Answers are grounded in this paper and cite their sources.
By 31 December 2020, Australia's National Bowel Cancer Screening Program (NBCSP) should be fully in place. All Australians aged 50–74 years will then be receiving invitations to screen for bowel (colorectal) cancer every two years using an immunochemical fecal occult blood test (iFOBT). The date will mark complete implementation of one of the most important public health programs in Australia and one which, in our view, should have regional significance. This issue of APJCO features a new study (the study) of the NBCSP's impact, co-authored by the Australian Institute of Health and Welfare and the Australian Government Department of Health.1 It analyses data relating to the at-risk population in 2006–2008 and compares clinical outcomes in invitees and non-invitees up until December 2011. The findings are compelling, particularly when viewed in the context of other analyses of the NBCSP's reported and predicted benefits.2-4 Having been involved in the NBCSP's development and promotion over many years (from a non-government perspective), we take this opportunity to comment on the study and to add context in relation to program developments since 2011. In our view, it provides evidence of benefit worthy of consideration across the Asia-Pacific region. The study examined outcomes of the NBCSP through linkage with Australian state and territory cancer registries and the National Death Index. Linkage enabled the investigators to assess success in relation to the principal goals of the NBCSP – that is reducing the morbidity and mortality of bowel (colorectal) cancer. The findings indicate that the NBCSP is doing just that. In addition, the authors were able to show that the particular iFOBT test used for screening is reliable and effective. As reported, the study involved linking 2006–2008 invitees (the first to enter the NBCSP) with bowel cancer incidence and mortality data through to 31 December 2011, comparing outcomes for invitees with a never-invited group. Further comparisons were made between sub-groups of the invitees – those with screen-detected cancer, interval (i.e. “missed”) cancers and cancers in non-responders to the invitation to screen. Comparisons were made in the same way for pathological stage of cancer, based on data from three jurisdictions. The authors also used linkage to evaluate performance of the NBCSP iFOBT, Magstream HemSp with a rabbit serum buffer (Fujirebio Inc, Tokyo), to assess its test sensitivity, specificity and positive and negative predictive values for bowel cancer. The authors are to be congratulated on the study's rigorous design, the care taken to identify and exclude potential biases and their informative interpretation of results. The study showed that, of 12 689 people diagnosed with bowel cancer in 2006–2008, the never-invited group (10 080 cases) had a 15% higher risk of bowel cancer death by 31 December 2011 compared with NBCSP invitees (after correcting for lead-time bias). Of bowel cancer cases with “summary stage at first presentation” data (27% of total), cancers in the never-invited group had 38% higher odds of being more advanced than those diagnosed among NBCSP invitees. NBCSP invitees therefore had less risk of dying from bowel cancer, and were more likely to have less-advanced bowel cancers at diagnosis than non-invitees. Put simply: bowel cancer screening works. As well as achieving its principal goals, the NBCSP is doing so with results that would compare favorably with a range of other public health programs.5 In this respect, the study is an instructive companion analysis to reports that were promoted to make the case for the NBCSP's completion and recurrent funding.6 The study applied a different methodology to several earlier studies evaluating the NBCSP's effectiveness7-9 and reported even more compelling results. Its findings are broadly consistent with recent modeling that, even on modest participation rates, the NBCSP would prevent 70 000 premature Australian deaths over the next 3–4 decades.4 Another special feature of the study is the way linkage has been used to identify cancers in those with negative screening tests. Of the 1063 cases of cancer in the 2006–2008 participants, 887 (83.4%) were detected following a positive screening test. With proximal cancers, the test sensitivity was 77% and with distal cancers an impressive 89%. Linkage could not assess the contribution of advanced adenomas but test specificity with regard to cancer was 92.6%. The positive predictive value (PPV) for cancer was 3.6% and the negative predictive value was 99.9%. While not discussed in the article, likelihood ratios can be calculated from the data in Table 3 of the study. Of 297 378 participants with negative screening results, 176 (0.06%) had false negative results, cancer being diagnosed over the following 2-year period. With a PPV for cancer of 3.6%, the likelihood of having cancer when the test is positive therefore is 60 times greater than when negative. Although just a test for low-grade bleeding from the large bowel, the screening test used in the Australian NBCSP is remarkably effective. The most important risk factor for bowel cancer at a population level is ageing. On published trends, the ageing of Australia's population equates to a projected increase in bowel cancer incidence of between 30 and 40% every decade until the middle of this century.10, 11 Bowel cancer is already the most common internal cancer diagnosed in both men and women in Australia and the cause of the most cancer deaths after lung cancer.12 Improved survival from other diseases, and an increase in the prevalence of bowel cancer risk factors such as physical inactivity and high body mass, are likely to further increase relative bowel cancer incidence. This means the morbidity- and mortality-reduction benefits of screening will be magnified. A number of other nations in the Asia-Pacific region share these demographic trends13 and, therefore, would have the potential to benefit from an iFOBT-based bowel cancer screening program. It is important to note, of course, that the NBCSP was designed following a rigorous pilot study in three varied Australian settings,14 and that a key factor in its success is the relative accessibility of primary, secondary and tertiary follow-up care in Australia.15 Engagement of general practitioners in particular is critical to the program's success.14 In any new screening program, careful attention must be given to choice of the screening test. The guaiac FOBTs used in the randomized trials16, 17 have low sensitivity for cancer18 and increasingly are being replaced by iFOBTs.19 With iFOBTs, a range of commercial tests are now available but their performance is affected by differences in test methodology, the number of test samples, the choice of threshold for positivity with quantitative tests, susceptibility to the effects of heat and delayed analysis and other factors.20-24 The information about reliable test performance of the NBCSP screening test is especially welcome as measures such as false-negative rate for cancer often receive little attention in population-based screening programs.25 It should also be noted that one of the key indicators of the program's success is its cost-effectiveness. As an example of the cost and clinical differential in screening, the NBCSP is highly effective at detecting advanced adenomas (precancerous polyps),26 the diagnosis and removal of which costs on average less than $2000 in the Australian health system. By contrast, health-system costs to treat one case of advanced bowel cancer are around $100 000 and rising. Bowel cancer is Australia's most expensive cancer to treat, by a substantial margin, with most of the expenditure incurred in the hospital setting and for high-cost drugs to treat late-stage disease.27 So, while screening is a budget cost, it is also highly cost-effective. Of course, a principle of screening is ensuring that appropriate interventions are available for suspected early-stage disease that might be detected. There is little or no benefit in screening if diagnosis and treatment are not available. For this reason, the suitability for bowel cancer screening may vary among countries. However, as populations age, patterns of demography and disease change and health systems evolve, the potential benefits of a program similar to the NBCSP may become available to an increasing number of countries in our region. We expect more data, including a longer follow-up of the present study, to be published, reinforcing what we know now about the benefits of the NBCSP. Like any public health program, and screening programs in particular, the NBCSP is not perfect. But the evidence published in this issue of the APJCO is another compelling indication that the National Bowel Cancer Screening Program is sustainable and effective in meeting its objectives – to reduce the morbidity and mortality of bowel cancer.
No takes yet. Share an insight, caveat, or question.
John et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: