and gastrectomy for cancer. Eur J Cancer 2011; 47: 2408–2414.
17. Wouters MW, Krijnen P, Le Cessie S, et al. Volume- or outcome-based referral to improve quality of care for esophageal cancer surgery in the Netherlands. J Surg Oncol 2009; 99: 481–487.
18. de Steur WO, Henneman D, Allum WH, et al. Common data items in seven Euro- pean oesophagogastric cancer surgery registries: towards a European upper GI cancer audit (EURECCA Upper GI). Eur J Surg Oncol 2014; 40: 325–329.
19. Low DE, Alderson D, Cecconello I, et al. International consensus on standardization of data collection for complications associated with esophagectomy: Esophagec- tomy Complications Consensus Group (ECCG). Ann Surg 2015; 262: 286–294.
