Medical Decision Making

Background: Although there are compelling moral arguments for patient participation in medical decisions, the link to health outcomes has not been systematically explored. Objective: Assess the extent to which patient participation in decision making within medical encounters is associated with measured patient outcomes. Methods: We conducted a primary search in PubMed—excluding non-English and animal studies—for articles on decision making in the context of the physician–patient relationship published through the end of February 2015, using the MeSH headings (Physician-Patient Relations [MeSH] OR Patient Participation [MeSH]) and the terms (decision OR decisions OR option OR options OR choice OR choices OR alternative OR alternatives) in the title or abstract. We also conducted a secondary search of references in all articles that met the inclusion criteria. Results: A thorough search process yielded 116 articles for final analysis. There was wide variation in study design, as well as measurement of patient participation and outcomes, among the studies. Eleven of the 116 studies were randomized controlled trials (RCTs). Interventions increased patient involvement in 10 (91%) of the 11 RCTs. At least one positive outcome was detected in 5 (50%) of the 10 RCTs reporting increased participation; the ratio of positive results among all outcome variables measured in these studies was much smaller. Although proportions differed, similar patterns were found across the 105 nonrandomized studies. Conclusions: Very few RCTs in the field have measures of participation in decision making and at least one health outcome. Moreover, extant studies exhibit little consistency in measurement of these variables, and results are mixed. There is a great need for well-designed, reproducible research on clinically relevant outcomes of patient participation in medical decisions.

Background. Values clarification is a recommended element of patient decision aids. Many different values clarification methods exist, but there is little evidence synthesis available to guide design decisions. Purpose. To describe practices in the field of explicit values clarification methods according to a taxonomy of design features. Data Sources. MEDLINE, all EBM Reviews, CINAHL, EMBASE, Google Scholar, manual search of reference lists, and expert contacts. Study Selection. Articles were included if they described 1 or more explicit values clarification methods. Data Extraction. We extracted data about decisions addressed; use of theories, frameworks, and guidelines; and 12 design features. Data Synthesis. We identified 110 articles describing 98 explicit values clarification methods. Most of these addressed decisions in cancer or reproductive health, and half addressed a decision between just 2 options. Most used neither theory nor guidelines to structure their design. “Pros and cons” was the most common type of values clarification method. Most methods did not allow users to add their own concerns. Few methods explicitly presented tradeoffs inherent in the decision, supported an iterative process of values exploration, or showed how different options aligned with users’ values. Limitations. Study selection criteria and choice of elements for the taxonomy may have excluded values clarification methods or design features. Conclusions. Explicit values clarification methods have diverse designs but can be systematically cataloged within the structure of a taxonomy. Developers of values clarification methods should carefully consider each of the design features in this taxonomy and publish adequate descriptions of their designs. More research is needed to study the effects of different design features.

The effectiveness of a pedometer intervention was affected by manipulating the goals given to participants and by providing social comparison feedback about how participants’ performance compared with others. In study 1 (n = 148), university staff members received a low, medium, or high walking goal (10%, 50%, or 100% increase over baseline walking). Participants walked 1358 more steps per day (95% confidence interval [CI], 729, 1985), when receiving a high goal than when receiving a medium goal, but a medium goal did not increase walking relative to a low goal (554 more steps; 95% CI, –71,1179). In study 2 (n = 64), participants received individual feedback only or individual plus social comparison feedback. Participants walked 1120 more steps per day (95% CI, 538, 1703) when receiving social comparison feedback than when receiving only individual feedback. Goals and the performance of others act as reference points and influence the effect that pedometer feedback has on walking behavior, illustrating the applicability of the principles of behavioral economics and social psychology to the design of health behavior interventions.

Objective. This qualitative study of a select sample of vaccine-hesitant parents (VHPs) explores perceived and constructed personal judgments about the risks and uncertainties associated with vaccines and vaccine-preventable diseases (VPDs) and how these subjective risk judgments influence parents’ decisions about childhood vaccination. Methods. The study employed semistructured focus group interviews with 42 VHPs to elicit parents’ perceptions and thought processes regarding the risks associated with vaccination and nonvaccination, the sources of these perceptions, and their approach to decision making about vaccination for their children. Results. VHPs engage in various reasoning processes and tend to perceive risks of vaccination as greater than the risks of VPDs. At the same time, VHPs engage in other reasoning processes that lead them to perceive ambiguity in information about the harms of vaccination—citing concerns about the missing, conflicting, changing, or otherwise unreliable nature of information. Conclusions. VHPs’ refusal of vaccination may reflect their aversion to both the risk and ambiguity they perceive to be associated with vaccination. Mitigating this vaccine hesitancy likely requires reconstructing the risks and ambiguities associated with vaccination—a challenging task that requires providing parents with meaningful evidence-based information on the known risks of vaccination versus VPDs and explicitly acknowledging the risks that remain truly unknown.

Background. Imprecision in estimates of benefits and harms around treatment choices is rarely described to patients. Variation in sampling error between treatment alternatives (e.g., treatments have similar average risks, but one treatment has a larger confidence interval) can result in patients failing to choose the option that is best for them. The aim of this study is to use a discrete choice experiment to describe how 2 methods for conveying imprecision in risk influence people’s treatment decisions. Methods. We randomized a representative sample of the Canadian general population to 1 of 3 surveys that sought choices between hypothetical treatments for rheumatoid arthritis based on different levels of 7 attributes: route and frequency of administration, chance of benefit, serious and minor side effects and life expectancy, and imprecision in benefit and side-effect estimates. The surveys differed in the way imprecision was described: 1) no imprecision, 2) quantitative description based on a range with a visual graphic, and 3) qualitative description simply describing the confidence in the evidence. Results. The analyzed data were from 2663 respondents. Results suggested that more people understood imprecision when it was described qualitatively (88%) versus quantitatively (68%). Respondents who appeared to understand imprecision descriptions placed high value on increased precision regarding the actual benefits and harms of treatment, equivalent to the value placed on the information about the probability of serious side effects. Both qualitative and quantitative methods led to small but significant increases in decision uncertainty for choosing any treatment. Limitations included some issues in defining understanding of imprecision and the use of an internet survey of panel members. Conclusions. These findings provide insight into how conveying imprecision information influences patient treatment choices.

We present results from a hypothetical framed field experiment assessing whether risk preferences significantly differ across the health and financial domains when they are elicited through the same multiple price list paired-lottery method. We consider a sample of 300 patients attending outpatient clinics in a university hospital in Athens during the Greek financial crisis. Risk preferences in finance were elicited using paired-lottery questions with hypothetical payments. The questions were adapted to the health domain by framing the lotteries as risky treatments in hypothetical health care scenarios. Using maximum likelihood methods, we estimated the degree of risk aversion, allowing for the estimates to be dependent on domain and individual characteristics. The subjects in our sample, who were exposed to both health and financial distress, tended to be less risk averse in the financial domain than in the health domain.

Background. Annual lung cancer screening using low-dose computed tomography (LDCT) scans is associated with a survival benefit, but it is also associated with potential harm. Unlike descriptive probability formats, experienced tasks have been shown to decrease perceptions of rare events. The objective of this study was to compare descriptive versus experienced probability formats on patients’ knowledge, beliefs, endorsement of screening for heavy smokers, and preference (choice predisposition) to undergo screening. Methods. A total of 276 patients attending an outpatient pulmonary practice were randomized to learn about screening using 1 of 3 formats: numbers only, numbers + icon arrays, numbers + a set of slides illustrating LDCT scans of 250 people in random order that displayed the number of normal scans, false-positive lung nodules, cancers found leading to a life saved, and cancers found leading to death despite treatment. Results. Knowledge differed between the 3 formats (P = 0.001), with participants randomized to the numbers + icon array format having the highest knowledge score. Beliefs were more favorable among participants randomized to the numbers + experienced format compared with the numbers + icon array format (difference between means [95% confidence interval]= 1.6 [0.4–2.8]). Differences in participants’ endorsement of screening (P = 0.4) and choice predisposition (P = 0.6) across probability format mirrored those of beliefs but were not statistically significant. Discussion. Contrary to what we expected, the experienced format increased propensity toward screening compared with the numbers + icon array format, as indicated by more favorable beliefs and nonsignificant trends toward stronger choice predisposition and endorsement. Experienced risk formats may not be a practical approach to improve risk communication for patients deciding whether or not to undergo annual lung cancer screening.

Background. Eliciting patient preferences within the context of shared decision making has been advocated for colorectal cancer (CRC) screening, yet providers often fail to comply with patient preferences that differ from their own. Purpose. To determine whether risk stratification for advanced colorectal neoplasia (ACN) influences provider willingness to comply with patient preferences when selecting a desired CRC screening option. Design. Randomized controlled trial. Setting/Participants. Asymptomatic, average-risk patients due for CRC screening in an urban safety net health care setting. Intervention. Patients were randomized 1:1 to a decision aid alone (n = 168) or decision aid plus risk assessment (n = 173) arm between September 2012 and September 2014. Outcomes. The primary outcome was concordance between patient preference and test ordered; secondary outcomes included patient satisfaction with the decision-making process, screening intentions, test completion rates, and provider satisfaction. Results. Although providers perceived risk stratification to be useful in selecting an appropriate screening test for their average-risk patients, no significant differences in concordance were observed between the decision aid alone and decision aid plus risk assessment groups (88.1% v. 85.0%, P = 0.40) or high- and low-risk groups (84.5% v. 87.1%, P = 0.51). Concordance was highest for colonoscopy and relatively low for tests other than colonoscopy, regardless of study arm or risk group. Failure to comply with patient preferences was negatively associated with satisfaction with the decision-making process, screening intentions, and test completion rates. Limitations. Single-institution setting; lack of provider education about the utility of risk stratification into their decision making. Conclusions. Providers perceived risk stratification to be useful in their decision making but often failed to comply with patient preferences for tests other than colonoscopy, even among those deemed to be at low risk of ACN.

Background. Evidence suggests that the average prediction across groups is more accurate than for individuals. Our goals were therefore to investigate accuracy of the average predictions for groups of clinicians and to compare this accuracy with a published statistical prediction model. Methods. Twenty-four expert clinicians attending an advisory board meeting were asked to make predictions for 25 patients from a research registry regarding the probability of having a positive bone scan 1 year from today if left untreated. Comparisons were made between the accuracy of average responses and that of an appropriate previously published statistical prediction model. Results. This study suggests that the mean of the clinicians’ predictions can quickly approach the accuracy of the best clinician using as few as 5 clinicians. When all 24 clinicians’ predictions were averaged, the concordance index reached 0.750, still far below that of the published statistical model with 0.812. Conclusions. Averaging clinician predictions may have merit over individual clinician predictions but still not reasonably replace a carefully built statistical model. However, averaging clinician predictions could prove helpful in situations where statistical models do not yet exist or where existing models are inadequate.

The unconscious thought theory argues that making complex decisions after a period of distraction can lead to better decision quality than deciding either immediately or after conscious deliberation. Two studies have tested this unconscious thought effect (UTE) in clinical diagnosis with conflicting results. The studies used different methodologies and had methodological weaknesses. We attempted to replicate the UTE in medical diagnosis by providing favorable conditions for the effect while maintaining ecological validity. Family physicians (N = 116) diagnosed 3 complex cases in 1 of 3 thinking modes: immediate, unconscious (UT), and conscious (CT). Cases were divided into short sentences, which were presented briefly and sequentially on computer. After each case presentation, the immediate response group gave a diagnosis, the UT group performed a 2-back distraction task for 3 min before giving a diagnosis, and the CT group could take as long as necessary before giving a diagnosis. We found no differences in diagnostic accuracy between groups (P = 0.95). The CT group took a median of 7 s to diagnose, which suggests that physicians were able to diagnose “online,” as information was being presented. The lack of a difference between the immediate and UT groups suggests that the distraction had no additional effect on performance. To assess the decisiveness of the evidence of this null result, we computed a Bayes factor (BF₀₁) for the 2 comparisons of interest. We found a BF₀₁ of 5.76 for the UT versus immediate comparison and of 3.61 for the UT versus CT comparison. Both BFs provide substantial evidence in favor of the null hypothesis: physicians’ diagnoses made after distraction are no better than diagnoses made either immediately or after self-paced deliberation.

Introduction. Participation in the decision-making process and health literacy may both affect health outcomes; data on how these factors are related among diverse groups are limited. This study examined the relationship between health literacy and decision-making preferences in a medically underserved population. Methods. We analyzed a sample of 576 primary care patients. Multivariable logistic regression was used to examine the independent association of health literacy (measured by the Rapid Estimate of Adult Literacy in Medicine–Revised) and patients’ decision-making preferences (physician directed or patient involved), controlling for age, race/ethnicity, and gender. We tested whether having a regular doctor modified this association. Results. Adequate health literacy (odds ratio [OR] = 1.7; P = 0.009) was significantly associated with preferring patient-involved decision making, controlling for age, race/ethnicity, and gender. Having a regular doctor did not modify this relationship. Males were significantly less likely to prefer patient-involved decision making (OR = 0.65; P = 0.024). Discussion. Findings suggest health literacy affects decision-making preferences in medically underserved patients. More research is needed on how factors, such as patient knowledge or confidence, may influence decision-making preferences, particularly for those with limited health literacy.