Geir Hoff and Michael Bretthauer are at the Institute of Population-Based Cancer Research, Oslo, Norway. Geir Hoff is also at the Telemark Hospital, Skien, Norway. Michael Bretthauer is also at the Department of Medicine, Rikshospitalet University Hospital, Oslo, Norway.
Published online 2006 January 3. doi: 10.1371/journal.pmed.0030036.
Politicians in developed countries have secured a certain standard of health care for the sick as well as vaccination programmes to reduce the risk of illness. For some of these countries, cancer screening is next on the agenda, with breast and cervical cancer screening programmes at the top of the list. When it comes to colorectal cancer (CRC), randomised trials of screening with the faecal occult blood test (FOBT) have shown that such screening reduces CRC incidence [1] and mortality [1–3]. Indirect evidence suggests that endoscopic screening (fl exible sigmoidoscopy and colonoscopy) may be far more effective, but results from large-scale randomised trials are still awaited [4–7].
In the United States, CRC screening, using any of these screening modalities, has been recommended for many years [8–10]. The European Union now encourages its member states to initiate CRC screening, provided that it is done in an organised fashion and linked to quality assurance programmes [11]. Although the EU guidelines point out that only FOBT screening has been proven in randomised trials to have an effect on CRC, some member states (Poland, Germany, and Italy) have gone straight for “gold standard” colonoscopy screening without evidence from randomised trials [12,13].
The ideal screening method has not yet been found, and the future may see more use of promising screening modalities such as stool-based DNA tests [14] and virtual colonoscopy [15]. At present, however, the fact that these newer methods are still at an early stage of development may only serve as an excuse for politicians to “wait and see”, and thus further postpone a decision on CRC screening. In this essay, we discuss the scientifi c and political controversies surrounding the introduction of a national CRC screening programme. We argue that it is in the best interest of governments to fund well-conducted randomised trials that can guide their decision making.
Public Awareness of CRC
Enthusiastic health professionals often find politicians and health authorities too slow to respond to the considerable evidence in favour of CRC screening [13], and they may take on a political role themselves in promoting their good cause. In addition, a number of celebrities—including Ronald Reagan, Katie Couric (an American television personality), and Lynn Faulds Wood (a British television presenter)—have helped to raise public awareness of CRC ([16]; http:⁄⁄www.bowelcancer. tv/cgi-bin/page.pl). His Holiness, the late Pope Paul John II, promoted CRC screening by being the first patron of the International Digestive Cancer Alliance, with strong support from United States Senator Hillary Clinton [17].
But while there is growing public awareness of CRC, there has been disturbingly little public discussion about the scientifi c pros and cons of CRC screening compared with what some countries have seen for breast cancer screening. One of the main reasons for nonparticipation in CRC screening has been that people are not convinced that they will gain anything by participating [18]. Similarly, the failure of some physicians to recommend screening may refl ect a general need for more convincing evidence and a convincing presentation of such evidence. Women being offered mammographic screening for breast cancer have expressed a particular need for information about logistics as well as about false positive and negative tests, and they want to play an active part in decision making on screening [19]. This need for balanced information on screening applies to other types of cancer screening and should be the “standard of care” for future cancer screening awareness campaigns.
By failing to recommend screening, physicians representing the medical profession may appear divided in their view to screen or not to screen, and undecided on which method to use, although the health problem of CRC is recognised by all. In many countries, CRC is one of the two most frequently encountered cancers for men and women collectively. Five-year survival is only about 50%, increasing to 80% if diagnosed at an early, asymptomatic stage [20]. Thus, early diagnosis and treatment is the obvious key to improved cure and survival. Some enthusiasts initially recommending national screening programmes with continuous quality assurance now accept opportunistic screening (i.e., screening done outside of a national screening programme), since “public health policy has not yet included population CRC screening” [13]. On the other hand, many doctors feel that there are still important questions to be answered before implementing national screening programmes. Besides, some are concerned that not enough is being done for patients who need basic health care rather than preventive measures. How Good Is the Evidence for Screening? So what kind of evidence can we offer to increase professional enthusiasm for CRC screening? Do we only have the support of celebrities and little else?
FOBT is the only screening method submitted to adequately designed randomised trials, which showed 15%–33% mortality reduction from CRC after 8–13 years’ follow-up [1–3]. The relative mortality reduction was signifi cant in all three trials, proving without doubt that “FOBT screening does work”. But how good is this in absolute, not relative, terms? With 5% lifetime risk of CRC and 50% mortality from the disease, the risk of dying from CRC is 2.5% without FOBT screening. With biennial FOBT screening, this may be reduced by 23% for those attending, i.e., reduced by 0.6% to 1.9% if attendance is 100%. About 98% of us will die from something else. How do you sell such figures both to doctors and to the public?
In contrast to FOBT, we only have case-control studies to support the usefulness of endoscopy screening, and such studies are particularly prone to overestimate the effectiveness of screening tests when compared with randomised trials [21]. The first reports from the four ongoing randomised trials on fl exible sigmoidoscopy screening are expected over the course of the next several years [4–7], while similar randomised trials on colonoscopy screening have not, so far, been launched. Mortality reduction by colonoscopy screening is expected, but not proven, to be much higher than the 23% for those attending FOBT screening, in addition to an incidence reduction expected through polypectomies.
There has been some concern about a possible excess number of non-CRC deaths in the screening group in CRC screening trials [22,23]. In addition, a recent 17-year follow-up of the Danish FOBT study showed that the signifi cant reduction in mortality seen after about ten years could no longer be observed, probably due to poorer attendance with an increasing number of screening rounds [24]. Since high attendance rates are crucial for the success of screening with the FOBT, the best choice among the present CRC screening options may differ between countries and cultures, thus indicating a need for research on CRC screening referable to the target population. Furthermore, there is only limited knowledge about what impact an increasing menu of screening services might have on the responsibility people take for their own health.
