Abstract
Objectives
To investigate the characteristics of participants screened for bowel cancer using a faecal immunochemical test for haemoglobin (FIT).
Setting
Scottish Bowel Screening Programme.
Methods
65909 men and women in two NHS Boards, aged 50 to 74, were invited to participate in an evaluation of FIT as a first-line test. Uptake was calculated by sex, age in quintiles, and deprivation in quintiles, and compared with a group who had completed a guaiac faecal occult blood test (gFOBT) and for whom details of sex, age and deprivation were well documented.
Results
FIT kits from 38672 participants were tested. The overall uptake of 58.7% was significantly higher than the 53.9% for gFOBT (p < 0.0001). Uptakes in the two NHS Boards were 57.6% and 54.4% for men and 63.2% and 59.1% for women, higher than the 49.5% and 58.1% completing gFOBT. Uptake was higher for FIT than gFOBT in all age and deprivation quintiles for both men and women in both NHS Boards. The difference in uptake fell with age for men but rose for women; the increase in uptake was greater for men than women. Uptake fell as deprivation decreased for both sexes, and was similar in both NHS Boards.
Conclusions
Use of FIT increases uptake over gFOBT, and the greatest increases are seen in men, younger participants, and more deprived individuals, groups for which an increase in uptake is likely to be beneficial. The results support a move to FIT as a first-line screening test for those countries still using gFOBT.
INTRODUCTION
An aspiration in all cancer screening programmes is increased uptake, particularly in the groups in which participation is low. A range of interventions to increase uptake is possible. 1 In randomized controlled trials of asymptomatic population bowel (colorectal cancer) screening, guaiac faecal occult blood tests (gFOBT) have been shown to reduce mortality. 2 These are currently used as the initial test in bowel screening programmes in all four countries of the United Kingdom, and elsewhere. The modest decrease in mortality found in randomized controlled trials has also been found in practice. 3,4 However, gFOBT have many disadvantages, 5 and current expert opinion is that faecal immunochemical tests for haemoglobin (FIT) are the best non-invasive tests for bowel screening. 6–8 FIT have many advantages over traditional gFOBT. One of the major merits, possibly because only one sample is usual instead of two samples from each of three faeces, and the faecal collection devices for many FIT are user-friendly, is that uptake with FIT has been found to be higher than with gFOBT; this was detailed in a recent systematic review and meta-analysis, in which it was stated that more research examining FIT from a participant perspective is warranted. 9
In Scotland, there are 14 NHS Boards responsible for the protection and improvement of their population's health and for the delivery of frontline healthcare services; these are very heterogeneous in terms of size and population characteristics. 10 In our assessment of a recent evaluation of FIT as a first-line test in two rather different NHS Boards in Scotland (NHS Tayside and NHS Ayrshire & Arran), we documented and compared uptake with that obtained contemporaneously in two other NHS Boards (NHS Fife and NHS Forth Valley), selected for their similarity to NHS Tayside and NHS Ayrshire & Arran, respectively, and also before and after the evaluation in all four NHS Boards. 11 The overall uptake in the two evaluation NHS Boards that participated in the study rose by 5.2% and 6.0% during the use of FIT, but then fell to values similar to those previously seen when the current gFOBT/FIT algorithm used in Scotland 12 was reinstituted. The two NHS Boards in which FIT was not used had small, random changes in uptake over time, but did not have the important rise in participation seen with use of FIT. It was not known what local health promotion efforts were undertaken in the four NHS Boards, if any, as these are local rather than national or regional responsibilities, but it may be that these efforts affect uptake. However, for the first time, in a fully rolled out national bowel screening programme, rather than in a simple trial comparing gFOBT and FIT, we demonstrated that use of FIT did increase uptake over gFOBT. It is well documented that uptake using gFOBT is lower in men than in women, increases with age, and is lower in the most deprived. 13,14 Thus, an aspiration of any bowel screening programme would be to find methods to increase uptake, particularly in men, the young and the deprived. We therefore investigated whether the overall increased uptake we found with FIT as a first-line test was achieving this aim, by comparison with a group from nine NHS Boards for whom we had detailed information on age, sex and deprivation.
METHODS
Several aspects of the FIT as a first-line test evaluation have been documented previously. 11,15,16 All individuals aged 50 to 74 in the two evaluation NHS Boards eligible to participate from 1 July 2010 to 12 January 2011 were sent an invitation pack. This contained an invitation letter, a “Know the Facts” leaflet on bowel cancer, a thin card wallet with printed written and pictorial instructions for specimen collection from a single faeces sample, a specimen collection device taking 10 mg faeces into 2.0 ml preservative buffer (Eiken Chemical Co., Tokyo, Japan), and a small zip-lock plastic bag containing absorbent material. The participant received an adhesive identification label that was integral to the invitation letter; this was attached to the outside of the zip-lock bag before return of the specimen to the Scottish Bowel Screening Laboratory in a foil mailing pouch. This label documented the name of the participant, the kit number and the Community Health Index (CHI) number, which is a unique 10-digit identifier used ubiquitously in NHS Scotland providing information on sex and age. The specimen collection devices were returned through the UK Post Office by first class freepost. On return to the Laboratory, the foil mailing pouches were opened and the receipt of a specimen captured electronically by the Scottish Bowel Screening System.
Uptake was calculated as the percentage of people with a screening test result (positive or negative) out of the total number invited. Sex and age were derived from the CHI number. Deprivation was categorized from individual postcodes using the Scottish Index of Multiple Deprivation (SIMD), which identifies small area concentrations of multiple deprivation across Scotland based on income level, employment, health, education, skills and training, housing, geographical access, and crime. The group used for comparison was the 10774 individuals for whom detailed information on uptake by sex, age, and deprivation were available and that we have described in detail previously. 17 These were resident in nine of the 14 NHS Boards in Scotland, and received the usual invitation gFOBT and associated materials used in the Scottish Bowel Screening Programme. Uptake was compared using chi-squared tests, and logistic regression was performed to determine the effect of the use of FIT as a first-line test independently of other factors that may influence uptake.
RESULTS
Of the 65909 people invited to participate in the FIT as a first-line test evaluation, 38672 (58.7%) provided a sample suitable for analysis, yielding a positive or negative screening test result. Uptake was 18037 (56.1%) for men, 9492 (57.6%) in NHS Tayside and 8545 (54.4%) in NHS Ayrshire & Arran, and 20635 (61.2%) for women, 10905 (63.2%) in NHS Tayside and 9730 (59.1%) in NHS Ayrshire & Arran. The uptake was 53.9% for the 19987 residents in the nine NHS Boards for whom gFOBT was the initial test, made up of 9704 (48.6%) men and 10283 (51.4%) women; uptake was 4801 (49.5%) and 5973 (58.1%) for men and women respectively. Uptake for FIT was higher than for gFOBT, overall and for men and women (p < 0.0001).
Uptake using guaiac faecal occult blood test (gFOBT) by sex and age quintile and comparison with faecal immunochemical test (FIT) in two NHS Boards
† T: NHS Tayside, A&A: NHS Ayrshire & Arran; *: p < 0.05, **: p < 0.01, ***: p < 0.001
Uptake using guaiac faecal occult blood test (gFOBT) by sex and Scottish Index of Multiple Deprivation (SIMD) quintile and comparison with faecal immunochemical test (FIT) in two NHS Boards
† T: NHS Tayside, A&A: NHS Ayrshire & Arran; *: p < 0.05, **: p < 0.01, ***: p < 0.001
Unadjusted and adjusted odds ratios for FIT uptake by sex, age quintile and Scottish Index of Multiple Deprivation (SIMD) quintile, with guaiac faecal occult blood test (gFOBT) uptake in that group the reference category
* Adjusted for all other variables, p < 0.001 for all comparisons
DISCUSSION
There are a number of plausible reasons why uptake with FIT is higher than with gFOBT. In addition to requiring only one sample for FIT as opposed to two samples from each of three faeces for gFOBT, the sample collection devices for FIT may appear to participants to be more “scientific” or “clinical” tools than the simple gFOBT cards. Moreover, the devices are less messy to use. Further, once the single sample is collected, the device is returned immediately, and the participant does not have to store and handle a faecal laden gFOBT card a number of times until the test is completed. The uptakes in NHS Tayside are generally higher for men and women in all age quintiles than in NHS Ayrshire & Arran. This may reflect the fact that screening in NHS Tayside began in May, 2000, on the commencement of the UK pilot of gFOBT screening for bowel cancer, whereas NHS Ayrshire & Arran began Screening only at the beginning of rollout of the Scottish Bowel screening Programme in September, 2007. Perhaps the participants in NHS Tayside, exposed to gFOBT, for a number of screening rounds, were keen to attempt the new FIT whereas those in NHS Ayrshire & Arran did not appreciate the advantages of FIT over gFOBT as they were unexposed to either test. Uptakes in the different deprivation classes, however, appeared similar in the two NHS Boards.
The major strength of this study, in contrast to previous work on comparing uptake of FIT and gFOBT, is that the data on uptake from the FIT as a first-line test evaluation and those from our comparison group were generated within an on-going fully rolled-out national bowel screening programme, and not in a research setting. Moreover, because of the use of the CHI as the participant identifier in NHS Scotland, checked by the participant and on receipt of the FIT or gFOBT kits in the Scottish Bowel Screening Centre Laboratory, the data on sex, age, and deprivation of all individuals were of very high quality. A further strength is that we compared uptakes by sex, age, and deprivation with a group for which we had very comprehensive data on these characteristics.
A possible weakness is that the data collection was done at different times for the two groups, FIT from 1 July 2010 to 12 January 2011 and gFOBT from 13 April 2009 to 29 May 2009, a shorter period. In addition, data on uptake for the FIT was from participants in two NHS Boards, whereas the data for gFOBT was from nine NHS Boards. As described previously 11 , we have compared the overall uptake of FIT with that of gFOBT in detail in four NHS Boards contemporaneously over four six month periods. Detailed exploration and comparison of the sex, age, and deprivation characteristics in participants in all four NHS Boards would have been of much interest, however, the detailed data required for this were unavailable from our two comparative participants in NHS Fife and NHS Forth Valley.
Although much less comprehensive, there are other data on the uptake in Scotland for groups with different sex, age and deprivation characteristics which confirm our findings. There are detailed data from the three pilot screening rounds in Scotland 13 and, although our results that uptake with FIT was higher in both sexes and through all ages and deprivation categories are confirmed by detailed comparison (not shown) of FIT and the gFOBT documented for the pilot screening rounds, the population offered screening was aged from 50 to 69 only, rather than from 50 to 74. In addition, the screening algorithms used differed, as gFOBT alone was used rather than the current gFOBT/FIT two-tier reflex screening algorithm developed from our research performed during the pilot screening rounds. 12 Further, only three NHS Boards were involved in the three pilot screening rounds, and there was little media publicity during the pilots. A more recent compilation of data on uptake and deprivation covering 13 of the 14 NHS Boards in Scotland for the two-year period from 1 November 2009 to 31 October 2011 has been made available. 18 Increases in uptake with FIT over these Key Performance Indicator data, for SIMD 1 through 5, were 6.5%, 5.0%, 4.3%, 3.3%, and 4.1% for both sexes, 7.0%, 5.8%, 5.6%. 4.4%, and 5.5% for men, and 6.0%, 4.3%, 3.5%, 2.2% and 2.8% for women. This data showed, as we also found, that uptake by men was higher than by women in both NHS Tayside and NHS Ayrshire & Arran, and the uptakes throughout the deprivation categories were very similar in both NHS Boards. Denominators are not published in the report 18 , however, two of the 13 NHS Boards contributing data were those in which the FIT as a first-line test evaluation was performed, and the increased uptake during this will have confounded the data, although on a population magnitude basis, only by a small amount. Through further comparison with these two additional groups, our findings that uptake with FIT rather than gFOBT was higher in both sexes, with age, and through deprivation categories are confirmed.
Ideally, in comparative studies, only one variable would be altered between study and control groups. We tried to achieve this in our study. The “Know the Facts leaflet” was identical for both evaluation NHS Boards and the other Health Boards in Scotland. Of necessity, there was a difference in the instructions for use for the gFOBT and the FIT. The different styles of these materials may have contributed to the increase in FIT uptake. Moreover, the invitation letter had: “You have been issued with a tube to collect your sample of bowel motion. It is different from the kit shown on the Know the Facts leaflet enclosed with this letter. You should find this easier to use.” Thus, an increase in uptake may have occurred simply because participants knew they were receiving a different kit. It would be difficult to dissect out the influences of all the possible factors that could increase uptake, however, as the introduction of FIT would require different materials for participants, this is somewhat irrelevant to our finding that uptake with FIT is higher than with gFOBT.
Most studies on comparisons of uptake of FIT and gFOBT have treated the screened population as a single entity. A study from Israel 19 has shown that overall test uptake was affected by sex, age, being immigrant, and socio-economic status, however, the overall uptakes with FIT and gFOBT were comparable. This is in contrast to all other studies, and is probably because participants were requested to prepare three consecutive daily samples, to keep the samples in their refrigerator, and bring the samples back to the clinic using a cooling bag provided with the kits. We used one FIT collection only and this is widely recognized to be appropriate for screening with FIT. 20,21 We have shown that the increases in uptake of FIT over gFOBT do vary with sex, age, and deprivation, but to the advantage of the population.
Although much information on screening for bowel cancer has been generated using gFOBT, the ever-growing view is that the disadvantages mean that many consider them dated. 22 Quantitative FIT kits are more expensive than gFOBT but, being based on automated immunoturbidimetric analytical systems, provide high-quality, reproducible results with fewer staff required to handle the workload in screening programmes. Visual interpretation of test results is eliminated. Lot-to-lot variation seen with gFOBT and qualitative FIT leading to varying positivity rates with time are also minimized. Moreover, with gFOBT in Scotland 8 , England 23 , and elsewhere, to keep the positivity rate low enough for the current colonoscopy resources to cope, two or three tests are required to generate an unequivocal positive result. This leads to increased expense for programmes, and a longer time to obtain a result, and therefore perhaps increased stress for participants. FIT provide results on first faecal specimen analysis, smoothing the organization and management of programmes, delivering a faster turnaround time, and possibly lowering stress for participants. However, there has been considerable investment in the current bowel screening programmes in all four countries of the UK, particularly in the laboratories set up for analysis of the returned gFOBT, and cogent and convincing business cases will be required to facilitate the move to FIT as a first-line test. Moreover, the introduction of flexible sigmoidoscopy 24 in England may lead to reappraisal of the optimal configuration of faecal tests (irrespective of whether gFOBT or FIT) and flexible sigmoidoscopy. This is still far from clear.
CONCLUSIONS
The many benefits of FIT include increased uptake in comparison with gFOBT. As shown by detailed multivariate analysis, the greatest increases in uptake of FIT over gFOBT are seen independently in men, younger participants, and more deprived individuals, the groups for which an increase in uptake will be beneficial. Using FIT narrows the gender, age, and deprivation gaps. Our results provide a compelling further rationale for the move to FIT as a first-line test for those countries still using gFOBT as the initial test, including the four UK health services.
Footnotes
ACKNOWLEDGEMENTS
We thank Iain McElarney of Mast Diagnostics Division, Bootle, Merseyside, UK, for his input into the preparation of material for potential participants and the setting up of the data capture for the automated analytical systems.