Capsule Colonoscopy Increases Uptake of CRC Screening
The present study for the first time analyses the effect of offering a new examination method on the uptake of CRC screening. In Germany, CRC screening by colonoscopy is based on spontaneous uptake and no invitation system exists as yet. In our study we were able to show that in addition to the presumed invitation effect – which per se led to an increase in colonoscopy rate by about 60 % – a fourfold increase in endoscopic screening could be attributed to the offer of capsule colonoscopy, with men in particular finding capsule colonoscopy more acceptable. However, the overall adenoma yield was not different in the two examination groups, although all patients with positive capsule findings and 40 % with incomplete capsule examination underwent secondary conventional colonoscopy. This points towards a need to improve capsule colonoscopy diagnostic efficacy before it can be considered for implementation into a CRC screening programme.
Our study results deserve several comments which have only partly to do with limitations of the present study:
In our study, a 'one-stop shop' process – using one colonic preparation for capsule endoscopy, followed by colonoscopy in case of lesions, requiring very rapid reading - was not attempted or done (and this was explained to patients during informed consent). Patients with positive or doubtful findings therefore had to undergo a second bowel preparation. This may be regarded as drawback of capsule endoscopy, even if it only applies to 20 %–30 % of patients. In previous capsule studies where colonoscopy was used as the gold standard in all cases, such as in the large multicenter trial, the protocol included a single preparation for capsule plus colonoscopy procedures, but this did not include patient selection for colonoscopy by capsule as a filter test when positive, since capsule results were not known at the time of colonoscopy. It may be quite difficult to establish such programmes of capsule endoscopy, with performance and reading, plus colonoscopy, all routinely done in 1 day. Further studies have to show how such ambitious programmes might possibly be implemented, for example by fast central reading services and an on-call colonoscopy service for screenees with polyps identified by the capsule, and whether such programmes might further increase uptake. Only if larger and stable numbers of capsule colonoscopies should ever be performed, and regarded as economically viable, might such a service appear more realistic. It will then also become more evident to which degree uptake is influenced by other factors such as capsule performance (sensitivity, need to still undergo colonoscopy) and logistical issues (such as one-stop shopping).
Discussions
The present study for the first time analyses the effect of offering a new examination method on the uptake of CRC screening. In Germany, CRC screening by colonoscopy is based on spontaneous uptake and no invitation system exists as yet. In our study we were able to show that in addition to the presumed invitation effect – which per se led to an increase in colonoscopy rate by about 60 % – a fourfold increase in endoscopic screening could be attributed to the offer of capsule colonoscopy, with men in particular finding capsule colonoscopy more acceptable. However, the overall adenoma yield was not different in the two examination groups, although all patients with positive capsule findings and 40 % with incomplete capsule examination underwent secondary conventional colonoscopy. This points towards a need to improve capsule colonoscopy diagnostic efficacy before it can be considered for implementation into a CRC screening programme.
Our study results deserve several comments which have only partly to do with limitations of the present study:
Uptake was rather low even under study conditions using an invitation process; however, uptake within the German opportunistic screening programme is generally limited (around 3 % annually) but was lower in the study area in the years preceding the study (1 %). It could be that the effect of offering capsule endoscopy may be less pronounced in areas with higher uptake, but this would have to be studied.
We performed a single-arm rather than a randomized study; from a purely scientific standpoint, only the latter, randomizing patients to being invited to either colonoscopy or capsule endoscopy would have enabled us to clearly differentiate between the invitation effect and the effect of offering a new technique. Nevertheless, our design mirrored the reality where often more than one options are offered. A further limitation to generalize our result was that only one patient group covered by one insurance was included in our study.
In the literature, direct comparisons of spontaneous versus invited screening colonoscopy uptake rates are not available, since it is difficult to measure the spontaneous uptake outside of a programme. However, there is ample indirect evidence that invitation methods increase uptake, and there are also randomized and other studies showing that more intensive invitation measures and the involvement of general practitioners in the invitation process lead to greater uptake. In our study, for reasons of data protection, 35 persons who underwent colonoscopy and who also had received invitation letters, but did not enter the project and/or sought information from one of the 4 project gastroenterologists, were not counted as study participants since their individual data could not be used. We decided that these cases would not be counted among the persons with concomitant spontaneous uptake. If we did this, the uptake increase would have been much higher for colonoscopy, further underlining the importance of an invitation process. Naturally, screening behavior would furthermore be different in setting with free access to the respective screening test versus self-paid methods; since colonoscopy is reimbursed over the age of 55 (without colonoscopy in the preceeding 10 years), we think that free access to capsule colonoscopy allowed for a fair comparison in the German setting. Very likely, this would change with changing reimbursement strategies.
Data on adenoma yield by capsule colonoscopy was disappointing, although our study was not powered to show differences in polyp yield. Firstly, it was noteworthy that all persons with polyps detected on capsule endoscopy also agreed to undergo secondary colonoscopy; in the group with incomplete capsule endoscopy, this rate was lower but still relevant (40 %). In the recent large multicenter trial on capsule endoscopy controlled by conventional colonoscopy, sensitivity for polyps less than 1 cm was only slightly over 60 %. Thus, it is conceivable that under routine conditions as in our study, sensitivity can be even lower; we think that sensitivity would have to be substantially improved before capsule colonoscopy would be an option for CRC screening. Polyps also tended to be smaller in the colonoscopy group which might be related to the lower sensitivity of capsule for smaller polyps. It appears possible that this would be the case with the second generation colon capsules, but this would have to be proven in either a similar study and/or a comparative trial
The offer of capsule endoscopy had a much better effect on uptake in men than in women. Men are known to have a lower uptake than women of colorectal screening measures, at least in Germany: a detailed analysis of colonoscopy screening participation in Germany shows a 10 % lower uptake of screening colonoscopy by men, which is increased in the age groups below 70. Experiences from other countries are however variable and more difficult to predict. Our results show that men might be better motivated by the offer of this new technology, but these results are of course preliminary. The durability and reproducibility of such effects at repeated investigation is not known either.
In our study, a 'one-stop shop' process – using one colonic preparation for capsule endoscopy, followed by colonoscopy in case of lesions, requiring very rapid reading - was not attempted or done (and this was explained to patients during informed consent). Patients with positive or doubtful findings therefore had to undergo a second bowel preparation. This may be regarded as drawback of capsule endoscopy, even if it only applies to 20 %–30 % of patients. In previous capsule studies where colonoscopy was used as the gold standard in all cases, such as in the large multicenter trial, the protocol included a single preparation for capsule plus colonoscopy procedures, but this did not include patient selection for colonoscopy by capsule as a filter test when positive, since capsule results were not known at the time of colonoscopy. It may be quite difficult to establish such programmes of capsule endoscopy, with performance and reading, plus colonoscopy, all routinely done in 1 day. Further studies have to show how such ambitious programmes might possibly be implemented, for example by fast central reading services and an on-call colonoscopy service for screenees with polyps identified by the capsule, and whether such programmes might further increase uptake. Only if larger and stable numbers of capsule colonoscopies should ever be performed, and regarded as economically viable, might such a service appear more realistic. It will then also become more evident to which degree uptake is influenced by other factors such as capsule performance (sensitivity, need to still undergo colonoscopy) and logistical issues (such as one-stop shopping).
SHARE
