Anticipating implementation of colorectal cancer screening in The Netherlands: a nation wide survey on endoscopic supply and demand

The demand for gastrointestinal endoscopic procedures and gastroenterological care in The Netherlands and surrounding countries has been steadily increasing over the years [1]. Demographic changes due to an aging population and technical developments like video capsule endoscopy, double balloon endoscopy, and magnifying techniques, are likely to have contributed to this increase in workload. The existing shortage of gastroenterologists and endoscopy staff in The Netherlands is clear from waiting lists and vacancies [1]. This shortage is likely to be compounded by the implementation of a nationwide screening program for colorectal cancer (CRC), as it would lead to a further increase in endoscopic demand.

CRC screening programs have to comply with the Wilson and Jungner criteria including the availability of sufficient facilities for diagnosis and treatment [2]. Several European countries favour population-based screening by faecal occult blood tests, which has been shown to decrease CRC-related mortality [3‐5]. Recently, the Dutch Minster of Health has decided to start the implementation of a CRC screening program in The Netherlands in 2013 [6]. In these program individuals aged 55-75 years are offered a faecal immunochemical test (FIT) biennially, at a cut-off value of ≥75 ng/ml [7]. All individuals with a positive FIT should subsequently be referred for colonoscopy.

To meet the Wilson and Jungner criteria, current endoscopic capacity in The Netherlands as well as other European countries needs to be determined. On behalf of the Dutch Society of Gastroenterology, this study aims to update current endoscopic production and determine the trends in endoscopic procedures performed in The Netherlands since 2004. In addition, the impact of implementing a FIT-based screening program on daily endoscopic practice is evaluated, and the number of colonoscopies currently being performed in several European countries is inventoried. In this study, a large increase in colonoscopies performed in The Netherlands is shown. For Europe, the number of colonoscopies performed appears to vary widely, but published data are lacking.

This nationwide endoscopy survey had a 100% response rate and has shown a 24% increase in endoscopies performed in The Netherlands over the past five years, and a 55% increase over the past 10 years. In particular, the number of colonoscopies increased by 64%. This is accompanied by only a minor increase in the number of endoscopists. Anticipating a national FIT-based screening program, an additional 10-22 colonoscopies per unit per week would be required in year six of the screening program, necessitating a 15% increase over the current endoscopic workload. Putting these figures into a European perspective, it was shown that almost half of the European countries that responded performed similar quantities of colonoscopies per 100,000 inhabitants. Yet, a striking heterogeneous distribution across Europe was observed.

An increase in the number of endoscopies was anticipated due to population growth and potentially changing morbidity patterns. However, the 64% increase in the number of colonoscopies over the past five years is remarkable. Several CRC screening trials have been ongoing in The Netherlands since 2009 [15‐17], but the cumulative number of extra colonoscopies in these trials is estimated not to exceed 2,500. Consequently, these trials will only have a minor effect on the total number of colonoscopies. Due to the increasing awareness of CRC and the need for a screening program, opportunistic screening has probably increased as well. In addition, although speculative, considering sigmoidoscopy rates decreased over the same time period, subjects who present with rectal bleeding may be being referred for total colonoscopy more frequently.

Although the total number of endoscopists increased slightly, the number of gastroenterologists increased substantially whereas the number of internists and surgeons performing endoscopies in 2009 declined markedly. The increase in the total number of endoscopies performed might be related to this change as gastroenterologists might spend more full time equivalents on performing endoscopies, what could result in an increase in endoscopic production. However, full time equivalent volumes on endoscopic procedures per endoscopist's specialty is unknown.

The number of endoscopies performed varied substantially over the different provinces of The Netherlands. Yet, per 100,000 inhabitants, this difference was at maximum 1.6 fold. These differences might be related to variation in available endoscopists and, although speculative, morbidity patterns. A difference in patient demographics among provinces as a potential explanation, could not be found [9].

A more extensive interpretation of the presented results in the context of the number of endoscopies performed in other European countries is difficult. Although sufficient endoscopic resources are clearly stated as a prerequisite before implementing a CRC screening program in the recent European CRC screening guidelines [18], very little data on European production are available. Only from Ireland and Romania have recent data been published [10, 11]. Other studies did not report on current and total capacity, but did report on models of capacity needed in CRC screening [19, 20]. The results of our pilot inquiry indicate that the number of colonoscopies per 100,000 inhabitants has a wide range from less than 126 in Turkey to 3,031 in Germany. If Romania and Turkey are considered representatives of Eastern Europe, then endoscopic capacity in Eastern Europe needs to be increased [11]. Half of the responding countries perform 950-1,263 colonoscopies per 100,000 inhabitants.

Anticipating a national screening program in The Netherlands, current capacity may be insufficient as the number of extra colonoscopies needed is expected to range from 10 to 22 per unit per week. Although subjective, 22% of endoscopy units expect to be able to cope with a 30% increase in workload in 2012, whereas the anticipated increase would only be 15%. The number of endoscopic procedures performed in 2009 are considered current minimal capacity. However, capacity also relies on endoscopy staff, medical co-workers, flexible endoscopes, (desinfecting) material, medication and effective use of resources. In addition, the number of endoscopies is also influenced by procedure guidelines and quality issues. For example, in 2004 the guideline for cleansing and desinfection of flexible endoscopes was intensified [21]. Current available facilities in Dutch endoscopy units are unknown. Improvements in operational efficiency and/or technological advancements could increase current capacity. Rest capacity is however unknown and warrants additional studies. Therefore, current capacity level may actually be higher or lower than the production level. Still, there might be no or little rest capacity as for years many vacancies for gastroenterologist exists in The Netherlands [22]. It can be expected that capacity needs to increase at least to some extent, as more colonoscopies need to be performed in a screening program. Also, an increase in workload for medical co-workers can be expected as e.g. in upper gastrointestinal endoscopies, it was shown that around 75% of total time was spent on pre-and postendoscopic operations [23]. Importantly, before implementing screening based on the present results, it should be ascertained that the increase in procedures performed does not hamper quality of the procedures. Due to the substantial increase over the recent years without a considerable increase in manpower, the upper limit of capacity might have been reached already, and a further increase without sufficient investment might result in low quality colonoscopies.

Whether investment in increasing capacity is needed or not, eventually the number of endoscopic procedures will decrease again. It is estimated that 10% of colonoscopies performed in daily practice are performed for opportunistic screening [24], which will become unnecessary after the implementation of a national screening program. In addition, less colonoscopies due to less frequent symptomatic presentations of CRC and a decrease in CRC incidence is expected [25‐28]. On the contrary, the implementation of a screening program will also result in more adenomas being detected, that consequently warrants ongoing colonoscopic surveillance. Capacity could be increased by critically reviewing guidelines for post-polypectomy surveillance [29, 30].

Some limitations of the present study need to be discussed. Firstly, this study describes the changes in daily endoscopic practice, but the factors that drive these changes are largely unknown. Secondly, due to missing data on full time equivalents for 2004, the trends in fulltime manpower could not be determined completely. A substantial increase in the number of endoscopies performed per endoscopist was found, so even if more endoscopists would work part-time in 2009, this number would in fact be even higher. Thirdly, the response rate for the data from 2004 was slightly lower than for 2009 (98% versus 100% respectively). Therefore, the increases presented could be slight overestimations and the decreases could be slight underestimations. However, modelling the missing data based on the complete data of 2009, the over/underestimation is expected not to be larger than 1-3% (data not shown). Fourthly, it was assumed that the 78,000 extra colonoscopies required for FIT-based CRC screening would distribute according to the number of inhabitants aged 55-75 in the different regions of The Netherlands, and that all individuals would get their colonoscopy in the province of residence. However, participation rates for FIT screening might show regional differences, and patients may prefer the closest hospital instead of a hospital in their own province. Fifth, the data presented for other European countries should be interpreted with caution. The data provided by European colleagues were mainly obtained from national or regional registries. As these data were not collected in the same standardized manner and therefore were not validated by the authors of this manuscript, the European data may not be the exact reflection of each country. Still, these data are provided with discretion by the European endoscopists, and we believe this effort is a good first inventory of current endoscopic variety. In addition, the lack of published data warrants more studies especially from countries in which CRC screening is advocated, planned or already implemented.

In conclusion, current Dutch endoscopic capacity has increased substantially since 2004. For the number of colonoscopies in particular, a remarkable increase was found, although the number of endoscopists increased marginally. When a national CRC screening program will be implemented in The Netherlands, the increase of colonoscopies needed per unit will be at maximum 22 per week. Finally, the distribution of colonoscopies performed in Europe is very heterogeneous and more solid data are needed for proper analysis.