INTRODUCTION
Shared decision-making is a bidirectional therapeutic partnership between patient and physician to promote discussion and reach an agreement based on patient values.1 In the context of uncertainty, communication processes are critical to understanding patients’ values and conveying relevant information.2 Whereas explicit physician-led recommendations have been shown to influence patient decision-making,3 the impact of implicit communication has not been well described. In everyday shared decision-making, and especially in highly charged contexts, behavioral economics concepts, such as framing and nudging,4 may strongly influence patient choices.5 In this randomized vignette study, we investigated the impact of emotionally salient and loaded words on the choice of intensity of care.
METHODS
We developed five clinical vignettes featuring a fictional patient with an acute health deterioration and a quantified prognosis requiring a decision about care. All five-vignette questionnaires administered to participants were identical except for the randomized presence or absence of loaded words, chosen by expert consensus among authors (i.e., “loss of independence,” “aggressive care,” “to end up in a vegetative state,” “to live with dignity,” or “to become a burden”). Participants assumed a patient or caregiver role and had to choose between two treatment options representing high or low intensity of care. Box 1 presents one example vignette with “loss of independence.”
Text Box 1. Clinical vignette with “loss of independence”
Notes. The vignette without the loaded words was identical except that it excluded the words in the brackets. Most participants were exposed to vignettes with and without loaded words
Following ethics approval, we recruited 100 French-speaking participants from the Centre hospitalier de l’Université de Montréal (CHUM) among patients and caregivers in outpatient clinics. Sociodemographic data (age, gender, ethnicity, religion, years of education, and occupation) were collected. The primary outcome was the level of intensity of care.
For the primary analysis comprising all vignettes, a mixed effects logistic regression with participants as random intercepts was used to account for clustering. We calculated odds ratios (OR) for individual vignettes. Significance was set at α = 0.05 (two-sided). Randomization and analyses were conducted using R 4.1.2 (R Foundation).
RESULTS
One hundred participants completed 500 vignettes between October and November 2022. The population included 70 women and 30 men, 54 patients, and 46 caregivers, with an average age of 54 (IQR = 38–66). Sociodemographic variables were evenly distributed. Overall, when the vignettes included loaded words, the proportion of participants choosing the lower intensity of care was 65% versus 38% when they were absent (OR = 3.7; CI: 2.4–5.7). Figure 1 presents the results for all vignettes and each vignette individually. The respective odd ratios and confidence intervals were OR = 4.2 (CI: 1.7–10.1) for “loss of independence,” OR = 5.8 (CI:2.4–14.1) for “aggressive care,” OR = 2.8 (CI: 2.4–14.1) for “vegetative state,” OR = 4.0 (CI: 1.6, 9.9) for “to live with dignity,” and OR = 1.8 (CI: 0.8, 4.1) for “to become a burden.”
Figure 1.
Choice of a lower intensity of care according to the presence of loaded words in randomized case vignettes.
DISCUSSION
The presence of loaded words in otherwise identical clinical vignettes had a substantial impact on decision outcomes related to intensity of care. Loaded words, baring emotions, and ethical and cultural values, are commonly used by physicians, patients, and proxy decision-makers.
The impact of nudging and framing, concepts that suggest that decision-making is not strictly rational and can be influenced by the way information is presented, has been previously described2,4; our study is the first to demonstrate their impact on the choice of intensity of care.
Although the vignettes were in French, loaded words can be found in every language. The effect of loaded words in all five vignettes suggests that their impact is not restricted to specific wording or language. Persuasive words used by physicians have been shown to orient patients towards a certain level of treatment.6
The strengths of our study are the randomization of loaded words and the inclusion of a diverse participant population. An important limitation is the use of written vignettes which do not capture variation in other verbal and non-verbal communicative elements, such as prosody. Our findings are likely an underestimation of the full impact of behavioral elements in decision-making. Other limitations include the inclusion of participants from a single center’s outpatient setting; the lack of power to detect an interaction between loaded words and gender, age, or other characteristics; and dichotomized decisions that may not reflect nuanced real-world decisions.
Our study shows that words have a significant impact on patients’ and carers’ decision-making. Since loaded words may be used both within shared decision-making and paternalism, exploring when or whether it may be appropriate to use them may be warranted.
Author Contribution
Dr. Nguyen had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Cheung, Nguyen, Paradis
Acquisition, analysis, or interpretation of data: All authors
Drafting of the manuscript: Nguyen, Paradis, Zhang
Critical revision of the manuscript for important intellectual content: All authors
Statistical analysis: Cheung, Nguyen
Administrative, technical, or material support: Cheung, Paradis
Supervision: Nguyen
Contributors: This article does not have any contributors that do not meet the criteria for authorship.
Funding
This work was funded in part by the Fonds de recherche du Québec – Santé.
Data Availability
Data will be made available upon reasonable request to the corresponding author.
Declarations:
Conflict of Interests:
The authors declare that they do not have a conflict of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Barry MJ, Edgman-Levitan S. Shared decision making--pinnacle of patient-centered care. N Engl J Med. 2012;366(9):780-1. 10.1056/NEJMp1109283. [DOI] [PubMed]
- 2.Truglio-Londrigan M, Slyer JT, Singleton JK, et al. A qualitative systematic review of internal and external influences on shared decision-making in all health care settings. JBI Libr Syst Rev. 2012;10(58):4633-4646. 10.11124/jbisrir-2012-432. [DOI] [PubMed]
- 3.Gurmankin AD, Baron J, Hershey JC, et al. The role of physicians' recommendations in medical treatment decisions. Med Decis Making. 2002;22(3):262–71. 10.1177/0272989X0202200314. [DOI] [PubMed]
- 4.Thaler RH, Sunstein CR. Nudge: Improving Decisions about Health, Wealth, and Happiness. Penguin Books; 2008.
- 5.Halpern SD, Loewenstein G, Volpp KG, et al. Default options in advance directives influence how patients set goals for end-of-life care. Health Aff (Millwood). 2013;32(2):408-17. 10.1377/hlthaff.2012.0895. [DOI] [PMC free article] [PubMed]
- 6.Martinez KA, Hurwitz HM, Rothberg MB. Qualitative Analysis of Patient-Physician Discussions Regarding Anticoagulation for Atrial Fibrillation. JAMA Intern Med. 2022;182(12):1260-1266. 10.1001/jamainternmed.2022.4918. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data will be made available upon reasonable request to the corresponding author.