Conflict of Interest

Conflict of Interest in Health Economics and Health Technology Assessment

A conflict of interest arises from a person's circumstances, such as consultancy fees from the maker of an appraised drug, so it can exist before any judgement has been distorted and whether or not bias ever occurs. In health economics the concern is practical, because manufacturers fund many of the cost-effectiveness analyses payers rely on, and sponsored analyses report more favourable results than others. A health technology assessment body responds by having interests declared and then managed, through rules on who takes part in a decision and independent critique of the company's evidence.

Primary and secondary interests

The Institute of Medicine (IOM) report of 2009 describes conflicts of interest as circumstances that create a risk that professional judgements or actions about a primary interest will be unduly influenced by a secondary interest. Primary interests include the integrity of research, the quality of medical education and the welfare of patients. In HTA the primary interest also includes a sound funding decision for the population a payer serves.

Secondary interests are wider than money. The IOM lists professional advancement and recognition and the wish to do favours for friends or colleagues, and the International Committee of Medical Journal Editors (ICMJE) adds personal rivalries, academic competition and intellectual beliefs. Policies still concentrate on financial gain because, in the IOM's account, it is relatively more objective, fungible and quantifiable.

In the IOM framework severity depends on two judgements. One is the likelihood that decisions made in the circumstances would be unduly influenced, which rises with the value of the secondary interest, its duration and depth, and the discretion the individual holds. The other is the seriousness of the harm or wrong that could follow. A finding that someone has a conflict does not imply that the person is unethical, and the IOM notes that policies generally emphasise prevention and management rather than punishment.

Why economic evaluation is exposed

Economic evaluation gives the analyst a great deal of discretion, which the IOM identifies as a driver of likelihood. Lane and colleagues note that investigators set the values assigned to quality of life, determine the price and duration of interventions, and make other methodological choices that can affect study findings. Xie and Zhou found that industry-sponsored model-based analyses reported significantly lower incremental cost-effectiveness ratios, whereas no significant difference was reached in trial-based studies.

Xie and Zhou also point out that manufacturers of new medicines are required to submit a cost-effectiveness analysis when applying for coverage in many countries, and that a technology with a lower ICER is more likely to be covered. The main evidence for a decision can therefore come from the party that profits from a favourable answer.

Evidence that sponsorship is linked to favourable results

Reviews of published economic evaluations have found the same direction of association for over two decades. The table gives thresholds in US dollars as the authors reported them.

Clinical evidence points the same way. The Cochrane review by Lundh and colleagues, covering 75 papers, found that industry-sponsored drug and device studies more often had favourable efficacy results (risk ratio 1.27) and favourable conclusions (risk ratio 1.34), with less agreement between results and conclusions. The authors concluded that there is an industry bias that cannot be explained by standard risk of bias assessments, which matters for any systematic review feeding a model.

These are associations across many studies. Friedberg and colleagues did not identify bias in individual studies, and the IOM notes that companies may fund only trials with a reasonable likelihood of success. Lane and colleagues add that retrospective designs can let investigators look early at the data and then choose the trials most likely to yield positive outcomes.

How HTA bodies declare and manage interests

The NICE policy on declaring and managing interests for advisory committees (version 1.7, November 2025) covers chairs, members, expert witnesses and organisations making submissions. Interests are declared if a reasonable person would think them relevant, and declaring an interest does not mean there is a conflict. It separates direct financial interests, such as consultancy or shareholdings in the commercial sector, from direct non-financial interests, such as having published a clear opinion on the matter, and from indirect interests held by close associates or an employer. Declarations cover the preceding 12 months and are updated at each meeting.

The response depends on whether the interest is specific to the topic, which includes underpinning research papers or economic analyses. There are three responses, open declaration with no further action, partial exclusion from developing recommendations and decision making, and complete exclusion from the topic. A member or standing committee chair with a specific financial interest normally leaves the meeting for that item unless the interest has ended and there is no scope to benefit, although a member with private practice or commercial income may take part if exclusion would reduce the committee's access to clinical expertise. Members should not be appointed if they are likely to be excluded from more than 50% of a committee's discussions. Under PMG36 (section 5.8.8), appraisal committee members declare interests in the public part of each meeting, recorded in the minutes.

The company's own interest is managed structurally. Under PMG36 (sections 1.3.24 and 1.3.25) NICE commissions an independent external assessment group to critique the evidence, which in a single technology evaluation may recommend extra analyses from the company or run exploratory analyses itself. The Guidelines International Network applies similar logic to clinical guidelines, with principles that chairs should have no direct financial or relevant indirect conflicts and that no member deciding the direction or strength of a recommendation should hold a direct financial conflict.

Reporting conflicts in published economic evaluations

Reporting standards make interests visible. The CHEERS 2022 checklist asks authors to describe how the study was funded and any role of the funder in the identification, design, conduct and reporting of the analysis (item 27), and to report conflicts of interest according to journal or ICMJE requirements (item 28). Item 4 asks whether a health economic analysis plan was developed, which matters because Lane and colleagues argue that making methodological choices transparently and before results are known could improve credibility. The ICMJE also requires authors to avoid agreements with sponsors that interfere with access to all the study's data or with independent analysis and publication.

Worked example of discretion in a sponsored model

All numbers in this example are illustrative. A company model compares a new drug with standard care. The incremental cost is GBP 27,000 per patient, fixed because the treatment course is fixed, and the trial shows a gain of 0.60 QALYs over 2 years of follow-up. The analyst must choose how long the effect lasts after the trial.

$$\text{ICER} = \frac{\Delta C}{\Delta E}$$

where $\Delta C$ is the incremental cost per patient in GBP and $\Delta E$ is the incremental QALYs per patient.

Against an illustrative cost-effectiveness threshold of GBP 30,000 per QALY, assumption A rejects the drug, B sits on the threshold and C accepts it. The move from B to C lowers the ICER by $(30{,}000 - 20{,}000) / 30{,}000 = 0.33$, or 33%, about the size of the adjusted gap Xie and Zhou reported between sponsored and other analyses. None of the three assumptions is falsified by the trial, so the choice is a judgement that a secondary interest could influence.

A pre-specified analysis plan fixes the base case before results are known, scenario analysis shows all three ICERs, and an independent group can test the evidence for lasting benefit. The sponsorship itself does not show that assumption C is wrong.

Boundaries with related concepts

A conflict of interest is a risk factor, and bias is a systematic error in results. The ICMJE separates the funding statement, which covers direct support of the work, from disclosure of relationships, which extends beyond it. Disclosure, the act of reporting an interest, is the first step in handling a conflict. Spin and reporting bias are possible consequences of a conflict.

The structure is that of the principal-agent problem, in which an agent whose actions cannot be fully observed may serve their own interest. Self-referral, where a physician refers patients to a facility in which they have an ownership interest, is a clinical example the IOM recognises.

Common misreadings of conflict of interest policy

Disclosure is often treated as a remedy, but the IOM calls it a critical but limited first step. Earlier research summarised by Sah and Loewenstein found that disclosure has little effect on what advisees do and can backfire by leading advisers to give more biased advice, and further experiments found that it can increase pressure on advisees to comply with advice they distrust. Disclosure did deter advisers from accepting avoidable conflicts in their experiments, so its value depends on whether the conflict can be avoided.

A second is that independent analysts have no interests. The IOM recognises professional advancement and recognition as secondary interests, and the NICE policy treats a published clear opinion on the topic as a non-financial interest. Excluding everyone with an interest has a cost, so NICE weighs access to expertise against the risk to objectivity.

A third is to discount sponsored evidence automatically. The association is consistent but does not show which individual analysis is biased. Critical appraisal of the assumptions, independent replication and scenario analysis are better guides than the funding line alone, especially as Lundh and colleagues found that standard risk of bias tools do not capture the effect.

Sources

• Bell CM, Urbach DR, Ray JG, Bayoumi A, Rosen AB, Greenberg D, Neumann PJ. Bias in published cost effectiveness studies: systematic review. BMJ. 2006;332(7543):699-703. https://doi.org/10.1136/bmj.38737.607558.80
• Friedberg M, Saffran B, Stinson TJ, Nelson W, Bennett CL. Evaluation of conflict of interest in economic analyses of new drugs used in oncology. JAMA. 1999;282(15):1453-1457. https://doi.org/10.1001/jama.282.15.1453
• Husereau D, Drummond M, Augustovski F, et al. Consolidated Health Economic Evaluation Reporting Standards 2022 (CHEERS 2022) statement: updated reporting guidance for health economic evaluations. BMC Health Services Research. 2022;22:114. https://doi.org/10.1186/s12913-021-07460-7
• Institute of Medicine. Conflict of Interest in Medical Research, Education, and Practice. Lo B, Field MJ, eds. Washington, DC: National Academies Press; 2009. https://doi.org/10.17226/12598
• International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. Updated January 2026. Section II.B, Disclosure of Financial and Non-Financial Relationships and Activities, and Conflicts of Interest. https://www.icmje.org/recommendations/browse/roles-and-responsibilities/author-responsibilities--conflicts-of-interest.html
• Lane JD, Friedberg MW, Bennett CL. Associations between industry sponsorship and results of cost-effectiveness analyses of drugs used in breast cancer treatment. JAMA Oncology. 2016;2(2):274-276. https://doi.org/10.1001/jamaoncol.2015.3928
• Lundh A, Lexchin J, Mintzes B, Schroll JB, Bero L. Industry sponsorship and research outcome. Cochrane Database of Systematic Reviews. 2017;(2):MR000033. https://doi.org/10.1002/14651858.MR000033.pub3
• National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). London: NICE; 2022, last updated 13 October 2026. Sections 1.3.24, 1.3.25 and 5.8.8.
• National Institute for Health and Care Excellence. Policy on declaring and managing interests for NICE advisory committees. Version 1.7. London: NICE; November 2025.
• Sah S, Loewenstein G. Nothing to declare: mandatory and voluntary disclosure leads advisors to avoid conflicts of interest. Psychological Science. 2014;25(2):575-584. https://doi.org/10.1177/0956797613511824
• Sah S, Loewenstein G, Cain DM. The burden of disclosure: increased compliance with distrusted advice. Journal of Personality and Social Psychology. 2013;104(2):289-304. https://doi.org/10.1037/a0030527
• Schünemann HJ, Al-Ansary LA, Forland F, et al. Guidelines International Network: principles for disclosure of interests and management of conflicts in guidelines. Annals of Internal Medicine. 2015;163(7):548-553. https://doi.org/10.7326/M14-1885
• Traversy G, Barnieh L, Akl EA, et al. Managing conflicts of interest in the development of health guidelines. CMAJ. 2021;193(2):E49-E54. https://doi.org/10.1503/cmaj.200651
• Xie F, Zhou T. Industry sponsorship bias in cost effectiveness analysis: registry based analysis. BMJ. 2022;377:e069573. https://doi.org/10.1136/bmj-2021-069573