Medical Decision Making

Bonner, Carissa; Trevena, Lyndal J.; Gaissmaier, Wolfgang; Han, Paul K. J.; Okan, Yasmina; Ozanne, Elissa; Peters, Ellen; Timmermans, Daniëlle; Zikmund-Fisher, Brian J.

doi: 10.1177/0272989X21996328pmid: 33660551

BackgroundShared decision making requires evidence to be conveyed to the patient in a way they can easily understand and compare. Patient decision aids facilitate this process. This article reviews the current evidence for how to present numerical probabilities within patient decision aids.MethodsFollowing the 2013 review method, we assembled a group of 9 international experts on risk communication across Australia, Germany, the Netherlands, the United Kingdom, and the United States. We expanded the topics covered in the first review to reflect emerging areas of research. Groups of 2 to 3 authors reviewed the relevant literature based on their expertise and wrote each section before review by the full authorship team.ResultsOf 10 topics identified, we present 5 fundamental issues in this article. Although some topics resulted in clear guidance (presenting the chance an event will occur, addressing numerical skills), other topics (context/evaluative labels, conveying uncertainty, risk over time) continue to have evolving knowledge bases. We recommend presenting numbers over a set time period with a clear denominator, using consistent formats between outcomes and interventions to enable unbiased comparisons, and interpreting the numbers for the reader to meet the needs of varying numeracy.DiscussionUnderstanding how different numerical formats can bias risk perception will help decision aid developers communicate risks in a balanced, comprehensible manner and avoid accidental “nudging” toward a particular option. Decisions between probability formats need to consider the available evidence and user skills. The review may be useful for other areas of science communication in which unbiased presentation of probabilities is important.

Yin, Siyuan; Arkes, Hal R.; McCoy, John P.; Cohen, Margot E.; Mellers, Barbara A.

doi: 10.1177/0272989X211001841pmid: 33764191

BackgroundPhysicians who communicate their prognostic beliefs to patients must balance candor against other competing goals, such as preserving hope, acknowledging the uncertainty of medicine, or motivating patients to follow their treatment regimes.ObjectiveTo explore possible differences between the beliefs physicians report as their own and those they express to patients and colleagues.DesignAn online panel of 398 specialists in internal medicine who completed their medical degrees and practiced in the United States provided their estimated diagnostic accuracy and prognostic assessments for a randomly assigned case. In addition, they reported the diagnostic and prognostic assessments they would report to patients and colleagues more generally. Physicians answered questions about how and why their own beliefs differed from their expressed beliefs to patients and colleagues in the specific case and more generally in their practice.ResultsWhen discussing beliefs about prognoses to patients and colleagues, most physicians expressed beliefs that differed from their own beliefs. Physicians were more likely to express greater optimism when talking to patients about poor prognoses than good prognoses. Physicians were also more likely to express greater uncertainty to patients when prognoses were poor than when they were good. The most common reasons for the differences between physicians’ own beliefs and their expressed beliefs were preserving hope and acknowledging the inherent uncertainty of medicine.ConclusionTo balance candor against other communicative goals, physicians tended to express beliefs that were more optimistic and contained greater uncertainty than the beliefs they said were their own, especially in discussions with patients whose prognoses were poor.

ten Broeke, Annebel; Hulscher, Jan; Heyning, Nicolaas; Kooi, Elisabeth; Chorus, Caspar

doi: 10.1177/0272989X211001320pmid: 33783246

We present a novel way to codify medical expertise and to make it available to support medical decision making. Our approach is based on econometric techniques (known as conjoint analysis or discrete choice theory) developed to analyze and forecast consumer or patient behavior; we reconceptualize these techniques and put them to use to generate an explainable, tractable decision support system for medical experts. The approach works as follows: using choice experiments containing systematically composed hypothetical choice scenarios, we collect a set of expert decisions. Then we use those decisions to estimate the weights that experts implicitly assign to various decision factors. The resulting choice model is able to generate a probabilistic assessment for real-life decision situations, in combination with an explanation of which factors led to the assessment. The approach has several advantages, but also potential limitations, compared to rule-based methods and machine learning techniques. We illustrate the choice model approach to support medical decision making by applying it in the context of the difficult choice to proceed to surgery v. comfort care for a critically ill neonate.

Rahn, Anne Christin; Jull, Janet; Boland, Laura; Finderup, Jeanette; Loiselle, Marie-Chantal; Smith, Maureen; Köpke, Sascha; Stacey, Dawn

doi: 10.1177/0272989X21997330pmid: 33759626

IntroductionIn 2005, the International Patient Decision Aid Standards (IPDAS) collaboration identified guidance and decision coaching as important dimensions of patient decision aids (PtDAs) and developed a set of quality criteria. We sought to update definitions, theoretical rationale, and evidence for guidance and/or decision coaching used within or alongside PtDAs for the IPDAS update 2.0.MethodsWe conducted 2 scoping reviews on guidance and decision coaching, including systematic searches and a hand search of the Cochrane Review on PtDAs. Eligible studies were randomized controlled trials (RCTs) on guidance or decision coaching used with/alongside PtDAs. Data, including conceptual models, were summarized narratively and with meta-analyses when appropriate.ResultsOf 1022 citations, we found no RCTs that evaluated guidance in PtDAs. The 2013 definition for guidance was endorsed, and we made minimal changes to the description of guidance. Of 3039 citations, we identified 21 RCTs on decision coaching informed by 5 conceptual models stating that people exposed to decision coaching are more likely to progress in making informed decisions consistent with their values. Compared to usual care, decision coaching with PtDAs led to improved knowledge mean difference [MD], 19.5/100; 95% confidence interval [CI], 10.0–29.0; 5 RCTs). Compared to decision coaching alone, PtDAs led to a small improvement in knowledge (MD, 3.6/100; 95% CI, 1.0–6.3; 3 RCTs). There were variable effects on other outcomes. We simplified the decision coaching definition slightly and defined minimal decision coaching elements.ConclusionWe found no evidence on which to propose changes in guidance in IPDAS. Decision coaching is continuing to be used alongside PtDAs, but there is inadequate evidence on the added effectiveness compared to PtDAs alone. The decision coaching definition was updated with minimal elements.

Achana, Felix; Gallacher, Daniel; Oppong, Raymond; Kim, Sungwook; Petrou, Stavros; Mason, James; Crowther, Michael

doi: 10.1177/0272989X211003880pmid: 33813933

Economic evaluations conducted alongside randomized controlled trials are a popular vehicle for generating high-quality evidence on the incremental cost-effectiveness of competing health care interventions. Typically, in these studies, resource use (and by extension, economic costs) and clinical (or preference-based health) outcomes data are collected prospectively for trial participants to estimate the joint distribution of incremental costs and incremental benefits associated with the intervention. In this article, we extend the generalized linear mixed-model framework to enable simultaneous modeling of multiple outcomes of mixed data types, such as those typically encountered in trial-based economic evaluations, taking into account correlation of outcomes due to repeated measurements on the same individual and other clustering effects. We provide new wrapper functions to estimate the models in Stata and R by maximum and restricted maximum quasi-likelihood and compare the performance of the new routines with alternative implementations across a range of statistical programming packages. Empirical applications using observed and simulated data from clinical trials suggest that the new methods produce broadly similar results as compared with Stata’s merlin and gsem commands and a Bayesian implementation in WinBUGS. We highlight that, although these empirical applications primarily focus on trial-based economic evaluations, the new methods presented can be generalized to other health economic investigations characterized by multivariate hierarchical data structures.

Mott, David J.; Shah, Koonal K.; Ramos-Goñi, Juan Manuel; Devlin, Nancy J.; Rivero-Arias, Oliver

doi: 10.1177/0272989X21999607pmid: 33733920

BackgroundAn important question in the valuation of children’s health is whether the preferences of younger individuals should be captured within value sets for measures that are aimed at them. This depends on whether younger individuals can complete valuation exercises and whether their preferences differ from those of adults. This study compared the preferences of adults and adolescents for EQ-5D-Y-3L health states using latent scale values elicited from a discrete choice experiment (DCE).MethodsAn online DCE survey, comprising 15 pairwise choices, was provided to samples of UK adults and adolescents (aged 11–17 y). Adults considered the health of a 10-year-old child, whereas adolescents considered their own health. Mixed logit models were estimated, and comparisons were made using relative attribute importance (RAI) scores and a pooled model.ResultsIn total, 1000 adults and 1005 adolescents completed the survey. For both samples, level 3 in pain/discomfort was most important, and level 2 in self-care the least important, based on the relative magnitudes of coefficients. The RAI scores (normalized on self-care) indicated that adolescents gave less weight relative to adults to usual activities (1.18 v. 1.51; P < 0.05), pain/discomfort (1.77 v. 3.12; P < 0.01), and anxiety/depression (1.64 vs. 2.65; P < 0.01). The pooled model indicated evidence of differences between the two samples in both levels in pain/discomfort and anxiety/depression.LimitationsThe perspective of the DCE task differed between the 2 samples, and no data were collected to anchor the DCE data to generate value sets.ConclusionsAdolescents could complete the DCE, and their preferences differed from those of adults taking a child perspective. It is important to consider whether their preferences should be incorporated into value sets.

Mieras, Adinda; Becker-Commissaris, Annemarie; Klop, Hanna T.; Pasman, H. Roeline W.; de Jong, Denise; Pronk, Lemke; Onwuteaka-Philipsen, Bregje D.

doi: 10.1177/0272989X21998951pmid: 33783264

BackgroundPrevious studies have investigated patients’ treatment goals before starting a treatment for metastatic lung cancer. Data on the evaluation of treatment goals are lacking.AimTo determine if patients with metastatic lung cancer and their oncologists perceive the treatment goals they defined at the start of systemic treatment as achieved after treatment and if in hindsight they believe it was the right decision to start systemic therapy.Design and ParticipantsA prospective multicenter study in 6 hospitals across the Netherlands between 2016 and 2018. Following systemic treatment, 146 patients with metastatic lung cancer and 23 oncologists completed a questionnaire on the achievement of their treatment goals and whether they made the right treatment decision. Additional interviews with 15 patients and 5 oncologists were conducted.ResultsAccording to patients and oncologists, treatment goals were achieved in 30% and 37% for ‘quality of life,’ 49% and 41% for ‘life prolongation,’ 26% and 44% for ‘decrease in tumor size,’ and 44% for ‘cure’, respectively. Most patients and oncologists, in hindsight, felt they had made the right decision to start treatment and also if they had not achieved their goals (72% and 93%). This was related to the feeling that they had to do ‘something.’ConclusionsBefore deciding on treatment, the treatment options, including their benefits and side effects, and the goals patients have should be discussed. It is key that these discussions include not only systemic treatment but also palliative care as effective options for doing ‘something.’

Revicki, Dennis A.; King, Madeleine T.; Viney, Rosalie; Pickard, A. Simon; Mercieca-Bebber, Rebecca; Shaw, James W.; Müller, Fabiola; Norman, Richard

doi: 10.1177/0272989X211003569pmid: 33813946

BackgroundThe EORTC QLU-C10D is a multiattribute utility measure derived from the cancer-specific quality-of-life questionnaire, the EORTC QLQ-C30. The QLU-C10D contains 10 dimensions (physical, role, social and emotional functioning, pain, fatigue, sleep, appetite, nausea, bowel problems). The objective of this study was to develop a United States value set for the QLU-C10D.MethodsA US online panel was quota recruited to achieve a representative sample for sex, age (≥18 y), race, and ethnicity. Respondents undertook a discrete choice experiment, each completing 16 choice-pairs, randomly assigned from a total of 960 choice-pairs. Each pair included 2 QLU-C10D health states and duration. Data were analyzed using conditional logistic regression, parameterized to fit the quality-adjusted life-year framework. Utility weights were calculated as the ratio of each dimension-level coefficient to the coefficient for life expectancy.ResultsA total of 2480 panel members opted in, 2333 (94%) completed at least 1 choice-pair, and 2273 (92%) completed all choice-pairs. Within dimensions, weights were generally monotonic. Physical functioning, role functioning, and pain were associated with the largest utility weights. Cancer-specific dimensions, such as nausea and bowel problems, were associated with moderate utility decrements, as were general issues such as problems with emotional functioning and social functioning. Sleep problems and fatigue were associated with smaller utility decrements. The value of the worst health state was 0.032, which was slightly greater than 0 (equivalent to being dead).ConclusionsThis study provides the US-specific value set for the QLU-C10D. These estimated health state scores, based on responses to the EORTC QLQ-C30 questionnaire, can be used to evaluate the cost-utility of oncology treatments.

Shiroiwa, Takeru; Ikeda, Shunya; Noto, Shinichi; Fukuda, Takashi; Stolk, Elly

doi: 10.1177/0272989X211001859pmid: 33754886

BackgroundEQ-5D-Y is a preference-based measure for children and adolescents (aged 8–15 y). This is the first study to develop an EQ-5D-Y value set for converting EQ-5D-Y responses to index values.MethodsWe recruited 1047 respondents (aged 20–79 y) from the general population, stratified by gender and age group, in 5 Japanese cities. All data were collected through face-to-face surveys. Respondents were asked to value EQ-5D-Y states for a hypothetical 10-y-old child from a proxy perspective using composite time tradeoff (cTTO) and a discrete choice experiment (DCE). The discrete choice data were analyzed using a mixed logit model. Latent DCE values were then converted to a 0 (death)/1 (full health) scale by mapping them to the cTTO values.ResultsThe mean observed cTTO value of the worst health state [33333] was 0.20. Analysis of the DCE data showed that the coefficients of the domains related to mental functions (“Having pain or discomfort” and “Feeling worried, sad, or unhappy”) were larger than those for the domains related to physical and social functions. By converting latent DCE values to a utility scale, we constructed a value set for EQ-5D-Y. No inconsistencies were observed. The minimum predicted score was 0.288 [33333], and the second-best score was 0.957 [12111].ConclusionA value set for EQ-5D-Y was successfully constructed. This is the first survey of an EQ-5D-Y value set. Interpreting the differences between EQ-5D-Y and EQ-5D-5L value sets is a future task with implications for health care policy.

van Delft, Frederik; Muller, Mirte; Langerak, Rom; Koffijberg, Hendrik; Retèl, Valesca; van den Broek, Daan; IJzerman, Maarten

doi: 10.1177/0272989X211002756pmid: 33813943

BackgroundAlthough immunotherapy (IMT) provides significant survival benefits in selected patients, approximately 10% of patients experience (serious) immune-related adverse events (irAEs). The early detection of adverse events will prevent irAEs from progressing to severe stages, and routine testing for irAEs has become common practice. Because a positive test outcome might indicate a clinically manifesting irAE that requires treatment to (temporarily) discontinue, the occurrence of false-positive test outcomes is expected to negatively affect treatment outcomes. This study explores how the UPPAAL modeling environment can be used to assess the impact of test accuracy (i.e., test sensitivity and specificity), on the probability of patients entering palliative care within 11 IMT cycles.MethodsA timed automata-based model was constructed using real-world data and expert consultation. Model calibration was performed using data from 248 non–small-cell lung cancer patients treated with nivolumab. A scenario analysis was performed to evaluate the effect of changes in test accuracy on the probability of patients transitioning to palliative care.ResultsThe constructed model was used to estimate the cumulative probabilities for the patients’ transition to palliative care, which were found to match real-world clinical observations after model calibration. The scenario analysis showed that the specificity of laboratory tests for routine monitoring has a strong effect on the probability of patients transitioning to palliative care, whereas the effect of test sensitivity was limited.ConclusionWe have obtained interesting insights by simulating a care pathway and disease progression using UPPAAL. The scenario analysis indicates that an increase in test specificity results in decreased discontinuation of treatment due to suspicion of irAEs, through a reduction of false-positive test outcomes.