Financial incentives for breast cancer screening undermine informed choiceBMJ 2022; 376 doi: https://doi.org/10.1136/bmj-2021-065726 (Published 10 January 2022) Cite this as: BMJ 2022;376:e065726
- Theodore Bartholomew, GP specialty registrar1,
- Mirela Colleoni, patient2,
- Harald Schmidt, assistant professor3
- 1Royal Surrey County Hospital NHS Foundation Trust, Guildford, UK
- 2Hauts-de-France, France
- 3Department of Medical Ethics and Health Policy, University of Pennsylvania, USA
- Correspondence to: T Bartholomew
Breast cancer is the most common cause of cancer death for women worldwide.1 Various measures have been explored to reduce breast cancer mortality. One approach is to encourage screening through financial incentives for patients or providers, as recently proposed in France and the UK, respectively.234 Several countries already use some form of incentive, and the extent to which they are being offered, in combination with ambitious screening targets, could suggest that their use is an appropriate way to promote population health. This, however, is far from clear: there are major concerns about breast cancer screening, the effectiveness of financial incentives is unclear, and there is an urgent need to ensure that women have given their valid consent—that is, informed and without undue influence—for screening. Some trans men and women and non-binary people are also eligible for breast cancer screening, and although the data mostly relate to cisgender women, the arguments on informed consent apply to everyone.
Health effects of breast cancer screening
Breast cancer screening is controversial on several grounds. For every woman who avoids a breast cancer death through screening, 3-10 women will be treated unnecessarily and over 200 will experience psychological distress because of false positive results (table 1).567 Although a 2012 UK review stated that screening conferred “significant benefit and should continue,” with one breast cancer death averted for every 235 women invited to screening over 20 years,5 its conclusions have been widely challenged.8910 Major concerns include the harms of overdiagnosis (box 1), substantial uncertainty over cost effectiveness,11 and that most reductions in breast cancer mortality can be attributed to improved breast cancer awareness and the use of adjuvant hormonal and chemotherapy.61213 A Cochrane review concluded that universal mammography screening should be reassessed given the “small at best” chance of benefit.6
Overdiagnosis of breast cancer
Overdiagnosis is defined as “the detection of cancers on screening which would not have become clinically apparent in the woman’s lifetime in the absence of screening.”5
Although the scientific modelling can be complex, in broad terms overdiagnosis is known to exist because in randomised studies that compare cancer incidence in cohorts that participate in screening and those that do not, there are cumulatively always more cancer cases in the screened cohort when, because of randomisation, they should be about the same.
Overdiagnosis is problematic because the lethality of screen-detected tumours is not always clear at the time of diagnosis and thus a breast cancer diagnosis almost always leads to treatment.
In the NHS, 99% of women diagnosed with breast cancer through screening have surgery, 72% have radiotherapy, 72% have adjuvant hormone therapy, and 27% have adjuvant chemotherapy.5
An overdiagnosis means that a woman receives interventions providing no benefits but entailing risks. The psychological distress of a breast cancer diagnosis and treatment have been well documented, and matter ethically because screening programmes knowingly cause a proportion of women who would never have been harmed from breast cancer to worry that they might be.56RETURN TO TEXT
Targets, rates, and incentives
Recommendations for breast cancer screening attract considerable political and public attention, and there have been calls for more open minded discussion and public education.14 Breast cancer screening rates vary widely between countries, but many nations with previously high screening rates have seen a decline,151617 which has partly been attributed to increased awareness of the harms of screening.18 Screening rates consistently lag behind targets set by health authorities: few European Union countries achieve the bloc’s screening target of 70%,19 and the latest figures in England and the US show that screening rates are below nationally set (“achievable”) targets.2021 In response, some countries have introduced or proposed financial incentives to increase screening rates.
In England, for example, the Independent Review of Adult Screening Programmes has recommended that NHS England should “urgently” consider using financial incentives for providers to increase cancer screening rates. Suggestions include introducing payment by activity for general practitioners (more money for more people screened) and targeted payments for enhanced services (giving a practice extra funding for providing additional screening services).15
In France, the National Institute of Cancer has adopted various measures to increase screening rates.22 These include trials of patient incentives and covering screening related expenses such as childcare, transport, and work.22 Breast cancer screening rates have been included as one of 29 clinical indicators that can provide GPs with additional remuneration since 2011, despite acknowledgment that the policy is controversial because of the known harms of screening.23 Financial incentives are also used elsewhere in Europe, including Croatia and Portugal.3
In the US, private health insurers encourage take up of breast cancer screening with paid time off, items in kind (T shirts, cinema tickets, etc), and financial incentives such as reduced insurance premiums, gift cards, or lotteries.4 Although lack of reporting requirements means that the frequency of use and size of incentives are unknown, they are typically $10-$200 (£7-£150; €9-€180).4 Incentives are also offered to women on low incomes in public insurance programmes in the US: four states offer incentives ranging from around $0.50 to $540, and two of these even offer incentives for yearly screening (more frequently than advised by the US Preventive Services Taskforce because of the large increase in harms; table 1).724 Breast cancer screening is also one of the most common indicators in pay-for-performance programmes for US physicians.25
Effectiveness of financial incentives
Whether and how financial incentives influence patient and physician behaviour remains unclear. Although patient financial incentives can increase uptake of some types of preventive care,262728 their effectiveness within cancer screening and, more specifically, in breast cancer screening is mixed (table 2).26293031323334 Internationally, the evidence for provider incentives within cancer screening is also “mixed,” with most studies on breast cancer showing partial or no effects on uptake.35
Ethics of financial incentives
The use of incentives requires consideration of ethical issues. The case in favour of incentives rests on behavioural economics. Motivation to act on long term health goals, even if much wanted, may be outweighed by short term needs, desires, or conflicting priorities. Failed New Year’s resolutions and abandoned weight loss regimens are familiar examples. One rationale for offering incentives for health promotion is that they nudge people to act by offering an immediate (financial) reward for a later health benefit.36 Consider the example of financial incentives for patients to stop smoking. Such incentives may be perceived as paternalistic and possibly exerting undue influence,2437 but since most smokers would like to quit,37 incentives that successfully motivate smoking cessation can be seen as enhancing autonomy. Not smoking also has overwhelming health benefits and little risk of harm.
Provider incentives have been used when policy makers think they can motivate professionals to adapt their behaviour in ways that better align with health system priorities. For example, provider incentives have been implemented to reduce antibiotic overprescribing38 and improve hand hygiene.39
The fundamental ethical concern with incentives is that they may lead people to make choices that they would not have made without the incentive and are harmed by this choice. Smoking cessation, reducing antibiotic overprescribing, and improving hand hygiene differ fundamentally from breast cancer screening because the benefits of these actions are objectively and substantially greater than the possible harms. Although there are, in principle, good reasons for exploring the potential of incentives to promote breast cancer screening, the starting position for breast cancer screening has to be that the decision is preference sensitive40—that is, equally well informed women may weigh the trade-off between harms and benefits differently and rationally may or may not decide to have screening.41
If incentives become unduly influential, they can undermine the principle and validity of consent.2437 Breast cancer incentives could also compromise the extent to which a provider offers unbiased information on harms and benefits when seeking patient consent. For example, financial gains for higher screening rates might lead providers to change, consciously or unconsciously, the time they spend discussing benefits and harms of screening with patients. This can be detrimental to shared decision making and genuine consent, and matters against the backdrop that in the US, for example, less than 10% of patients report that their physician informed them about overdiagnosis when discussing cancer screening.42 The presence of a target screening rate further exacerbates this problem, with providers likely to feel pressure to meet targets.
Understanding screening and communicating risk
Across society, understanding of breast cancer screening is generally poor. Most women overestimate the mortality benefits of screening (92% of those surveyed in nine European countries overestimated the mortality reduction from screening by a factor of 10-200 or reported that they did not know)43 and most are unaware that inconsequential disease can be detected by screening.41 Such findings are perhaps unsurprising given that organisations that promote screening sometimes overstate benefits and underplay harms.644
The scientific community agrees that people should be given better information about screening.45 The fact that people with adequate knowledge of the overall benefit, false positive results, and overdiagnosis may be less likely to choose breast cancer screening,41 and that those with an understanding of overdiagnosis may be more likely to express an intent to discontinue screening, further underlines this need.46 Additional research, however, is needed to clarify how understanding affects longer term participation.47
Doctors can also have inadequate understanding of screening. In a survey of over 400 US primary care doctors in 2011-12, almost half incorrectly answered that detection of more cancers proves that the screening test reduces mortality (more detection could just mean more overdiagnosis). 48 Three quarters also answered that better five year survival among patients with screen detected cancers than their non-screened comparators would prove that the screening test reduces mortality (when it could be the result of lead time bias).48
Effective communication is also a concern. Among 151 UK GPs assessed with a hypothetical patient seeking advice on cancer screening, only 44% were deemed to communicate “complete and meaningful” information on risk.49 Physicians should have a robust understanding of the interventions they offer and be able to communicate this to patients in a way they understand, but this does not seem to be the case.
Decision aids to support informed choice
Ideally information provided on the harms and benefits of screening should be individualised to a person’s specific risk profile.50 At the very least people should be supported in their screening decisions with evidence based and suitably comprehensive decision aids41 (fig 1) in conjunction with impartial and informed advice from their physician. Given the complexity of the information that needs to be weighed up in breast cancer screening and the fact that the topic requires confronting your mortality, determining whether your preferences align or conflict with screening for breast cancer is a process, rather than something that is easily decided instantaneously. Decision aids can help surface a person’s values and preferences.
Decisions aids for breast cancer screening have been successfully implemented across various settings.4151 Measures include integrating decisions aids into routine patient care and allocating physician “champions” to encourage adoption.51 Decision aids have been shown to increase the proportion of women making informed choices, and can also correct a woman’s pre-existing bias on the perceived benefits and harms of breast cancer screening to more accurately reflect the evidence.4152
Correspondingly for providers, the focus should shift from the narrow aim of encouraging screening uptake, towards supporting and educating health professionals to provide better quality information about screening. This is in line with their obligations to provide patients with information based on the best available evidence and strongly aligns with the UK General Medical Council’s guidance on decision making and consent.53
Achieving this shift will also require increased clinical and physician support for routine use of decision aids rather than considering them as optional, improved system support and resources for decision aids, and addressing time constraints for patient education and support.51 If incentives are considered, a preferable option would be to tie them to the use of decision aids, to overcome barriers to adoption.
We also propose breast cancer screening targets be abolished as they risk raising conflicts of interest for providers in facilitating preference sensitive decisions. An alternative would be to set targets for rates of informed choice, which would not face this challenge and are a meaningful alternative. Evidence based decision aids that enable scoring of different levels of understanding mean that this proposal is readily implementable (as long as decision makers genuinely value promoting women’s informed choices).4154 Research is needed to understand how to incorporate informed choice rates into performance indicators for breast cancer screening and other preference sensitive behaviours.
Use of incentives to increase uptake of breast cancer screening requires urgent reconsideration, as they are ethically problematic. A better approach would be to support women with their screening decisions through the provision of evidence based decision aids, as well as ensuring the availability of healthcare professionals who are both adequately trained and have no conflicts of interest in facilitating preference sensitive decisions.
Financial incentives tied to screening mammography are being used to increase uptake
Such incentives risk unduly influencing decisions since the benefits and harms are finely balanced
Health authorities should instead focus on promoting informed choice by providing evidence based decision aids
Targets for breast cancer screening should consider informed choice rather than uptake
We thank Samuel Finnikin for comments on an earlier draft of the manuscript as well as the reviewers, whose comments enabled us to clarify key points.
Contributors and sources: TB has a special interest in health inequalities and overdiagnosis, and has a master’s in bioethics. MC has had personal experience of the breast cancer screening process in France and participated in the citizen’s consultation on the new Cancer Plan, which included the proposal of financial incentives for screening. In her day job she works as a food scientist. HS’s research focuses on reducing disadvantage and promoting opportunity in health promotion and priority setting. The idea for the article was conceived by HS. TB wrote the first draft and led all subsequent revisions. All authors contributed to subsequent drafts. All have read and agreed to the final version. TB is guarantor.
Competing interests: We have read and understood BMJ policy on declaration of interests and have no interests to declare.
Patient and public involvement: MC alerted HS and TB to the citizens’ consultation that has been run by France’s National Institute of Cancer and provided invaluable input regarding her personal experiences of the screening process, broader insights into breast cancer screening in France, and the articulation of the substantive arguments set out here.
Provenance and peer review: Not commissioned; externally peer reviewed.