Evidence-based health information (EBHI) can support informed choice regarding whether or not to attend colorectal cancer (CRC) screening. The present study aimed to assess if German leaflets and booklets appropriately inform consumers on the benefits and harms of CRC screening.
A systematic search for print media on CRC screening was performed via email enquiry and internet search. The identified documents were assessed for the presence and correctness of information on benefits and harms by two reviewers independently using a comprehensive list of criteria.
Many of the 28 leaflets and 13 booklets identified presented unbalanced information on the benefits and harms of CRC screening: one-third did not provide any information on harms. Numeracy information was often lacking. Ten cross-language examples of common misinterpretations or basically false and misleading information were identified.
Most of the CRC screening leaflets and booklets in Germany do not meet current EBHI standards. After the study, the publishers of the information materials were provided feedback, including a discussion of our findings. The results can be used to revise existing information materials or to develop new materials that provide correct, balanced, quantified, understandable and unbiased information on CRC screening.
Citation: Dreier M, Borutta B, Seidel G, Münch I, Kramer S, Töppich J, et al. (2014) Communicating the Benefits and Harms of Colorectal Cancer Screening Needed for an Informed Choice: A Systematic Evaluation of Leaflets and Booklets. PLoS ONE 9(9): e107575. https://doi.org/10.1371/journal.pone.0107575
Editor: Hiromu Suzuki, Sapporo Medical University, Japan
Received: January 28, 2014; Accepted: August 20, 2014; Published: September 12, 2014
Copyright: © 2014 Dreier et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by the German Federal Ministry of Health. Further support came from internal funding sources supplied by the Institute for Epidemiology, Social Medicine and Health System Research. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have read the journal's policy and have the following conflicts. All authors have completed the ICMJE uniform disclosure at www.icmje.org/coi_disclosure.pdf and declare that UW, MLD, MD, BB, GS and IM had financial support from the Federal Ministry of Health, Germany for the submitted work; UW and MLD are members of the National Cancer Plan, Germany; JT works for the Federal Centre for Health Education (BZgA) which is a specialist authority within the portfolio of the Federal Ministry of Health, Germany; MLD, GS and IM received financial support from the Institute for Quality and Efficiency in Health Care (IQWIG); their spouses, partners, or children have no financial relationships that may be relevant to the submitted work. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
In the past, cancer screening procedures gained increasing significance for public health, and were promoted in many countries . However, in recent years, it became clear that the benefits of some procedures may have been overemphasised and their harms underemphasised . Potential harms of cancer screening include adverse effects from false-positive test results (mental stress, diagnostic evaluation), from the procedure itself, and from overdiagnosis –. As screening addresses healthy people who have a very small risk of ever developing the target disease, even very rare potential harms of the screening procedure are relevant. Thus, the individual should have the possibility to make an informed choice for or against the screening test , . In Germany, statutory health insurance covers colorectal cancer (CRC) screening tests, including the Guaiac-based faecal occult blood test (gFOBT) and screening colonoscopy, for persons aged 50 years and older, comprising about 85% of the population. In contrast to the FOBT, the colonoscopy is more a preventive than an early diagnostic screening test because endoscopic measures allow the detection and removal of CRC precursors (polyps, adenoma) . As colonoscopy is an invasive procedure , , special efforts are needed to promote informed decision-making. The informed choice to attend or not to attend CRC screening and which test to choose may be supported with appropriate evidence-based health information (EBHI) materials , . EBHI requires balanced, unbiased, quantified, understandable, and evidence-based information about CRC and both the benefits and harms of CRC screening , .
To estimate the extent of benefits and harms, numerical figures are essential. It is important to present the statistics in a comprehensible and unbiased manner , , . To begin with, probabilities should be presented as natural frequencies ,  because natural proportions are easier to understand than percentages (e.g. 80 out of 100 instead of 80%). A reference number (denominator) is needed to estimate the magnitude of effects. If possible, the same denominator should be used to ensure the comparability of the numbers. Regarding information on “lifetime” measures, the following messages are recommended: 1. When discussing risks, limited time frames (e.g. “in the next 10 years”) are easier to understand than “lifetime” figures . 2. When comparing the benefits of having a given screening test, use of the amount of absolute risk reduction (ARR) is preferred because relative risk reduction (RRR) tends to overestimate the effect –. For example: If five out of 1,000 screening participants develop CRC compared to 25 out 1,000 without screening, RRR = 80% (or: of 25 persons, only 5 out of 1,000 develop CRC = 80% less). In contrast, ARR = 20 out of 1,000 or 2%. This means that 20 out of 1,000 screening participants benefit from screening and do not develop the disease, compared to those not participating in screening. The temporal dimension of the risk reductions should also be indicated (for example: “1,000 people have an annually FOBT over 10 years”). Otherwise, essential information for benefit estimation would be lacking.
In Germany, information about CRC screening is provided by different healthcare stakeholders, including governmental organisations, foundations, healthcare providers, and health insurance companies. Some of the existing information materials do not meet EBHI standards, but rather focuses on the benefits of screening and/or strongly encourages participation in screening . In the context of the National Cancer Plan , the Federal Ministry of Health funded a CRC screening project to identify information materials that meet EBHI standards. We developed a list of criteria to assess whether print media like leaflets and booklets provide reliable, correct, understandable and unbiased information on CRC screening (gFOBT and colonoscopy) . This article reports whether and what kind of information on benefits and harms of CRC screening is contained in such leaflets and booklets in Germany. The aim is explicit not to discuss, which screening test to choose or to assess CRC screening strategies in Germany.
Materials and Methods
The study protocol was approved by the ethics committee of Hannover Medical School (Application No. 1803–2013).
The study methods included the following steps: 1) development of an assessment tool for expert use , 2) identification of consumer information materials on CRC screening, 3) assessment of the identified materials from an expert's view, and 4) assessment of the materials from the consumer's view. This paper reports the results from 2) and 3).
Identification of leaflets and booklets on CRC screening
In August 2010, stakeholders such as governmental organisations, foundations, healthcare providers, statutory health insurance companies, medical and scientific societies, and professional associations in Germany (excluding the pharmaceutical industry) were systematically identified and asked via email for leaflets and booklets on CRC screening. Information materials from the pharmaceutical industry were excluded due to potential conflicts of interest. A reminder email was sent two weeks later, if necessary. The websites of these institutions were also searched. The latest versions of available print media and online printable documents (PDF or MS Word format) on CRC screening (in German) addressing persons with an average risk of CRC were included in the analysis. Eligible materials had to provide information at least on the two CRC screening methods reimbursed by German statutory health insurances: gFOBT and colonoscopy. Information materials designed only to help persons determine their individual CRC risk, regional print media, or leaflets and booklets targeting CRC patients or people with an increased risk of CRC (e.g. persons with inflammatory bowel disease, familial adenomatous polyposis, hereditary non-polyposis colorectal cancer, or increased incidence of CRC in first- or second-degree relatives) were excluded. The provision of additional information on treatment options was no reason for exclusion.
Assessment of the identified health information
The information materials were assessed independently by two out of four reviewers (MD, BB, GS, IM) using an extensive list of 230 criteria  designed to evaluate information on the gFOBT and colonoscopy. Any discrepancies between the reviewers were solved by consens, if necessary, including a third person. The list is divided into four domains: A. Content issues, B. Formal issues, C. Presentation and understandability and D. Neutrality and balance (Table S1), representing a maximum of possible information. Not all of the criteria are essential for high-quality information. A and B criteria are rated multi-dimensionally, i.e. on reporting, correctness, presentation and evidence level. For each item an accompanying text input was possible that helps to make the reviewers' decisions transparent. C criteria were rated with “yes”, “rather yes”, “rather no”, “no”, “not applicable”. A manual supported the assessment by giving the correct answers based on systematic reviews, HTA reports, and guidelines. To illustrate examples for false or misleading information, these examples were translated into English language and reviewed by a certified translator who is a native English speaker. The original German wordings are provided in the supplement. Differences between leaflets and booklets were restricted to qualitative comparisons because the cell count was too small for sound statistical analysis.
Baseline risks, namely the risk of developing the target disease and dying from it, give an impression of the disease severity. These risks are derived from incidence and mortality rates (frequencies of getting or dying from a disease). The corresponding criteria on the list gather information on the precursors, incidence and mortality of CRC, including age- and sex-specific rates, comparisons with risks associated with other cancers, everyday risks, and the natural course of disease (Table 1).
The following parameters were included to assess the potential benefits of CRC screening:
They were assessed with respect to the benefits of CRC screening in general, and to the benefits of colonoscopy and the FOBT in particular (Table 2). Both qualitative and quantitative information (e.g. “screening prevents deaths from CRC” and “screening prevents death from CRC in 3 out of 1000 cases”, respectively) was collected. Quantitative information was further divided into ARR, RRR and the number needed to screen (NNS), which is calculated as the reciprocal value of the ARR.
The potential harms of CRC screening were examined with reference to screening colonoscopy and FOBT. Three main areas (modified from ) were considered:
- Harms of the screening test including those due to preparation, the procedure itself, and accompanying measures (anaesthesia/sedation)
- Risks of inaccurate test results, including false-positive results (that may lead to complications of follow-up, emotional distress, and lost work days) and false-negative results (that miss disease and thus cause a delay in treatment)
- Harms of overdiagnosis from true-positive identification of CRC or precancerous abnormalities (that would not have been harmful and led to death , , )
Numerical data in the leaflets and booklets were assigned to the corresponding criteria. Free text was added, citing the wording and information on whether the number was presented as a natural frequency, with a denominator and time frame. In addition, domain C provided aggregated criteria on the presentation of numerical data (e.g. the use of natural frequencies instead of percentages, reference values, and same denominators).
Seventy-two of the 142 stakeholders identified (50.7%) responded. In combination with the internet search, a total of 71 information materials were identified, 41 of which (28 leaflets and 13 booklets) met the inclusion criteria. A detailed list of the included materials is provided in the supplement (Table S2). The majority of leaflets were published by non-profit organisations such as foundations (17/28) and state ministries (6/28), while the booklets were mainly published by scientific societies (6/13).
Information on baseline risks
Four of five materials explained the meaning of precursors for the development of CRC, but information on the frequency of these polyps or adenomas was clearly less often given (Table 1). We identified some false statements suggesting, for example, that many of the precursors become cancerous, though this is only true in 5–15% of cases (Table 2, Example 1) , . Information on CRC incidence was presented in 61% of the leaflets (17/28) and 85% of the booklets (11/13), but mainly without stratification by sex and age. Information on CRC mortality was included in nearly two-thirds of the leaflets (18/28) and half of the booklets (6/13). Information on the remaining lifetime risk was not given and only two leaflets and two booklets mentioned age-related risks of developing CRC and/or dying from CRC. More frequently, the information materials compared CRC risks to other risks. In most cases, this was done by describing CRC as the second most common cause of developing or dying from cancer out of all types of cancerous diseases. Only a minority of materials described the natural progression of the disease.
In 11/28 leaflets and 4/13 booklets, we identified a common source of misleading information, which arises from describing CRC incidence and mortality of CRC in a single sentence (Table 2, Example 3). As a result, the reader may get the impression that about one-third of the newly diseased will die within one year and thus will overestimate the case fatality rate of the disease. This example also shows that we accepted the absolute number of newly diseased and the absolute number of cases of death as incidence and mortality even though the correct number should include a reference population as a denominator. Further results on the rating of numerical data are described in detail below.
Information on the benefits of CRC screening
Most of the leaflets and booklets described general benefits of CRC screening in terms of decreasing the incidence and mortality of CRC, for example: “cancer can be prevented”, “cancer is curable if detected early”, “screening reduces deaths caused by CRC” (Table 3). In contrast, the specific benefits of a given screening test were clearly less frequently mentioned. While about one-third of the booklets mainly indicated the positive effect of the FOBT or colonoscopy on the CRC-mortality, leaflets rather focused on the reduction of CRC incidence by colonoscopy.
False messages were rare. Statements, such as “cancer disease that can be prevented in 100% of cases” were rated as false. Table 2 shows further important examples for basically false and misleading information on screening benefits. Example 4 claims that it is an advantage for everyone to attend screening, neglecting to state that screening is only useful for individuals who actually develop CRC. Example 5 is also misleading because it mixes information about screening versus diagnostic examinations, which may increase widespread false beliefs about screening, especially in those who refuse screening because they have no symptoms.
Information on the harms of CRC screening
Harms attributed to the screening procedure.
Unlike colonoscopy, there are no direct adverse effects attributed to the FOBT test. Those associated with colonoscopy are divided into harms that occur during colonoscopy preparation, sedation, and the colonoscopy procedure itself (Table 4). Only 3 out of 41 (7%) information materials indicated any harms due to the preparation or sedation. Harms of the colonoscopy procedure itself were reported more frequently. Half of the leaflets and nearly two-thirds of the booklets gave information on common risks, for example, by characterising colonoscopy as being a “low-risk” procedure or having “few complications”, and causing possible pain. Information on further adverse effects like bleeding, infection, and perforations was included less often. Only one leaflet and two booklets indicated the risk of death. More than one-third of the leaflets (10/28) and one-fifth of the booklets (2/11) did not mention harms at all.
In some cases (especially leaflets) the risk reporting was rated as being incorrect. Most frequently, the colonoscopy procedure was falsely characterised as being “pain-free” or “non-painful”. Common statements describing colonoscopy as being “without any problems” or “safe” were likewise considered false. Some materials were rated as having excessive or very low estimates of the frequencies of possible adverse effects. For further examples, see Table 2, No. 6–8.
Harms attributed to inaccurate tests.
Statements on the quality of the screening tests were assessed (Table 5). About two-thirds of the leaflets and booklets provided general information on the test accuracy of colonoscopy. However, this information was misleading in half of the leaflets, as shown in Table 2, Example 9: Colonoscopy was sometimes described as being “ the safest (German: sicherste) way to prevent CRC” or as “providing the highest safety (German: Sicherheit)”. This obviously refers to the high sensitivity and specificity of the colonoscopy in the sense of being “reliable”, but may be misinterpreted as meaning that the procedure is completely safe and without any possible adverse effects.
As for the FOBT, general information on test accuracy was given in one-third of the leaflets and in nearly two-thirds of the booklets, often in comparing the FOBT to colonoscopy as “less reliable” or “not sufficiently reliable”. Accuracy criteria for the FOBT test were indicated more often than those for colonoscopy, especially sensitivity and negative predictive value data. Table 2, Example 10 shows a false and very misleading description of false-negatives, which are presented as negative predictive values, giving the reader the impression, that someone with a negative FOBT has a high risk of CRC.
Regarding the information on test properties, predictive values give consumers the best information because they display the actual proportions of correct or false results out of all positive or negative results. Thus, these values represent the consumers' view and qualify as patient-oriented information. However, such information was provided for the FOBT in only about half of the identified materials, and even less often for the colonoscopy.
Overall, about half of the leaflets and booklets presented numerical data as natural frequencies, and one quarter of these materials used the same denominator within one subject (Figure 1).
Results from the criteria characterising the quality of numerical data: 1. Natural frequencies instead of percentages are used. 2. Same denominators are used.
More detailed analysis revealed that most of the information materials (21/28 leaflets, 12/13 booklets) included some quantification of baseline risks in terms of incidence and mortality. However, these numbers were mainly presented as absolute numbers of cases in Germany without a reference population and without a proportion such as 100 out of 100,000.
Only some of the leaflets and booklets provided quantitative information on benefits and harms. Regarding colonoscopy, three leaflets and one booklet describing the RRR for CRC incidence reported, for example, that “80% of all CRC is preventable”. We raise two points of criticism against such statements. First, presenting numerical data as natural frequencies (e.g. 80 out of 100) is considered more comprehensible than percentages , . Second, failing to indicate the ARR along with the RRR leads to an enormous exaggeration of the benefit , , .
The positive effects of colonoscopy on the CRC mortality were quantified in six information materials (15%), four of which correctly indicated, in accordance with the current evidence base, that the extent of benefit is currently unknown. The other two only described the RRR without giving the ARR (for example: “according to experts, early colonoscopy could save more than three-quarters of …”). Regarding the FOBT, estimates of the reduction of mortality are available from meta-analysis of randomised controlled trials, providing the highest level of evidence . These figures were found in less than 10% of the identified information materials (3/41), most of which described the amount of ARR rather than RRR (for example: “1 to 3 out of 1,000 persons less die with screening than without screening”). One material reported the number needed to screen (NNS) to prevent one case of death caused by CRC.
Information on the harms of the screening colonoscopy seldom included numerical data. Quantitative information on the following was provided in leaflets versus booklets: common adverse effects (3/28 vs. 4/13), bleedings (1/28 vs. 3/13), perforations (2/28 vs. 3/13), and mortality (1/28 vs. 1/13). Most of these numbers were comprehensibly presented as natural frequencies.
Balance of benefits and harms
Whereas almost all leaflets and booklets provided information on the general benefits of CRC screening, only about half of the leaflets and booklets described the general harms of the colonoscopy (Figure 2). The reverse is true for specific benefits and harms, whereby specific harms were presented more often than specific benefits in the case of both colonoscopy and FOBT. Overall, specific information was more often provided in booklets than in leaflets.
The figure indicates whether a leaflet contains any information on the benefits of CRC screening a) in general or specifically for b) colonoscopy or c) the FOBT, and whether it contains any information on the d) general and e) specific harms of colonoscopy. To be rated positive for harms, it was not sufficient if the only information about possible harms referred to pain, stating incorrectly that there is no pain involved.
Thirteen German booklets and 28 leaflets on CRC screening were identified. Their content and correctness of reporting the benefits and harms of CRC screening was assessed using a comprehensive list of criteria. We found that the standards of evidence-based health information were met only partially. First, information on benefits and harms was presented in an unbalanced manner, generally in favour of the benefits; however specific harms were reported more often than specific benefits. The vast majority of the materials provided at least qualitative information on potential benefits, whereas nearly one-third of the materials did not mention potential harms associated with CRC screening at all. Quantitative information about benefits and harms was mostly lacking. The numerical data presentations showed deficits in understandability and were partly misleading in favour of screening. Some materials contained false and/or distorted information.
We used a two-way approach to searching for leaflets and booklets: written email enquiries, and systematic searches of the websites of a broad range of stakeholders. Fifty percent of the stakeholders responded, which might be because not all of them provide information materials on CRC screening. As we additionally searched the websites, we are rather confident that we found almost all leaflets and booklets that met the inclusion criteria. The reviewers were not blinded to the identity of producers of the information materials. Thus, ratings could be biased in favour of the leaflets and booklets from scientific organisations, but this was probably rare, as false or misleading information was found in nearly all of the leaflets and booklets. As materials provided by the pharmaceutical industry were excluded, our overview of the quality of leaflets and booklets might not be representative of all CRC screening leaflets and booklets in Germany. This might be especially true for certain screening tests (e.g. immunological stool tests) that are not reimbursed by statutory health insurance companies. Despite the extremely high number of criteria assessed, our list was not entirely complete. First, we focused on traditional outcomes and did not use quality of life (QoL) to characterise the benefits of CRC screening in supplement to CRC incidence and mortality data . However, QoL was partially taken into account in the context of describing adverse effects of the screening procedure. Thus, QoL criteria will be added to the list as parameters of patient-related outcomes . Second, as we focused on tests reimbursed by German statutory health insurance companies, flexible sigmoidoscopy was not included though there is highest-level evidence indication that it reduces CRC incidence and mortality . However, our aim was to assess the information materials; to evaluate the current screening strategies in Germany is of high importance but would go beyond the scope of this study. Third, we collected information on the quality of the screening tests but not on the consequences of false-positive or -negative test results, including physical or psychosocial harms from the test results as well as unnecessary follow-up procedures . These consequences will be acknowledged in detail in future information material assessments.
A quality assessment that goes beyond previous tools by considering the correctness of information was used in the present study. To our knowledge, this study on the quality of written information materials provides the most thorough and in-depth evaluation of reported benefits and harms of CRC screening available. The use of a comprehensive list of criteria enabled us to detect missing information in detail. However, it is still unclear which information should be included in either brief or extensive information materials because the criteria list reflects the maximum content requirements. The main strength of this approach, apart from its comprehensiveness, is its ability to rate the correctness of information. Previous tools, which focus on criteria that characterise structural and process quality as surrogate parameters for content quality, might not always give true results . Other tools for the assessment of content quality  do not assess the correctness of information and might even rate false information as being of good quality. The amount of false information detected in the leaflets and booklets underscores the need to directly assess the correctness of such information. Ten examples of common misinterpretations or basically false and misleading information were presented in this paper. Although these examples originate from materials written in German, they have cross-language validity and thus might be very helpful for health information providers worldwide. For example, when searching the National Cancer Institute's website for information about colorectal cancer screening , we found the simultaneous reporting of CRC incidence and mortality data in one sentence. In the present study, we showed that this may lead to exaggeration of the anticipated CRC case fatality rate (Table 2, Example 3).
Many of the leaflets and booklets must be revised to comply with EBHI standards. Thus, most of the information materials do not meet the prerequisites for informed choice. The identified deficits clearly indicate that the principles of good communication practice have not been adopted by the producers of information leaflets and booklets on CRC screening . Obviously, the effects of promotional media campaigns that focus on the benefits of CRC screening are still predominant, while the new political strategy of promoting informed choice in an unbiased manner has caught these consumer information producers unprepared. Besides, people might be afraid that an informed choice will decrease participation in screening. High participation rates are necessary to support the quality of screening programs and to be able to measure benefits on a population level. From an ethical point of view, the informed choice is prior to the aim of increasing the uptake . However, the effect of informed-decision making in screening remains unclear as studies found either increased, or decreased attendance, or no effects –.
There is an urgent need to address the identified shortcomings. As a first step, the results of this study were already presented to different providers of the evaluated leaflets and booklets at a workshop organised by the German Federal Centre for Health Education (BZgA) to. In addition, individual rating results were submitted to the providers in writing. The public may not be prepared for EBHI that includes information on harms as well as numerical data or statements regarding the uncertainties of research results . Moreover, not all people are capable of making an informed choice. Therefore, special effort is needed to develop information materials for people with different levels of health literacy , . Further effort is needed to improve and implement current knowledge about the production of evidence-based health information and to establish minimum content requirements to enable informed choices.
EBHI on benefits and harms may support an informed decision making ,  regarding whether or not to attend CRC screening , . Our list of criteria examines whether health information meet EBHI standards, but cannot directly assess whether it is sufficient to support informed choice . The criteria list is used together with a manual that provides correct answers in order to minimise subjectivity and to facilitate the rating process. Challenges of the manual include the need for large time resources for regular updates to incorporate the latest evidence, as well as conflicts arising from different interpretations of the current evidence, for example, when experts disagree on the actual numbers characterising the benefits and harms in a leaflet on breast cancer screening . The criteria list might also be used to revise existing information materials or to produce new ones. The rating principle should be adapted when assessing screening information for other tests or types of cancer.
We would like to thank the panel of external reviewers who supported the development of our list of criteria: Dr. Berndt Birkner, Prof. Dr. Eckhard Breitbart, Dr. Regine Hagmann, Prof. Dr. David Klemperer, Dr. Klaus Koch, Prof. Dr. Jürgen Riemann, and Petra Uschold. We acknowledge support by Deutsche Forschungsgemeinschaft. We thank Suzyon O'Neal Wandrey who critically reviewed our translations as well as main parts of the manuscript.
Conceived and designed the experiments: UW MLD JT. Performed the experiments: MD BB GS IM SK. Analyzed the data: MD BB. Contributed reagents/materials/analysis tools: MD BB GS IM SK JT MLD UW. Wrote the paper: MD BB UW. Reviewed data analyses and interpretation: UW MLD JT. Read and approved the final manuscript: UW, MLD, and JT. Participated in interpretation of the results: MD BB GS IM SK JT MLD UW. Critically revised subsequent versions of the paper for important intellectual content: MD BB GS IM SK JT MLD UW.
- 1. Raffle A, Muir Gray JA (2007) How screening started. In Screening - Evidence and Practice Oxford University Press. pp. 1–30.
- 2. Woloshin S, Schwartz LM, Black WC, Kramer BS (2012) Cancer screening campaigns–getting past uninformative persuasion. N Engl J Med 367: 1677–1679.
- 3. Eckstrom E, Feeny DH, Walter LC, Perdue LA, Whitlock EP (2013) Individualizing cancer screening in older adults: a narrative review and framework for future research. J Gen Intern Med 28: 292–298.
- 4. Elmore JG, Fletcher SW (2012) Overdiagnosis in breast cancer screening: time to tackle an underappreciated harm. Ann Intern Med 156: 536–537.
- 5. Jorgensen KJ, Zahl PH, Gotzsche PC (2009) Overdiagnosis in organised mammography screening in Denmark. A comparative study. BMC Womens Health 9: 36.
- 6. Austoker J, Giordano L, Hewitson P, Villain P (2012) European guidelines for quality assurance in colorectal cancer screening and diagnosis. First Edition - Communication. Endoscopy 44 Suppl 3SE164–SE185.
- 7. General Medical Counsil (2008) Consent: patients and doctors making decisions together. Available: http://www.gmc-uk.org/guidance/ethical_guidance/consent_guidance/index asp. Accessed 2014 Jan 12.
- 8. Bretthauer M, Kalager M (2013) Principles, effectiveness and caveats in screening for cancer. Br J Surg 100: 55–65.
- 9. Pox CP, Altenhofen L, Brenner H, Theilmeier A, Von SD, et al. (2012) Efficacy of a nationwide screening colonoscopy program for colorectal cancer. Gastroenterology 142: 1460–1467.
- 10. Patera N (2010) Screening for colorectal cancer. Part 1: screening-tests and program design, HTA project.
- 11. Coulter A, Ellins J (2007) Effectiveness of strategies for informing, educating, and involving patients. BMJ 335: 24–27.
- 12. Bunge M, Mühlhauser I, Steckelberg A (2010) What constitutes evidence-based patient information? Overview of discussed criteria. Patient education and counseling 78: 316–328.
- 13. Jorgensen KJ, Brodersen J, Hartling OJ, Nielsen M, Gotzsche PC (2009) Informed choice requires information about both benefits and harms. J Med Ethics 35: 268–269.
- 14. Gigerenzer G, Gassmaier W, Kurz-Milcke E, Schwartz L, Woloshin S (2008) Helping Doctors and Patients Make Sense of Health Statistics. Psychological Science in the Public Interest 8: 53–96.
- 15. Gigerenzer G, Edwards A (2003) Simple tools for understanding risks: from innumeracy to insight. BMJ 327: 741–744.
- 16. Gigerenzer G (2011) What are natural frequencies? BMJ 343: d6386.
- 17. Akl EA, Oxman AD, Herrin J, Vist GE, Terrenato I, et al. (2011) Using alternative statistical formats for presenting risks and risk reductions. Cochrane Database Syst Rev CD006776. 10.1002/14651858.CD006776.pub2 [doi].
- 18. Nuovo J, Melnikow J, Chang D (2002) Reporting number needed to treat and absolute risk reduction in randomized controlled trials. JAMA 287: 2813–2814.
- 19. Steckelberg A, Balgenorth A, Muhlhauser I (2001) [Analysis of German language consumer information bookllets on screening for colorectal cancer]. Z Arztl Fortbild Qualitatssich 95: 535–538.
- 20. National Cancer Plan Germany (2012) Aktueller Stand des Nationalen Krebsplans. Handlungsfelder, Ziele und Umsetzungsempfehlungen. Stand: 04. Januar 2012. Available: https://www.webgate.ec.europa.eu/sanco/heidi/images/2/23/Germany_Working_Document_on_NCP_German_4.1.2012.pdf. Accessed 2014 Jan 12.
- 21. Dreier M, Borutta B, Seidel G, Kreusel I, Toppich J, et al. (2013) Development of a comprehensive list of criteria for evaluating consumer education materials on colorectal cancer screening. BMC Public Health 13: 843.
- 22. Welch HG, Black WC (2010) Overdiagnosis in cancer. J Natl Cancer Inst 102: 605–613.
- 23. Winawer SJ, Zauber AG, Ho MN, O'Brien MJ, Gottlieb LS, et al. (1993) Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 329: 1977–1981.
- 24. Hewitson P, Glasziou P, Watson E, Towler B, Irwig L (2008) Cochrane Systematic Review of Colorectal Cancer Screening Using the Fecal Occult Blood Test (Hemoccult): An Update. Am J Gastroenterol 103: 1541–1549.
- 25. Elmunzer BJ, Hayward RA, Schoenfeld PS, Saini SD, Deshpande A, et al. (2012) Effect of flexible sigmoidoscopy-based screening on incidence and mortality of colorectal cancer: a systematic review and meta-analysis of randomized controlled trials. PLoS Med 9: e1001352.
- 26. Heleno B, Thomsen MF, Rodrigues DS, Jorgensen KJ, Brodersen J (2013) Quantification of harms in cancer screening trials: literature review. BMJ 347: f5334.
- 27. Khazaal Y, Chatton A, Zullino D, Khan R (2012) HON label and DISCERN as content quality indicators of health-related websites. Psychiatr Q 83: 15–27.
- 28. Elwyn G, O'Connor AM, Bennett C, Newcombe RG, Politi M, et al. (2009) Assessing the quality of decision support technologies using the International Patient Decision Aid Standards instrument (IPDASi). PloS one 4: e4705.
- 29. National Cancer Institute. Tests to detect colorectal cancer. 1. What is colorectal cancer. Available: http://www.cancer.gov/cancertopics/factsheet/detection/colorectal-screening. Accessed 2014 Jan 12.
- 30. van Dam L, Korfage IJ, Kuipers EJ, Hol L, van Roon AH, et al. (2013) What influences the decision to participate in colorectal cancer screening with faecal occult blood testing and sigmoidoscopy? Eur J Cancer 49: 2321–2330.
- 31. Steckelberg A, Hulfenhaus C, Haastert B, Muhlhauser I (2011) Effect of evidence based risk information on “informed choice” in colorectal cancer screening: randomised controlled trial. BMJ 342: d3193.
- 32. Smith SK, Trevena L, Simpson JM, Barratt A, Nutbeam D, et al. (2010) A decision aid to support informed choices about bowel cancer screening among adults with low education: randomised controlled trial. BMJ 341: c5370.
- 33. Marteau TM, Mann E, Prevost AT, Vasconcelos JC, Kellar I, et al. (2010) Impact of an informed choice invitation on uptake of screening for diabetes in primary care (DICISION): randomised trial. BMJ 340: c2138.
- 34. Wagner CV (2011) A decision aid to support informed choice about bowel cancer screening in people with low educational level improves knowledge but reduces screening uptake. Evid Based Nurs 14: 36–37.
- 35. Hirschberg I, Seidel G, Strech D, Bastian H, Dierks ML (2013) Evidence-based health information from the users' perspective - a qualitative analysis. BMC Health Serv Res 13: 405.
- 36. Smith SK, Trevena L, Barratt A, Dixon A, Nutbeam D, et al. (2009) Development and preliminary evaluation of a bowel cancer screening decision aid for adults with lower literacy. Patient Educ Couns 75: 358–367.
- 37. Hawkes N (2011) Breast cancer screening is to be reviewed, cancer tsar announces. BMJ 343: d6905.