Eliciting preferences for alternative cancer drug treatments. The influence of framing, medium, and rater variables

Putting the Choice in Choice Tasks: Incorporating Preference Elicitation Tasks in Health Preference Research

Choice-based preference elicitation methods such as the discrete choice experiment (DCE) present hypothetical choices to respondents, with an expectation that these hypothetical choices accurately reflect a 'real world' health-related decision context and that consequently the choice data can be held to be a true representation of the respondent's health or treatment preferences. For this to be the case, careful consideration needs to be given to the format of the choice task in a choice experiment. The overarching aim of this paper is to highlight important aspects to consider when designing and 'setting up' the choice tasks to be presented to respondents in a DCE. This includes the importance of considering the potential impact of format (e.g. choice context, choice set presentation and size) as well as choice set content (e.g. labelled and unlabelled choice sets and inclusion of reference alternatives) and choice questions (stated choice versus additional questions designed to explore complete preference orders) on the preference estimates that are elicited from studies. We endeavoure to instil a holistic approach to choice task design that considers format alongside content, experimental design and analysis.

How to Present a Decision Object in Health Preference Research: Attributes and Levels, the Decision Model, and the Descriptive Framework

In health preference research (HPR) studies, data are generated by participants'/subjects' decisions. When developing an HPR study, it is therefore important to have a clear understanding of the components of a decision and how those components stimulate participant behavior. To obtain valid and reliable results, study designers must sufficiently describe the decision model and its components. HPR studies require a detailed examination of the decision criteria, detailed documentation of the descriptive framework, and specification of hypotheses. The objects that stimulate subjects' decisions in HPR studies are defined by attributes and attribute levels. Any limitations in the identification and presentation of attributes and levels can negatively affect preference elicitation, the quality of the HPR data, and study results. This practical guide shows how to link the HPR question to an underlying decision model. It covers how to (1) construct a descriptive framework that presents relevant characteristics of a decision object and (2) specify the research hypotheses. The paper outlines steps and available methods to achieve all this, including the methods' advantages and limitations.

Changing Mindsets About Side Effects of the COVID-19 Vaccination: A Randomized Controlled Trial

BACKGROUND

Side-effect concerns are a major barrier to vaccination against COVID-19 and other diseases. Identifying cost- and time-efficient interventions to improve vaccine experience and reduce vaccine hesitancy-without withholding information about side effects-is critical.

PURPOSE

Determine whether a brief symptom as positive signals mindset intervention can improve vaccine experience and reduce vaccine hesitancy after the COVID-19 vaccination.

METHODS

English-speaking adults (18+) were recruited during the 15-min wait period after receiving their second dose of the Pfizer COVID-19 vaccination and were randomly allocated to the symptom as positive signals mindset condition or the treatment as usual control. Participants in the mindset intervention viewed a 3:43-min video explaining how the body responds to vaccinations and how common side effects such as fatigue, sore arm, and fever are signs that the vaccination is helping the body boost immunity. The control group received standard vaccination center information.

RESULTS

Mindset participants (N = 260) versus controls (N = 268) reported significantly less worry about symptoms at day 3 [t(506)=2.60, p=.01, d=0.23], fewer symptoms immediately following the vaccine [t(484)=2.75, p=.006, d=0.24], and increased intentions to vaccinate against viruses like COVID-19 in the future [t(514)=-2.57, p=.01, d=0.22]. No significant differences for side-effect frequency at day 3, coping, or impact.

CONCLUSIONS

This study supports the use of a brief video aimed at reframing symptoms as positive signals to reduce worry and increase future vaccine intentions.

CLINICAL TRIAL INFORMATION

Australian New Zealand Clinical Trials Registry: ACTRN12621000722897p.

A systematic review of the effects of positive versus negative framing on cancer treatment decision making

OBJECTIVE

Presenting treatment outcomes positively or negatively may differently influence treatment preferences and lead to sub-optimal decision in a medical context. This review systematically organised how positive versus negative framing of treatment outcomes influenced cancer treatment decisions of cancer patients and individuals without a cancer diagnosis.

DESIGN

Three databases (PubMed, PsycInfo and Scopus) were searched for studies reporting the effects of positive versus negative framing on cancer treatment decision-making from 1981 to December 2020.

MAIN OUTCOME MEASURE

The effects of positive versus negative framing on cancer treatment preferences and the elimination of framing effect were evaluated.

RESULTS

A total of 12 studies that met inclusion criteria were reviewed. Framing effect was consistently observed in individuals without a cancer diagnosis. There was not enough evidence to suggest a robust framing effect in cancer patients. Surgery was preferred in positive framing, whereas adjuvant therapy was preferred in negative framing. Justification intervention significantly eliminated framing effect. Mixed framing failed to eliminate framing effect.

CONCLUSION

Current recommendations for presenting treatment options are based on research in cancer-screening decision-making. Knowledge of how positive versus negative framing affect cancer patients' treatment decisions is still limited. Our review highlighted the need for continued research in this area.

Evaluation of adverse drug reaction formatting in drug information databases

Objective:

The research evaluated the differences in formatting of adverse drug reaction (ADR) information in drug monographs in commonly used drug information (DI) databases.

Methods:

A cross-sectional analysis of formatting of ADR information for twenty commonly prescribed oral medications in seven commonly used DI databases was performed. Databases were assessed for presentation of ADR information, including presence of placebo comparisons, severity of ADR, onset of ADR, formatting of ADRs in percentile (quantitative) format or qualitative format, whether references were used to cite information, whether ADRs are grouped by organ system, and word count of the ADR section. Data were collected by two study investigators and discrepancies were resolved via consensus. Chi-square analyses and one-way analysis of variance (ANOVA) were used to evaluate for mean group differences in categorical and continuous data, respectively.

Results:

The seven DI databases varied significantly on each analyzed ADR variable, including variables known to impact interpretation such as placebo comparisons and qualitative versus quantitative formatting. Placebo comparisons were most common among monographs in Micromedex In-Depth Answers (70%) but were absent among monographs in Epocrates, Lexicomp, and Micromedex. Quantitative information was commonly used in most databases but was absent in Epocrates. Average word counts were higher in Clinical Pharmacology and Micromedex In-Depth answers compared to other databases.

Conclusion:

Substantial variation in ADR formatting exists between the most common DI databases. These differences may translate into alternative interpretations of medical information and, thus, impact clinical judgment. Further studies are needed to assess whether these differences impact clinical practice.

Framing of clinical information affects physicians' diagnostic accuracy

BACKGROUND

Framing bias occurs when people make a decision based on the way the information is presented, as opposed to just on the facts themselves. How the diagnostician sees a problem may be strongly influenced by the way it is framed. Does framing bias result in clinically meaningful diagnostic error?

METHODS

We created three hypothetical cases and asked consultants and registrars in Emergency Medicine and Internal Medicine to provide their differential diagnoses and investigations list. Two of the presentations were written two ways to frame the case towards or away from a particular diagnosis (Presentation 2 - pulmonary embolus (PE) and Presentation 3 - interstitial lung disease (ILD)) and these were randomly assigned to the participants. Both versions were however entirely identical in terms of the objective facts. Physician impressions and diagnostic plan were compared. A third presentation was identical for all and served as a control for clinician baseline 'risk-averseness'.

RESULTS

There were significant differences in the differential diagnoses generated depending on the presentation's framing. PE and ILD were considered and investigated for the majority of the time when the presentation was framed towards these diagnoses, and the minority of the time when it was not. This finding was most striking in Presentation 2, where 100%versus50% of clinicians considered PE in their diagnosis when the presentation was framed towards PE. This result remained robust when undertaking stratified analysis and logistic regression to account for differences in seniority and baseline risk-averseness- neither of the latter variables had any effect on the result.

CONCLUSION

We demonstrate a clinically meaningful effect of framing bias on diagnostic error. The strength of our study is focus on clinically meaningful outcomes: investigations ordered. This finding has implications for the way we conduct handovers and teach juniors to communicate clinical information.

The choice dilemma in chronic hematological conditions: Why choosing is not only a medical issue? A psycho-cognitive perspective

Research in cognitive psychology focused on risk perception and decision making was shown to facilitate treatment choice and patient's satisfaction with decision in a number of medical conditions, increasing perceived alliance between patient and physician, and adherence to treatment. However, this aspect has been mostly neglected in the literature investigating choice of treatment for chronic hematological conditions. In this paper, a patient centered model and a shared decision making (SDM) approach to treatment switch in chronic hematological conditions, in particular chronic myeloid leukemia, atrial fibrillation, and β-thalassemia is proposed. These pathologies have a series of implications requiring important decisions about new available treatments. Although new generation treatments may provide a significant improvement in patient's health and health-related quality of life (HrQoL), a significant percentage of them is uncertain about or refuse treatment switch, even when strongly suggested by healthcare guidelines. Possible cognitive and emotional factors which may influence decision making in this field and may prevent appropriate risk-and-benefits evaluation of new treatment approaches are reviewed. Possible adaptive strategies to improve quality of care, patient participation, adherence to treatment and final satisfaction are proposed, and implications relatively to new treatment options available are discussed.

Treatment preference and patient centered prostate cancer care: Design and rationale

Prostate cancer is a slow progressing cancer that affects millions of men in the US. Due to uncertainties in outcomes and treatment complications, it is important that patients engage in informed decision making to choose the "optimal treatment". Patient centered care that encompasses informed decision-making can improve treatment choice and quality of care. Thus, assessing patient treatment preferences is critical for developing an effective decision support system. The objective of this patient-centered randomized clinical trial was to study the comparative effectiveness of a conjoint analysis intervention compared to usual care in improving subjective and objective outcomes in prostate cancer patients. We identified preferred attributes of alternative prostate cancer treatments that will aid in evaluating attributes of treatment options. In this two-phase study, in Phase 1 we used mixed methods to develop an adaptive conjoint task instrument. The conjoint task required the patients to trade-off attributes associated with treatments by assessing their relative importance. Phase 2 consisted of a randomized controlled trial of men with localized prostate cancer. We analyzed the effect of conjoint task intervention on the association between preferences, treatment and objective and subjective outcomes. Our conjoint task instrument can lead to a values-based patient-centered decision aid tool and help tailor treatment decision making to the values of prostate cancer patients. This will ultimately improve clinical decision making, clinical policy process, enhance patient centered care and improve prostate cancer outcomes.

Cognitive biases and heuristics in medical decision making: a critical review using a systematic search strategy

BACKGROUND

The role of cognitive biases and heuristics in medical decision making is of growing interest. The purpose of this study was to determine whether studies on cognitive biases and heuristics in medical decision making are based on actual or hypothetical decisions and are conducted with populations that are representative of those who typically make the medical decision; to categorize the types of cognitive biases and heuristics found and whether they are found in patients or in medical personnel; and to critically review the studies based on standard methodological quality criteria.

METHOD

Data sources were original, peer-reviewed, empirical studies on cognitive biases and heuristics in medical decision making found in Ovid Medline, PsycINFO, and the CINAHL databases published in 1980-2013. Predefined exclusion criteria were used to identify 213 studies. During data extraction, information was collected on type of bias or heuristic studied, respondent population, decision type, study type (actual or hypothetical), study method, and study conclusion.

RESULTS

Of the 213 studies analyzed, 164 (77%) were based on hypothetical vignettes, and 175 (82%) were conducted with representative populations. Nineteen types of cognitive biases and heuristics were found. Only 34% of studies (n = 73) investigated medical personnel, and 68% (n = 145) confirmed the presence of a bias or heuristic. Each methodological quality criterion was satisfied by more than 50% of the studies, except for sample size and validated instruments/questions. Limitations are that existing terms were used to inform search terms, and study inclusion criteria focused strictly on decision making.

CONCLUSIONS

Most of the studies on biases and heuristics in medical decision making are based on hypothetical vignettes, raising concerns about applicability of these findings to actual decision making. Biases and heuristics have been underinvestigated in medical personnel compared with patients.

Does framing of cancer survival affect perceived value of care? A willingness-to-pay survey of US residents

AIMS

To investigate the relationship between the framing of survival gains and the perceived value of cancer care.

METHODS

Through a population-based survey of 2040 US adults, respondents were randomized to one of the two sets of hypothetical scenarios, each of which described the survival benefit for a new treatment as either an increase in median survival time (median survival), or an increase in the probability of survival for a given length of time (landmark survival). Each respondent was presented with two randomly selected scenarios with different prognosis and survival improvements, and asked about their willingness to pay (WTP) for the new treatments.

RESULTS

Predicted WTP increased with survival benefits and respondents' income, regardless of how survival benefits were described. Framing therapeutic benefits as improvements in landmark rather than median time survival increased the proportion of the population willing to pay for that gain by 11-35%, and the mean WTP amount by 42-72% in the scenarios we compared.

CONCLUSION

How survival benefits are described may influence the value people place on cancer care.

Is a picture worth a thousand words? The interaction of visual display and attribute representation in attenuating framing bias

The attribute framing bias is a well-established phenomenon, in which an object or an event is evaluated more favorably when presented in a positive frame such as “the half full glass” than when presented in the complementary negative framing. Given that previous research showed that visual aids can attenuate this bias, the current research explores the factors underlying the attenuating effect of visual aids. In a series of three experiments, we examined how attribute framing bias is affected by two factors: (a) The display mode—verbal versus visual; and (b) the representation of the critical attribute—whether one outcome, either the positive or the negative, is represented or both outcomes are represented. In Experiment 1 a marginal attenuation of attribute framing bias was obtained when verbal description of either positive or negative information was accompanied by corresponding visual representation. In Experiment 2 similar marginal attenuation was obtained when both positive and negative outcomes were verbally represented. In Experiment 3, where the verbal description represented both positive and negative outcomes, significant attenuation was obtained when it was accompanied by a visual display that represented a single outcome, and complete attenuation, totally eliminating the framing bias, was obtained when it was accompanied by a visual display that represented both outcomes. Thus, our findings showed that interaction between the display mode and the representation of the critical attribute attenuated the framing bias. Theoretical and practical implications of the interaction between verbal description, visual aids and representation of the critical attribute are discussed, and future research is suggested.

Framing of health information messages

BACKGROUND

The same information about the evidence on health effects can be framed either in positive words or in negative words. Some research suggests that positive versus negative framing can lead to different decisions, a phenomenon described as the framing effect. Attribute framing is the positive versus negative description of a specific attribute of a single item or a state, for example, "the chance of survival with cancer is 2/3" versus "the chance of mortality with cancer is 1/3". Goal framing is the description of the consequences of performing or not performing an act as a gain versus a loss, for example, "if you undergo a screening test for cancer, your survival will be prolonged" versus "if you don't undergo screening test for cancer, your survival will be shortened".

OBJECTIVES

To evaluate the effects of attribute (positive versus negative) framing and of goal (gain versus loss) framing of the same health information, on understanding, perception of effectiveness, persuasiveness, and behavior of health professionals, policy makers, and consumers.

SEARCH METHODS

We searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, issue 3 2007), MEDLINE (Ovid) (1966 to October 2007), EMBASE (Ovid) (1980 to October 2007), PsycINFO (Ovid) (1887 to October 2007). There were no language restrictions. We reviewed the reference lists of related systematic reviews, included studies and of excluded but closely related studies. We also contacted experts in the field.

SELECTION CRITERIA

We included randomized controlled trials, quasi-randomised controlled trials, and cross-over studies with health professionals, policy makers, and consumers evaluating one of the two types of framing.

DATA COLLECTION AND ANALYSIS

Two review authors extracted data in duplicate and independently. We graded the quality of evidence for each outcome using the GRADE approach. We standardized the outcome effects using standardized mean difference (SMD). We stratified the analysis by the type of framing (attribute, goal) and conducted pre-planned subgroup analyses based on the type of message (screening, prevention, and treatment). The primary outcome was behaviour. We did not assess any adverse outcomes.

MAIN RESULTS

We included 35 studies involving 16,342 participants (all health consumers) and reporting 51 comparisons.In the context of attribute framing, participants in one included study understood the message better when it was framed negatively than when it was framed positively (1 study; SMD -0.58 (95% confidence interval (CI) -0.94 to -0.22); moderate effect size; low quality evidence). Although positively-framed messages may have led to more positive perception of effectiveness than negatively-framed messages (2 studies; SMD 0.36 (95% CI -0.13 to 0.85); small effect size; low quality evidence), there was little or no difference in persuasiveness (11 studies; SMD 0.07 (95% CI -0.23 to 0.37); low quality evidence) and behavior (1 study; SMD 0.09 (95% CI -0.14 to 0.31); moderate quality evidence).In the context of goal framing, loss messages led to a more positive perception of effectiveness compared to gain messages for screening messages (5 studies; SMD -0.30 (95% CI -0.49 to -0.10); small effect size; moderate quality evidence) and may have been more persuasive for treatment messages (3 studies; SMD -0.50 (95% CI -1.04 to 0.04); moderate effect size; very low quality evidence). There was little or no difference in behavior (16 studies; SMD -0.06 (95% CI -0.15 to 0.03); low quality evidence). No study assessed the effect on understanding.

AUTHORS' CONCLUSIONS

Contrary to commonly held beliefs, the available low to moderate quality evidence suggests that both attribute and goal framing may have little if any consistent effect on health consumers' behaviour. The unexplained heterogeneity between studies suggests the possibility of a framing effect under specific conditions. Future research needs to investigate these conditions.

Effective or ineffective: attribute framing and the human papillomavirus (HPV) vaccine

OBJECTIVE

To experimentally test whether presenting logically equivalent, but differently valenced effectiveness information (i.e. attribute framing) affects perceived effectiveness of the human papillomavirus (HPV) vaccine, vaccine-related intentions and policy opinions.

METHODS

A survey-based experiment (N=334) was fielded in August and September 2007 as part of a larger ongoing web-enabled monthly survey, the Annenberg National Health Communication Survey. Participants were randomly assigned to read a short passage about the HPV vaccine that framed vaccine effectiveness information in one of five ways. Afterward, they rated the vaccine and related opinion questions. Main statistical methods included ANOVA and t-tests.

RESULTS

On average, respondents exposed to positive framing (70% effective) rated the HPV vaccine as more effective and were more supportive of vaccine mandate policy than those exposed to the negative frame (30% ineffective) or the control frame. Mixed valence frames showed some evidence for order effects; phrasing that ended by emphasizing vaccine ineffectiveness showed similar vaccine ratings to the negative frame.

CONCLUSION

The experiment finds that logically equivalent information about vaccine effectiveness not only influences perceived effectiveness, but can in some cases influence support for policies mandating vaccine use.

PRACTICE IMPLICATIONS

These framing effects should be considered when designing messages.

How to reduce the effect of framing on messages about health

BACKGROUND

Patients must be informed about risks before any treatment can be implemented. Yet serious problems in communicating these risks occur because of framing effects.

OBJECTIVE

To investigate the effects of different information frames when communicating health risks to people with high and low numeracy and determine whether these effects can be countered or eliminated by using different types of visual displays (i.e., icon arrays, horizontal bars, vertical bars, or pies).

DESIGN

Experiment on probabilistic, nationally representative US (n = 492) and German (n = 495) samples, conducted in summer 2008.

OUTCOME MEASURES

Participants' risk perceptions of the medical risk expressed in positive (i.e., chances of surviving after surgery) and negative (i.e., chances of dying after surgery) terms.

KEY RESULTS

Although low-numeracy people are more susceptible to framing than those with high numeracy, use of visual aids is an effective method to eliminate its effects. However, not all visual aids were equally effective: pie charts and vertical and horizontal bars almost completely removed the effect of framing. Icon arrays, however, led to a smaller decrease in the framing effect.

CONCLUSIONS

Difficulties with understanding numerical information often do not reside in the mind, but in the representation of the problem.

Media response to colon cancer campaigns in Switzerland 2005-2007: regional newspapers are the most reliable among the printed media

Background

Health campaigns are frequently covered by printed media, but coverage is not homogeneous across different types of newspapers. Switzerland as a multilinguistic country with many newspapers offers a good field for study. A better understanding of how printed media report on national campaigns against colon cancer in the three main linguistic regions may help to improve future public health interventions. Therefore, we analyzed articles published between 2005 and 2007 during the campaigns "Darmkrebs-nie?" and "Self-Care" in the German, French and Italian regions of Switzerland.

Findings

Some 65% of articles reporting on colon cancer were in German, 23% and 12% were in French and Italian respectively. During the campaign, topics linked to colon cancer were increasingly covered by the media. Regional newspapers (66%) reported significantly more about colon cancer and produced the most detailed articles.

Both gain- and loss-framed messages have been used by journalists, whereas the campaigns used merely gain-framed messages. Latin (French and Italian) newspapers mixed gain- and loss-framed messages in the same articles, while German articles mainly used a single frame throughout.

Conclusions

Swiss-German papers reported more about the topic and the reporting was quantitatively and qualitatively more prominent in regional papers. The press followed the campaigns closely only during the period of campaigning, with high coverage. We propose to consider the regional press as an important vehicle of health information. Moreover, slight differences in framing can be observed between German and Latin articles.

Dual equipoise shared decision making: definitions for decision and behaviour support interventions

BACKGROUND

There is increasing interest in interventions that can support patients who face difficult decisions and individuals who need to modify their behaviour to achieve better outcomes. Evidence for effectiveness is used to categorize patients care. Effective care is where evidence of benefit outweighs harm: patients should always receive this type of care, where indicated. Preference-sensitive care describes a situation where the evidence for the superiority of one treatment over another is either not available or does not allow differentiation; in this situation, there are two or more valid approaches, and the best choice depends on how individuals value the risks and benefits of treatments.

DISCUSSION

Preference-sensitive decisions are defined by equipoise: situations where options need to be deliberated. Moreover, where both healthcare professionals and patients agree that equipoise exists, situations may be regarded as having 'dual equipoise'. Such conditions are ideal for shared decision making. However, there are many situations in medicine where dual equipoise does not exist, where health professionals hold the view that scientific evidence for benefit strongly outweighs harm. This is often the case where people suffer from chronic conditions, and where behaviour change is recommended to improve outcomes. However, some patients, are either ambivalent or find it difficult to sustain optimal behaviours, i.e., patients will be in varying degrees of equipoise. Therefore, situations where dual equipoise exists (or not) help to clarify the definitions of two classes of support, namely, decision and behaviour change support interventions. Decision support interventions help people think about choices they face; they describe where and why choice exists, in short, conditions of dual equipoise; they provide information about options, including, where reasonable, the option of taking no action. These interventions help people to deliberate, independently or in collaboration with others, about options by considering relevant attributes; they support people to forecast how they might feel about short, intermediate, and long-term outcomes that have relevant consequences, in ways that help the process of constructing preferences and eventual decision making appropriate to their individual situation. Whereas, behavioural support interventions describe, justify, and recommend actions that, over time, lead to predictable outcomes over short, intermediate, and long-term timeframes, and that have relevant and important consequences for those who are considering behaviour change.

SUMMARY

Decision and behaviour support interventions have divergent aims, different relationships to equipoise, and form two classes of interventions.

How should the impact of different presentations of treatment effects on patient choice be evaluated? A pilot randomized trial

Background

Different presentations of treatment effects can affect decisions. However, previous studies have not evaluated which presentations best help people make decisions that are consistent with their own values. We undertook a pilot study to compare different methods for doing this.

Methods and Findings

We conducted an Internet-based randomized trial comparing summary statistics for communicating the effects of statins on the risk of coronary heart disease (CHD). Participants rated the relative importance of treatment consequences using visual analogue scales (VAS) and category rating scales (CRS) with five response options. We randomized participants to either VAS or CRS first and to one of six summary statistics: relative risk reduction (RRR) and five absolute measures of effect: absolute risk reduction, number needed to treat, event rates, tablets needed to take, and natural frequencies (whole numbers). We used logistic regression to determine the association between participants' elicited values and treatment choices. 770 participants age 18 or over and literate in English completed the study. In all, 13% in the VAS-first group failed to complete their VAS rating, while 9% of the CRS-first group failed to complete their scoring (p = 0.03). Different ways of weighting the elicited values had little impact on the analyses comparing the different presentations. Most (51%) preferred the RRR compared to the other five summary statistics (1% to 25%, p = 0.074). However, decisions in the group presented the RRR deviated substantially from those made in the other five groups. The odds of participants in the RRR group deciding to take statins were 3.1 to 5.8 times that of those in the other groups across a wide range of values (p = 0.0007). Participants with a scientific background, who were more numerate or had more years of education were more likely to decide not to take statins.

Conclusions

Internet-based trials comparing different presentations of treatment effects are feasible, but recruiting participants is a major challenge. Despite a slightly higher response rate for CRS, VAS is preferable to avoid approximation of a continuous variable. Although most participants preferred the RRR, participants shown the RRR were more likely to decide to take statins regardless of their values compared with participants who were shown any of the five other summary statistics.

Trial Registration

Controlled-Trials.com ISRCTN85194921

Framing effect debiasing in medical decision making

OBJECTIVE

Numerous studies have demonstrated the robustness of the framing effect in a variety of contexts. The present study investigated the effects of a debiasing procedure designed to prevent the framing effect for young adults who made decisions based on hypothetical medical decision-making vignettes.

METHODS

The debiasing technique involved participants listing advantages and disadvantages of each treatment prior to making a choice. One hundred and two undergraduate students read a set of three medical treatment vignettes that presented information in terms of different outcome probabilities under either debiasing or control conditions.

RESULTS

The framing effect was demonstrated by the control group in two of the three vignettes. The debiasing group successfully avoided the framing effect for both of these vignettes.

CONCLUSION

These results further support previous findings of the framing effect as well as an effective debiasing technique. This study improved upon previous framing debiasing studies by including a control group and personal medical scenarios, as well as demonstrating debiasing in a framing condition in which the framing effect was demonstrated without a debiasing procedure.

PRACTICE IMPLICATIONS

The findings suggest a relatively simple manipulation may circumvent the use of decision-making heuristics in patients.

Further insight into the perception of quantitative information: judgments of gist in treatment decisions

PURPOSE

To compare relative accuracy and relative response times (RTs) as well as impact of foreground and background colors in a treatment decision context of judging larger/smaller when the following elements are added to the graphics studied previously: 1) a number (the displayed percentage), 2) a referent scale, and 3) a number and a referent scale.

METHOD

An experiment compared pie charts, vertical bars, horizontal bars, digits, systematic ovals, and random ovals. On each trial, participants saw 2 percentages (in 1 format) and were asked to choose the larger chance of survival or the smaller chance of side effects. Outcomes were errors and RT. Formats were either black and white or blue and yellow; background color was either white or blue. Participants were 216 volunteers from the community older than 50 years.

RESULTS

Formats with a number produced the same relative errors and relative RT as the formats with a number and scale. Formats with only a scale, however, shifted relative performance: Errors increased with more difficult formats (pie charts and random ovals by 3%-4% v. approximately 1% with other formats), but RT decreased with easier formats (vertical bars, horizontal bars, and systematic ovals decreased 100-200 ms v. an increase of 0-300 ms with other formats). Vertical bars with scales were the fastest and most accurately processed. Neither foreground nor background color had any impact on either outcome.

CONCLUSIONS

For supporting older people's judgments of relative extent, risk information is best presented using vertical bars with a scale; the format systematic ovals with a scale are among the next most easily processed.

The influence of the ratio bias phenomenon on the elicitation of health states utilities

This paper tests whether logically equivalent risk formats can lead to different health state utilities elicited by means of the traditional standard gamble (SG) method and a modified version of the method that we call “double lottery.” We compare utilities for health states elicited when probabilities are framed in terms of frequencies with respect to 100 people in the population (i.e., X out of 100 who follow a medical treatment will die) with SG utilities elicited for frequencies with respect to 1,000 people in the population (i.e., Y out of 1,000 who follow a medical treatment will die). We found that people accepted a lower risk of death when success and failure probabilities were framed as frequencies type “Y deaths out of 1,000” rather than as frequencies type “X deaths out of 100” and hence the utilities for health outcomes were higher when the denominator was 1000 than when it was 100. This framing effect, known as Ratio Bias, may have important consequences in resource allocation decisions.

Results of a Survey of Hospital Coagulation Laboratories in the United States, 2001

Context.—Coagulation and bleeding problems are associated with substantial morbidity and mortality, and inappropriate testing practices may lead to bleeding or thrombotic complications.

Objective.—To evaluate practices reported by hospital coagulation laboratories in the United States and to determine if the number of beds in a hospital was associated with different practices.

Design.—From a sampling frame of institutions listed in the 1999 directory of the American Hospital Association, stratified into hospitals with 200 or more beds (“large hospitals”) and those with fewer than 200 beds (“small hospitals”), we randomly selected 425 large hospitals (sampling rate, 25.6%) and 375 small hospitals (sampling rate, 8.8%) and sent a survey to them between June and October 2001. Of these, 321 large hospitals (75.5%) and 311 small hospitals (82.9%) responded.

Results.—An estimated 97.1% of respondents reported performing some coagulation laboratory tests. Of these, 71.6% reported using 3.2% sodium citrate as the specimen anticoagulant to determine prothrombin time (81.3% of large vs 67.7% of small hospitals, P < .001). Of the same respondents, 45.3% reported selecting thromboplastins insensitive to heparin in the therapeutic range when measuring prothrombin time (59.4% of large vs 39.8% of small hospitals, P < .001), and 58.8% reported having a therapeutic range for heparin (72.9% of large vs 53.2% of small hospitals, P < .001). An estimated 96.3% of respondents assayed specimens for activated partial thromboplastin time within 4 hours after phlebotomy, and 89.4% of respondents centrifuged specimens within 1 hour of collection. An estimated 12.1% reported monitoring low-molecular-weight heparin therapy, and to do so, 79% used an assay for activated partial thromboplastin time (58% of large vs 96% of small hospitals, P = .001), whereas 38% used an antifactor Xa assay (65% of large vs 18% of small hospitals, P = .001).

Conclusions.—Substantial variability in certain laboratory practices was evident. Where significant differences existed between the hospital groups, usually large hospitals adhered to accepted practice guidelines to a greater extent. Some reported practices are not consistent with current recommendations, showing a need to understand the reasons for noncompliance so that better adherence to accepted standards of laboratory practice can be promoted.

Testing the internal consistency of the standard gamble in ‘success’ and ‘failure’ frames

Decision making behaviour has often been shown to vary following changes in the way in which choice problems are described (or 'framed'). Moreover, a number of researchers have demonstrated that the standard gamble is prone to internal inconsistency, and loss aversion has been proposed as an explanation for this observed bias. This study attempts to alter the influence of loss aversion by framing the treatment arm of the standard gamble in terms of success (where we may expect the influence of loss aversion to be relatively weak) and in terms of failure (where we may expect the influence of loss aversion to be relatively strong). The objectives of the study are (1) to test whether standard gamble values vary when structurally identical gambles are differentially framed, and (2) to test whether the standard gamble is equally prone to internal inconsistency across the two frames. The results show that compared to framing in terms of treatment success, significantly higher values were inferred when the gamble was framed in terms of treatment failure. However, there was no difference in the quite marked levels of internal inconsistency observed in both frames. It is possible that the essential construct of the standard gamble induces substantial and/or widespread loss aversion irrespective of the way in which the gamble is framed, which offers a fundamental challenge to the usefulness of this value elicitation instrument. It is therefore recommended that further tests are undertaken on more sophisticated corrective procedures designed to limit the influence of loss aversion.

Describing treatment effects to patients

OBJECTIVE

To examine the impact of different presentations of equivalent information (framing) on treatment decisions faced by patients.

DESIGN

A systematic review of the published literature was conducted. English language publications allocating participants to different frames were retrieved using electronic and bibliographic searches. Two reviewers examined each article for inclusion, and assessed methodological quality. Study characteristics were tabulated and where possible, relative risks (RR; 95% confidence intervals) were calculated to estimate intervention effects.

MEASUREMENTS AND MAIN RESULTS

Thirty-seven articles, yielding 40 experimental studies, were included. Studies examined treatment (N = 24), immunization (N = 5), or health behavior scenarios (N = 11). Overall, active treatments were preferred when outcomes were described in terms of relative rather than absolute risk reductions or number needed to treat. Surgery was preferred to other treatments when treatment efficacy was presented in a positive frame (survival) rather than a negative frame (mortality) (relative risk [RR] = 1.51, 95% confidence interval [CI], 1.39 to 1.64). Framing effects were less obvious for immunization and health behavior scenarios. Those with little interest in the behavior at baseline were influenced by framing, particularly when information was presented as gains. In studies judged to be of good methodological quality and/or examining actual decisions, the framing effect, although still evident, was less convincing compared to the results of all included studies.

CONCLUSIONS

Framing effects varied with the type of scenario, responder characteristics, scenario manipulations, and study quality. When describing treatment effects to patients, expressing the information in more than one way may present a balanced view to patients and enable them to make informed decisions.

Australian women's prediagnostic decision-making styles, relating to treatment choices for early breast cancer treatment

Women diagnosed with early breast cancer are now asked by their doctors to choose from a range of options for their preferred medical treatment plan. Little information is known about women's treatment decision-making and therefore nurses do not have evidence to guide this decision support. The aim of this descriptive survey was to investigate the prediagnostic decision-making behavior of a sample (N = 377) of Australian women, regarding their treatment choices for early breast cancer. The data were collected using the Pre-Decision Portfolio Questionnaire (PDPQ) by Pierce (1996), which includes the Michigan Assessment of Decision Styles (MADS). Of 366 participating women, 19.9% strongly agreed to all three items of the MADS factor Deferring Responsibility; 0.3% strongly agreed to all four factors of Avoidance; 32.7% strongly agreed on all four items of Information Seeking; and 63.4% strongly agreed to all five items of Deliberation. Women showed a variety of preferred decision styles, depending on age, education, occupation and employment status. Only 36% of women indicated it was critically important to "get the treatment over as soon as possible;" 55% to "participate in selecting treatment;" and 53% to "read a lot of information:" The understanding of factors that are important to women when they are making decisions for medical treatment is a mandatory step in designing customized evidence-based decision support, which can be delivered by nurses to help women during this distressing experience.

Patient comprehension of information for shared treatment decision making: state of the art and future directions

How does the way that information is presented potentially influence patients' consent for health treatments, in a shared decision making process? The goal of this paper is to present an overview of selected recent literature concerning patient health information presentation/use for treatment decision making. Recent work with patient populations has begun to extend early cognitive psychological work showing systematic biases in thinking. Key research findings are organized by type of format (probability, graphic, and qualitative/quantitative dimensions). The applied literature on this topic is amenable to only limited integration in regard to key findings, and relatively few novel approaches to improving information comprehension have been described in the health literature. Promising approaches being proposed, developed, and tested are described, such as enhanced-access computerized patient choice modules, "debiasing" techniques, and tailoring of information. Additional theoretical and practical issues are discussed, as well as selected policy implications of current knowledge.

Cancer patients' attitudes toward treatment options for advanced non-small cell lung cancer: implications for patient education and decision support

The purpose of this study was to determine how people weigh both median survival time and 1-year survival probability when considering a choice between palliative Cisplatin-based chemotherapy with best supportive care (C+BSC) versus best supportive care alone (BSC) as treatment for advanced non-small cell lung cancer (NSCLC). Sixty people, previously treated for cancer, were interviewed as surrogate patients making a treatment decision. The interview included a structured description of the treatment options, and trade-off exercises used to clarify the participants' attitudes pertaining to the survival probabilities associated with each treatment.Participants' attitudes ranged from choosing the more toxic treatment if it offered no survival advantage to declining C+BSC no matter how large its advantage. Fifty-seven percent of participants would choose chemotherapy if the 1-year survival were 10% higher with C+BSC than with BSC alone. For 44 participants (76%), both their median survival and 1-year survival thresholds for accepting C+BSC were consistent, and for two (3%), neither threshold was consistent with their stated treatment preference. For the remaining 12 (21%), one threshold was discordant, but in all cases, this threshold was less relevant to his/her decision. Participants' thresholds could not be predicted reliably on the basis of patient age, sex, education, preferred role in treatment decision making, or previous treatment with chemotherapy. All but one participant recommended the interview as a decision-support strategy for actual patients. The findings suggest that patients with advanced NSCLC should be offered more than one treatment option, and that a systematic process for educating patients and for eliciting their preferences is desirable. The process described herein has potential for use in this clinical setting.

Preference for fractures and other glucocorticoid-associated adverse effects among rheumatoid arthritis patients

OBJECTIVE

The objective of this study was to determine rheumatoid arthritis (RA) patients' preferences for validated health state scenarios depicting glucocorticoid adverse events, predictors of these preferences, and psychometric properties of different preference techniques in this population.

METHODS

Preferences were elicited by rating scale and time trade-off methods. Time trade-offs included trading current health for either time spent alive in an adverse health state for chronic conditions (time trade-off) or time spent in a sleeplike state for acute conditions (sleep trade-off).

RESULTS

A total of 107 subjects with long-standing RA participated in the preference interviews. Mean preference values (rating scale/trade-off) were lowest for serious fracture adverse events, including hip fracture requiring a nursing home stay (0.55+/-0.22/0.76+/-0.36) and vertebral fracture with chronic pain (0.59+/-0.23/0.67+/-0.35), and highest for cataracts (0.84 + 0.17/0.96 0.09) and wrist fracture (0.82+/-0.18/0.81+/-0.29). Rating scales had a stronger correlation (r= 0.88) with physician ranking of scenarios than trade-off methods (r = 0.31). All methods were feasible and demonstrated good reliability, while rating scale method showed better construct validity than trade-off techniques.

CONCLUSION

Relative to their current health, RA patients assigned low preference values to many glucocorticoid adverse events, particularly those associated with chronic fracture outcomes. Results varied with the preference measure used, indicating that methodological attributes of preference determinations must be considered in clinical decision making.

Phase I study of a decision aid for patients with locally advanced non-small-cell lung cancer

PURPOSE

Many patients with locally advanced non-small-cell lung cancer (LA-NSCLC) are eligible for combined-modality therapy (CMT; chemotherapy and radiotherapy). Although CMT offers slightly higher chances of survival than radiotherapy alone (RT), it also carries a higher probability of toxicity, raising the possibility that some patients may prefer to decline CMT. We report a pilot study of a decision aid designed for patients in this setting.

PATIENTS AND METHODS

The aid included a structured description of the treatment options and trade-off exercises designed to help clarify the patient's values for the relevant outcomes by determining the patient's survival advantage threshold (SAT; the increase in survival conferred by CMT over RT that the patient deemed necessary for choosing CMT). Additional outcome measures included each patient's strength of treatment preference, decisional conflict, objective understanding of survival information, and decisional role preference.

RESULTS

Twenty-seven patients met the eligibility criteria for the study. Of these, seven declined the decision aid because they had a clear treatment preference. The remaining 20 participants completed the decision aid; 18 chose CMT, and two chose RT. All 20 patients wished to participate in the decision to some extent. All patients reported that using the decision support was useful to them and recommended its use for others. No patient or physician reported that the aid interfered with the physician-patient relationship. Patients' 3-year SATs and median SATs were each strongly correlated with their strengths of treatment preference (rho = 0.83, P <.001 and rho = 0.67, P =.02, respectively). For all but one patient, either their 3-year or median survival threshold was consistent with their final treatment choice. Ten patients reported a stronger treatment preference after using the decision aid.

CONCLUSION

We conclude that implementing the decision-aid for patients with LA-NSCLC is feasible, that it demonstrates convergent validity, and that it is favorably evaluated by patients and their physicians. The aid seems to help patients understand the benefits and risks of treatment and to choose the treatment that is most consistent with their values. Further evaluation of the aid is warranted.

A treatment trade-off based decision aid for patients with locally advanced non-small cell lung cancer

Purpose To describe the structure and use of a decision aid for patients with locally advanced non-small cell lung cancer (LA-NSCLC) who are eligible for combined-modality treatment (CMT) or for radiotherapy alone (RT). METHODS: The aid included a structured description of the treatment options and trade-off exercises designed to help clarify the patient's values for the relevant outcomes by determining the patient's survival advantage threshold (the increase in survival conferred by CMT over RT that the patient deemed necessary for choosing CMT). Additional outcome measures included each patient's strength of treatment preference, decisional conflict, objective understanding of survival information, decisional role preference, and evaluation of the aid itself. RESULTS: Twenty-five patients met the eligibility criteria for study. Of these, seven declined the decision aid because they had a clear treatment preference (four chose CMT and three chose RT). The remaining 18 participants completed the decision aid; 16 chose CMT and two chose RT. All 18 patients wished to participate in the decision to some extent. All patients reported that using the decision support was useful to them and recommended its use for others. No patient or physician reported that the aid interfered with the physician-patient relationship. Patients' 3-year survival advantage thresholds, and their median survival advantage thresholds, were each strongly correlated with their strengths of treatment preference (rho=0.80, P < 0.001 and rho=0.77, P < 0.001, respectively). For all but one patient, either their 3-year or median survival threshold was consistent with their final treatment choice. Eight patients reported a stronger treatment preference after using the decision aid. CONCLUSIONS: We conclude that a treatment trade-off based decision aid for patients with locally advanced non-small cell lung cancer is feasible, that it demonstrates internal consistency and convergent validity, and that it is favourably evaluated by patients and their physicians. The aid seems to help patients understand the benefits and risks of treatment and to choose the treatment that is most consistent with their values.

Framing effects in choices between multioutcome life-expectancy lotteries

AIM

To explore framing or editing effects and a method to debias framing in a clinical context.

METHOD

Clinical scenarios using multioutcome life-expectancy lotteries of equal value required choices between two supplementary drugs that either prolonged or shortened life from the 20-year beneficial effect of a baseline drug. The effects of these supplementary drugs were presented in two conditions, using a between-subjects design. In segregated editing (n = 116) the effects were presented separately from the effects of the baseline drug. In integrated editing (n = 100), effects of supplementary and baseline drugs were combined in the lottery presentation. Each subject responded to 30 problems. To explore one method of debiasing, another 100 subjects made choices after viewing both segregated and integrated editings of 20 problems (dual framing).

RESULTS

Statistically significant preference reversals between segregated and integrated editing of pure lotteries occurred only when one framing placed outcomes in the gain domain, and the other framing placed them in the loss domain. When both editings resulted in gain-domain outcomes only, there was no framing effect. There was a related relationship of framing-effect shifts from losses to gains in mixed-lottery-choice problems. Responses to the dual framing condition did not consistently coincide with responses to either single framing. In some situations, dual framing eliminated or lessened framing effects.

CONCLUSION

The results support two components of prospect theory, coding outcomes as gains or losses from a reference point, and an s-shaped utility function (concave in gain, convex in loss domains). Presenting both alternative editings of a complex situation prior to choice more fully informs the decision maker and may help to reduce framing effects. Given the extent to which preferences shift in response to alternative presentations, it is unclear which choice represents the subject's "true preferences."

Health values and prospect theory

Health values are important components of medical decisions. Experimental data suggest that people value health in complex and dynamic ways. Prospect theory is a descriptive theory of choice that may accurately characterize how people assign values to health states. The authors first provide background on prospect theory and how it can be applied to health values. Next, they review the relevant health research and find mixed support for prospect theory. Last, they discuss implications of prospect theory for cost-effectiveness analysis. The application of prospect theory to health deserves further research because it may help clarify the link between health and values.

Utility valuations for outcome states of colorectal cancer

OBJECTIVE

Utilities for the outcome states of colorectal cancer (CRC) must be measured to evaluate the cost-utility of screening and surveillance strategies for this disease. We sought to measure utilities for stage-dependent outcome states of CRC.

METHODS

We identified persons who had previously undergone removal of colorectal adenoma. We conducted individual interviews in which these participants were presented with stage-dependent outcome states and were asked to assess utilities for them using the standard gamble technique.

RESULTS

A total of 90 participants were interviewed; nine were excluded, leaving 81 for analysis. We obtained the following utility valuations: stage I rectal or stage I/II colon cancer (mean 0.74, median 0.75); stage III colon cancer (mean 0.67, median 0.75); stage II/III rectal cancer without ostomy (mean 0.59, median 0.60), stage II/III rectal cancer with ostomy (mean 0.50, median 0.55), stage IV rectal or colon cancer (mean 0.25, median 0.20). These valuations were statistically different from each other.

CONCLUSIONS

We measured utilities for stage-dependent outcome states of CRC in a sample of persons who had previously undergone removal of colorectal adenoma. We found that our participants were able to differentiate between the presented outcome states and assigned lower utility to increasingly morbid states. Our results show that stage-dependent morbidity is an important consideration in CRC and should be incorporated into any decision analysis model evaluating the cost-effectiveness of CRC screening or surveillance.

Third line chemotherapy in patients with metastatic breast cancer: an evaluation of quality of life and cost

Many patients with metastatic breast cancer receive several types of chemotherapy, although it is recognized that there is a declining probability of response. A major problem confronts oncologists in deciding when to recommend to patients that no further chemotherapy should be given. To address this problem we have assessed prospectively, health-related quality of life (HRQL) and costs of health care for 35 patients with metastatic breast cancer receiving third line chemotherapy in a representative clinical situation. HRQL and utilities were measured longitudinally using the EORTC QLQ-C30 questionnaire and the time trade-off method. Patients received a median of 2 cycles of chemotherapy and lived a median of 4.3 months. Twelve patients (34%) had substantial (> 10 points) improvement in the Global QL subscale and more than 30% of patients had similar changes in emotional and social function. The median baseline utility score was 0.9 and utilities correlated poorly with HRQL subscale. Eighteen patients had measurable disease and one patients experienced a partial response. Grade 3/4 toxicity occurred in 30% of patients. The average cost of management from study entry to death was CDN$ 17,260 (approximately US$ 12,000). Sixteen percent of this cost was associated directly with chemotherapy while hospital admissions and outpatient visits accounted for 50% and 14% of the total cost respectively. We conclude that: (a) many patients receiving third line chemotherapy maintain or improve indices of HRQL despite short survival and a low response rate: this might be due to chemotherapy, placebo effect, or a shift in frame of reference for HRQL; (b) patients were unwilling to trade quantity for quality of life; and (c) response rates and survival may be overestimated in patients selected for clinical trials.

The Influence of Framing on Risky Decisions: A Meta-analysis

In framing studies, logically equivalent choice situations are differently described and the resulting preferences are studied. A meta-analysis of framing effects is presented for risky choice problems which are framed either as gains or as losses. This evaluates the finding that highlighting the positive aspects of formally identical problems does lead to risk aversion and that highlighting their equivalent negative aspects does lead to risk seeking. Based on a data pool of 136 empirical papers that reported framing experiments with nearly 30,000 participants, we calculated 230 effect sizes. Results show that the overall framing effect between conditions is of small to moderate size and that profound differences exist between research designs. Potentially relevant characteristics were coded for each study. The most important characteristics were whether framing is manipulated by changing reference points or by manipulating outcome salience, and response mode (choice vs. rating/judgment). Further important characteristics were whether options differ qualitatively or quantitatively in risk, whether there is one or multiple risky events, whether framing is manipulated by gain/loss or by task-responsive wording, whether dependent variables are measured between- or within- subjects, and problem domains. Sample (students vs. target populations) and unit of analysis (individual vs. group) was not influential. It is concluded that framing is a reliable phenomenon, but that outcome salience manipulations, which constitute a considerable amount of work, have to be distinguished from reference point manipulations and that procedural features of experimental settings have a considerable effect on effect sizes in framing experiments. Copyright 1998 Academic Press.

Getting meaningful informed consent from older adults: a structured literature review of empirical research

OBJECTIVES

To perform a structured literature review of the published empirical research on informed consent with older adults in order to make recommendations to improve the informed consent process and to highlight areas needing further examination.

DESIGN

Relevant literature was identified by searching electronic databases (AGELINE, BIOETHICSLINE, CancerLit, Ethics Index, Health, LegalTrac, MEDLINE, PAIS International, PsycInfo, and Sociofile). Studies were included if they were reports of primary research data about informed consent and, if patients or other subjects were used, older subjects were included in the sample. Data related to the aspect of informed consent under study (recruitment, decision-making capacity, voluntariness, disclosure of information, understanding of information, consent forms, authorization, and policies and procedures) were abstracted and entered into a specially designed database.

MEASUREMENTS

Characterization of the population, age of subjects, setting, whether informed consent was being studied in the context of research or treatment, study design, the nature of outcome or dependent variables, independent variables (e.g., experimental conditions in a randomized controlled trial or patient/subject characteristics in a nonrandomized comparison), and results according to the aspect of informed consent under study.

RESULTS

A total of 99 articles met all the inclusion criteria and posed 289 unique research questions covering a wide range of aspects of informed consent: recruitment (60); decision making capacity (21); voluntariness (6); disclosure (30); understanding (139); consent forms (7); authorization (11); policies (13); and other (2). In the secondary analyses of numerous studies, diminished understanding of informed consent information was associated with older age and fewer years of education. Older age was also sometimes associated with decreased participation in research. Studies of disclosure of informed consent information suggest strategies to improve understanding and include a variety of novel formats (e.g., simplified, storybook, video) and procedures (e.g., use of health educators, quizzing subjects, multiple disclosure sessions).

CONCLUSIONS

A systematic review of the published literature on informed consent reveals evidence for impaired understanding of informed consent information in older subjects and those with less formal education. Effective strategies to improve the understanding of informed consent information should be considered when designing materials, forms, policies, and procedures for obtaining informed consent. Other than empirical research that has investigated disclosure and understanding of informed consent information, little systematic research has examined other aspects of the informed consent process. This deficit should be rectified to ensure that the rights and interests of patients and of human subjects who participate in research are adequately protected.

Framing effects on expectations, decisions, and side effects experienced: the case of influenza immunization

OBJECTIVE

To examine the effects of using positive or negative frames to describe influenza vaccine benefits and side effects on patients' expectations, decisions, decisional conflict, and reported side effects.

METHODS

292 previously unimmunized patients with chronic respiratory or cardiac disease were randomly assigned to receive benefit/risk information that was framed: (1) positively as the percentage who remain free of influenza and have no vaccine side effects, or (2) negatively as the percentage who acquire influenza and have vaccine side effects. Questionnaires elicited expectations, decisions, and decisional conflict. Vaccines were telephoned 3 days later for a self-report of local and systemic side effects and work absenteeism.

RESULTS

Both groups had similar immunization rates and decisional conflict scores. The positive frame group had lower and more realistic expectations of vaccine side effects, fewer systemic side effects, and less work absenteeism (p < 0.05).

CONCLUSION

In contrast to previous studies of health care workers, framing did not influence patients' decisions, possibly due to the patients' awareness of their higher risk of influenza complications and greater desire to follow recommendations. The common practice of using negative frames when describing probabilities of side effects may need to be reexamined, considering its deleterious influence on self-reported side effects and work absenteeism.

Assessing preferences about the DNR order: does it depend on how you ask?

Despite increasing emphasis on advance directives, there has been little methodologic work to assess preferences about the "do not resuscitate" (DNR) order. This developmental work assessed, in a non-patient group, the performance of a probability-trade-off task designed to assess DNR attitudes, in terms of framing effects and stability of preferences. 105 female nursing students each completed one of two versions of the task. In version I (n = 58), the trade-off moved to increasingly negative descriptions of the outcomes of resuscitation (decreasing chance of survival and increasing risk of brain death), whereas in version II (n = 47), the trade-off moved to increasingly positive descriptions. One week later, repeat assessments were obtained for versions I (n = 35) and II (n = 28). The DNR preference scores were lower and more stable when the task moved to increasingly positive descriptions; perhaps this version of the task tends to weaken risk aversion. These results imply that care should be used in applying a probability trade-off task to the assessment of DNR preferences, since artefactual effects could be induced.

Cancer patients' decision making and trial-entry preferences: the effects of "framing" information about short-term toxicity and long-term survival

The study purpose was to determine whether the framing of treatment information influenced patients' reported preferences for participating in treatment decision making and for trial entry. Ninety cancer patients read either neutrally-, positively-, or negatively-framed information about a chemotherapeutic treatment, then indicated their preferences for participating in the treatment decision, and whether they would participate in a clinical trial incorporating this protocol. There was no difference across information groups in preferences for participating in treatment decision making or willingness to enter such a clinical trial. Preference for participation in treatment decision making was significantly related to age (t = 2.54; p = 0.022), sex (x2 = 3.89; p = 0.05), and education (t = 2.54; p = 0.018); trial entry preferences were unrelated to these demographic variables. These results imply that, in this clinical context, attitudes towards participation in treatment decision making may be associated with characteristics of the patient, and attitudes towards trial entry may be dependent upon the clinical characteristics of a particular trial, but neither set of attitudes is influenced by the framing of protocol information.

Patient preference for high or low dose rate brachytherapy in carcinoma of the cervix

High and low dose rate are two competing methods of brachytherapy. Existing data do not support choosing one method over the other for treating carcinoma of the uterine cervix. Arguments include clinical efficacy, monetary cost, radiation safety, and patient preference. There are no published data on patient preference. We developed a questionnaire to elicit patient preference and to measure its strength. Subjects received descriptions of both treatment options and their probable outcomes. We elicited preference for one low or three high dose rate fractions, and for two low or five high dose rate fractions, assuming both methods to be isoeffective. Strength of initial preference was measured by asking subjects how much of a change, in either the changes for cure or the chances for toxicity, would make them change preference. The questionnaire was completed by female staff at our centre (n = 90), by a group of previously treated patients (n = 18), and by a group of newly diagnosed patients (n = 20). When both methods were assumed to be isoeffective, only 34% of the 38 patients preferred three fractions of high dose rate to one fraction of low dose rate. However, when high dose rate was assumed to be 2% more curative, or 6% less toxic, a simple majority of 50% then said they would prefer high dose rate. Both preference and strength of preference for low dose rate were significantly associated with a greater travelling distance for treatments. Age, marital status, family structure, education, employment, and family income were not associated. In summary, a majority of our patients preferred low dose rate brachytherapy.(ABSTRACT TRUNCATED AT 250 WORDS)

Can there be a more patient-centred approach to determining clinically important effect sizes for randomized treatment trials?

Sample sizes for treatment trials with categorical outcomes are conventionally derived by balancing three elements: a difference between alternative treatments in the event rates for the outcomes of interest (commonly termed the clinically important difference), the alpha error tolerance (false positive risk) and the beta error tolerance (false negative risk). Clinically important differences used to plan trials are chosen in part based on earlier experience with similar interventions (i.e. biological or clinical plausibility). Methodological conventions and clinicians' perceptions will also affect choices. Lastly, practical concerns about the feasibility of accruing large numbers of subjects may drive trialists to specify bigger differences as clinically important, with a view to containing sample size requirements. We suggest that patients or other members of the public be given an active role in determining the magnitude of the clinically important treatment effect for trial planning. Probability trade-offs could be constructed to enable patients and/or healthy volunteers to indicate the degree of benefit they would want from a "new" treatment, given the potential side-effects of the same treatment. This method has the advantage of respecting patient autonomy and principles of informed consent. It provides an additional consideration when plausible effect sizes and error tolerances on hypothesis tests are balanced against feasibility of accruing various sample sizes. Its primary disadvantage is inconvenience, as it adds another step to trial design. On the other hand, if patient-based clinically important differences are generated for a variety of disease states and types of treatments, specific trade-off exercises may be needed only for unusual trials.(ABSTRACT TRUNCATED AT 250 WORDS)

The effect of physician's explanations on patients' treatment preferences: five-year survival data

OBJECTIVE

To evaluate the influence of physicians' explanations on patients' choices.

SETTING

A university-based Department of Veterans Affairs Medical Center.

PARTICIPANTS

136 patients seen in a continuity-care general medicine clinic.

MEASUREMENTS AND RESULTS

Patients were randomized to two groups [Limited Explanation (LE) and Extensive Explanation (EE)] and asked to choose between two alternative treatments (differing in short-term vs long-term survival benefits) for an unidentified medical condition, based on the information given in the explanations. LE consisted of a brief orientation to graphs summarizing the treatment results, while EE consisted of a detailed verbal description of the graphs. Significantly (p < 0.001) more patients receiving EE changed their preferences across the three pairs of five-year survival curves, compared with patients receiving LE. Of the patients receiving EE, 57% reported either medium-term (year 0-to-intercept or intercept-to-year 5) data or the average life expectancy for the five-year period contained in the curves (ALE-5) as most influencing their decision making; whereas 78% of patients receiving LE reported only endpoint (year 0 or year 5) data as most influencing their preferences.

CONCLUSIONS

The patients' treatment preferences for long-term vs short-term survival benefits were influenced by the amounts of verbal explanation provided to them about five-year survival graphs summarizing treatment results. The patients appeared to minimize the importance of medium-range data when those data were not specifically pointed out to them.

Opinion polling and decision making: a critical appraisal of quality of life assessment

The relationship between quality of life (QOL) assessments and decision making, in relation to the delivery of health services, is subjected to critical appraisal. Three levels of decision making in the health care system are taken into account in the analysis. Criticisms of opinion polling provide the basis for the appraisal. Examples of criticisms considered are: Might the use of QOL information be manipulative? Could the interviews or questionnaires used to obtain QOL data influence personal opinions? Are the methods used sometimes defective and/or superficial? Will QOL information always be used in decision making in ways that are ascertainable and justifiable? It is concluded that the time has come for the main focus of critical appraisal in QOL research to shift, from an emphasis on evaluation of the quality of methods used for assessments of QOL, toward an emphasis on the practical usefulness of QOL data.

Test-retest reliabilities of two treatment-preference instruments in measuring utilities

The authors assessed the test-retest reliabilities of two treatment-preference instruments recently applied to the measurement of the utilities of health states after different treatment modalities for cancer. The first instrument measures the strengths of preferences concerning a choice between a wait-and-see policy, and treatment with radiotherapy after an initial surgical breast-conserving procedure for early breast cancer. The second measures the strengths of preferences concerning a choice between two hypothetical surgical treatment outcomes in cancer of the rectum with different probabilities of expected five-year survival. Both measure the strength of a subject's treatment preference given probabilities of treatment-related costs and benefits. The subjects were radiotherapy technicians (n = 20) and cancer patients (n = 20) who were interviewed in weeks 2 and 4 of radiotherapy. The test-retest reliabilities of both instruments were inconsistent and moderately high, with Spearman's rank correlations ranging from 0.38 to 0.81 and weighted kappas ranging from 0.38 to 0.69. To investigate whether the start of treatment with radiotherapy influenced the utilities that patients assigned to health states, the same procedure was applied in another, comparable, group of patients with cancer (n = 20). For this group, the first assessment was made prior to the start of treatment and the second during the second week of radiation therapy. The scores of this group of patients indeed appeared to be less stable than the scores of the patients assessed in weeks 2 and 4 of radiotherapy. However, the instability of the scores could have been the result of test bias.

Decision analysis in oncology: panacea or chimera?

A review of the literature on the use of decision analysis in clinical oncology shows that, although these techniques have been available for more than 25 years, they have not been widely applied: only 19 decision analyses of therapeutic management in clinical oncology were found. The main disadvantages concern the difficulty of accurately assessing probabilities and defining measures of outcome. Time-consuming analysis may produce results that are either equivocal or simply confirm the expectations of common sense. If the basic design fails to include all relevant factors then any conclusions will be of little value. The main advantages are that, by demanding that problems be explicitly stated and analysed in a logical fashion, deficiencies in current knowledge, belief and practice are identified. The usefulness of these techniques lies in formulating management guidelines, either for treatment or for follow-up. They have only a limited role in decision making for individual patients.

Interpretation of graphic data by patients in a general medicine clinic

OBJECTIVE

To assess how patients use graphic data to decide on preferences between alternative treatments.

DESIGN

Cross-sectional survey of patients, physicians, and medical students. The physicians and medical students served as a control group with which to compare the patients' responses.

SETTING

A university-based Department of Veterans Affairs Medical Center.

PARTICIPANTS

152 patients seen in a general medicine clinic, 57 medical students, and 11 physicians.

MEASUREMENTS AND RESULTS

Subjects were given a survival graph showing the patient outcomes for two different unidentified treatments for an unidentified serious disease. They were asked to indicate which treatment they preferred and which portion(s) of the curves most influenced their preference. A large majority of both patients and health professionals preferred the treatment that had worse short-term and better long-term survival. Eleven percent of patients and 51% of health professionals identified mid-curve data (points other than the curve end-points) as most influencing their preferences.

CONCLUSIONS

A graphic survival curve appears to provide enough information to assess patient preferences between two alternative treatments. Patients appeared to differ from physicians and medical students in their interpretation of the curves.