[Skip to Content]
[Skip to Content Landing]
Views 13,460
Citations 0
March 9, 2020

Improving Physician Communication About Treatment Decisions: Reconsideration of “Risks vs Benefits”

Author Affiliations
  • 1Department of Epidemiology and Public Health, University of Maryland School of Medicine & Hospital Epidemiology, Baltimore
  • 2Veterans Affairs Maryland Health Care System, Baltimore
  • 3Adult and Child Consortium of Health Outcomes Research and Delivery Science (ACCORDS), Division of Cardiology, University of Colorado School of Medicine, Aurora
  • 4Center of Innovation for Veteran-Centered and Value-Driven Care, VA Denver, Denver, Colorado
  • 5Center for Health Policy and Outcomes, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, New York
  • 6Department of Medicine, Weill Cornell Medical College, New York, New York
JAMA. Published online March 9, 2020. doi:10.1001/jama.2020.0354

Patient-physician conversations around prevention and treatment decisions are fundamental to good medical care. Optimal communication is needed for shared decision-making that enables patients to make evidence-informed choices.1 Clearly, the words physicians use have a critical function in this communication. However, many physicians may use language that could mislead patients and alter their decisions, specifically involving the phrase “risks vs benefits.” Among the most commonly used medical phrases, “risks vs benefits” is usually an inaccurate comparison and may have potential implications for patient-physician decision-making.

Referring to harms as “risks” emphasizes that the unfavorable outcome may or may not happen, whereas there is no parallel language that highlights the equally probabilistic nature of “benefits.” Presenting treatment decisions as a comparison of risks vs benefits creates an inherent imbalance in which benefits simply exist, whereas harms are uncertain. This imbalance is widespread and is present in how physicians have discussions with patients, how physicians likely approach decisions, and even how studies are reported in the medical literature. Recent guidelines for research reporting, such as Grading of Recommendations Assessment, Development and Evaluation, have used the phrase “harms vs benefits,” but the term has likely had little effect in discussions between clinicians and patients. Comparing “risks vs benefits” instead of “harms vs benefits” could also potentially contribute to unnecessary and, sometimes, harmful medical care because everyday language highlights the uncertainty of harms but not benefits.2

How Language Biases Decisions

Behavioral science has shown that minor changes in language or framing can significantly alter judgments and decisions, including in health care settings. In a study on perceptions of ductal carcinoma in situ (DCIS) that included 394 women, 69% indicated that they preferred watchful waiting when DCIS was framed as “abnormal cells,” but 53% preferred treatment when DCIS was framed as “noninvasive cancer.”3 Such framing effects are thought to be driven by emotional or cognitive associations and expectations triggered by specific words and phrases.

Framing trade-offs in terms of risks and benefits may have similar effects. Not only is “risk-benefit” framing less clear than “harm-benefit,” it also may bias decisions by emphasizing the uncertainty related to harms while implying certainty around benefits. This may contribute to a tendency among both physicians and patients to perceive medical therapy predominantly in terms of gains, with less awareness of potential harms.2 A change in words may prompt reflection on the part of physicians to give equal weight to harms and benefits of treatments.

Evolution of Risk Terminology

The formal incorporation of probability and risk into medical decisions began with the development of informed consent and US Food and Drug Administration (FDA) oversight in the 1960s. The legal requirement for informed consent for medical treatment grew out of the need to give patients autonomy over their medical decisions. In Canterbury v Spence “true consent” was defined as “the informed exercise of a choice, and that entails an opportunity to evaluate knowledgeably the options available.” This included the “degree of the harm threatened” and “potential benefit of the therapy.”4

The concept of “risk” in place of “harm” seems to have emerged in 1962, when the US Congress passed the Kefauver-Harris amendment for the FDA, which affirmed the need for drugmakers to demonstrate the efficacy of their products for human consumption by stating, “For a drug to be approved for marketing, FDA must determine that the drug is effective and that its expected benefits outweigh its potential risks to patients.”5 Since then, the use of the term “risk in this context has proliferated.

What Is Better Than “Risks vs Benefits”?

Language regarding medical therapy must be accurate and easy to understand for medical professionals, the lay public, and patients, and should facilitate, rather than impede, patient-physician communication. Avoiding the term “risks” and replacing it with “harms” is a good first step. Discussions about tests and treatment options should incorporate a description of potential harms and benefits as well as the likelihood of both. Language that emphasizes the uncertainty of both desirable and undesirable outcomes includes “chance,” “probability,” and “likelihood.” Language for treatment decisions should focus on the understandable “chance” of harms and benefits, using absolute and not relative numbers to compare the chance of each outcome with and without treatment. The persistent use of relative differences in medical journals and news reports of scientific findings adds to the confusion of true risk.

For example, a clinician might explain to a patient: “You have atrial fibrillation, and 4 of 100 people like you with atrial fibrillation will have a stroke every year. Let’s talk about the chance of benefits and harms of treatment with anticoagulation, which involves taking medication to decrease the ability of your blood to form clots. Studies have shown that treatment with these drugs reduces the chance of stroke occurring from 4 of 100 people with atrial fibrillation per year to 2 of 100 people per year. However, 1 to 3 people of 100 who are treated with these anticoagulation drugs will experience significant bleeding that could involve bleeding into the brain or intestinal tract. In other words, most patients like you do not benefit from or experience harms from this treatment.”

Challenges and Nuances to Communication of Harms and Benefits

The process of shared decision-making is difficult for many reasons, including limitations in the evidence, challenges in comparing diverse outcomes, and the time needed to have these discussions. The imbalance of “risks vs benefits” language adds to this difficulty. The wide spectrum of harms and benefits of importance to a patient should be discussed, including treatment burden, such as taking medications and attending appointments, and possible psychological effects, such as depression and anxiety. However, in many cases these nonphysical harms are poorly described and quantified. Accurately balancing pros and cons of treatment has been recognized in the shared decision-making literature, but the centrality of replacing “risks” with “harms” has not been emphasized.1

Furthermore, trade-offs may be difficult to conceptualize when outcomes vary widely in frequency and severity. A patient could face a comparison of rare significant benefits with common mild harms. For example, framing breast cancer screening with mammography as having “risks and benefits” makes benefits seem more certain than harms, whereas the actual likelihood of benefit, less than 1 in 1000 chance of avoiding death from breast cancer, should be balanced against much more certain harms, including a 3 in 1000 chance of experiencing breast cancer overdiagnosis and a 60% chance of a false-positive mammography screening result with subsequent anxiety and consequences related to follow-up testing.6 In this example, the use of balanced language of benefits and harms can at least help to facilitate understanding of this complex trade-off with a description of possible harms and benefits and the probabilities of each.

For an older woman considering breast cancer screening, framing the discussion around harms vs benefits rather than the more nebulous “risks” might facilitate the decision to discontinue nonrecommended screening. The fact that screening is more likely to be harmful than beneficial to an individual patient may be a more compelling reason to discontinue screening than the vague notion of limited life expectancy, which, in the context of an assumed net benefit from screening, can be interpreted as abandonment.7


Language guides decisions. It is incumbent upon physicians to use the most accurate and simple language possible when communicating with patients about medical science. The phrase “chance of harms and benefits” should be considered as the standard language for physicians, scientific journals, and policy makers to convey a scientifically accurate understanding of medical choices. Adopting this phrase and concept for communication can enable decisions that are more appropriate, potentially more understandable, and better aligned with patients.

Back to top
Article Information

Corresponding Author: Daniel J. Morgan, MD, University of Maryland, Baltimore, 685 W Baltimore St, MSTF 334, Baltimore, MD 21201 (dmorgan@som.umaryland.edu).

Published Online: March 9, 2020. doi:10.1001/jama.2020.0354

Conflict of Interest Disclosures: Dr Morgan reported receiving grants from the National Institutes of Health, the Agency for Healthcare Research and Quality, the Centers for Disease Control and Prevention, and the VA Health Services Research and Development Service outside the submitted work. Dr Scherer reported receiving grants from the National Institutes of Health. Dr Korenstein reported being supported in part by a Cancer Center Support grant to the Memorial Sloan Kettering Cancer Center from the National Cancer Institute (P30 CA008748) and her spouse serving on the scientific advisory board of Vedanta Biosciences and consulting for Takeda Pharmaceutical. No other disclosures were reported.

Hoffmann  TC, Montori  VM, Del Mar  C.  The connection between evidence-based medicine and shared decision making.  JAMA. 2014;312(13):1295-1296. doi:10.1001/jama.2014.10186PubMedGoogle ScholarCrossref
Hoffmann  TC, Del Mar  C.  Clinicians’ expectations of the benefits and harms of treatments, screening, and tests: a systematic review.  JAMA Intern Med. 2017;177(3):407-419. doi:10.1001/jamainternmed.2016.8254PubMedGoogle ScholarCrossref
Omer  ZB, Hwang  ES, Esserman  LJ, Howe  R, Ozanne  EM.  Impact of ductal carcinoma in situ terminology on patient treatment preferences.  JAMA Intern Med. 2013;173(19):1830-1831. doi:10.1001/jamainternmed.2013.8405PubMedGoogle ScholarCrossref
Canterbury v Spence, 464 F2d 772, 775 (DC Cir 1972).
Benefit-Risk Assessment in Drug Regulatory Decision-Making. US Food and Drug Administration; 2018. Accessed August 30, 2019. https://www.fda.gov/media/112570/download
Keating  NL, Pace  LE.  Breast cancer screening in 2018: time for shared decision making.  JAMA. 2018;319(17):1814-1815. doi:10.1001/jama.2018.3388PubMedGoogle ScholarCrossref
Schoenborn  NL, Lee  K, Pollack  CE,  et al.  Older adults’ views and communication preferences about cancer screening cessation.  JAMA Intern Med. 2017;177(8):1121-1128. doi:10.1001/jamainternmed.2017.1778PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    Language Complexity: 'RIsk' And 'Harm' Are Not The Same.
    Peter Shah, BSc MA FRCOphth FRCP Edin | University Hospitals Birmingham NHS Foundation Trust
    Morgan et al eloquently argue for a shift in the language of consent for treatment. However, language is complex, and one cannot use the terms 'risk' and 'harm' interchangeably.

    For example, I may need to perform high-stakes surgery on the only seeing eye of a young patient with glaucoma. There may be a high risk of needing to return to theatre within the first seven days post-operatively to manage challenges such as hypotony, but return to theatre is not the same as the patient suffering harm - i.e. loss of vision in their only eye.

    is a place for both 'risk' and 'harm' in these complex dialogues. One could discuss that the risk of needing to return to theatre is 3 to 5% in the most challenging cases, but that the chance of the patient suffering serious harm and becoming totally blind as a result of the operation is very small at about 1 in a 1000.

    Equating risk and harm as equivalent has the potential to fuel an already complex medico-legal landscape.
    Language is important, and underpins the power relations within the doctor-patient relationship. In discussing treatment and consent as we journey with our patients, there is one word upon which all the others rest - compassion.