Help patients make better decisions

Using principles of behavioural economics, the app makes it easier for patients to make good choices even when information is complex.

Rationale: Behavioural economic research has identified a host of circumstances in which people fail to make good choices when presented with the appropriate information. The use of traditional decision aids (DAs) is based on conventional economic theory that assumes once patients are informed, they will make rational judgments based on reflection, deliberation and maximizing behaviour, driven by their values and intentions.1 However, behavioural economic research acknowledges that, in reality, many individuals fail to act rationally, in particular when information is unfamiliar, complex and overwhelming2,3—all traits of the information in most DAs. Instead, people switch to an automatic, affective system, abandoning logic, rationality, and information, and relying instead on heuristics and intuition to guide their decision.2,4 While for some simple decisions, these strategies can be effective because they economize on time and can often lead to sensible decisions. However, departures from rationality can also lead to decision errors – errors that lead to poor quality decisions. In the context of DA, this means some patients fail to identify the option that is best for them.2,4

The errors caused by cognitive biases and simplifying heuristics are well established in the medical literature, but have had relatively little attention in the design and evaluation of DA. Our recent review concluded that the key features of DAs that may lead to inappropriate choices could be attributed to the three decision errors: the framing heuristic; the psychological phenomenon of overweighting rare and unfamiliar events; and uncertainty aversion.

  • Framing heuristic: For a rational decision maker, the order that information presented should not have an influence on the choice that is made. However, people tend to remember information at the beginning (primacy effect) or the end (recency effect).5-7 For example, a study found that patients who learned first about the risks of a treatment thought more favorably of the drug than women who learned first about the benefits. Other studies have found when there is a long list of information, what is listed first is generally considered first and consequently given more weight than information given further down the list.8 Nevertheless, either the harms or the benefits must be presented first in a DA, and therefore, the designer of the DA may inadvertently influence the patient to choose an option.
  • Overweighting of rare or unfamiliar events: Most treatments have multiple complications. DAs are expected to inform patients about any treatment complication that is reasonably likely to occur. Although there is no absolute cut-off for how probable a complication must be in order to be included for meeting informed consent, most DAs show information on any moderately severe complication that occurs at least 1% of the time, and on serious complications that occur even less often. Kahnemann and Tversky’s prospect theory -the origins of behavioural economics for which they were awarded the Nobel prize in Economics – describes how people systematically overweight small probabilities in terms of their impact on decisions.10 Further, the consequence of this phenomenon is that patients using DA can be dissuaded from a treatment with multiple rare side-effects when rationally it would have been the best treatment option.9,10
  • Uncertainty aversion: When individuals receive conflicting, incomplete, uncertain or excessive information, they experience ambivalence and make contradictory decisions.11 The role of information overload causing ambiguity in investment decision-making has been well documented. When the complexity of decision-making increases, people tend to expend less effort to actually make their decision, rather seeking others to make decisions for them or select default options, if available.12

Development DCIDA: The DCIDA seeks to employ principles of behavioural economics to reduce the cognitive effort required to engage complex decision-making processes. This is achieved by structuring the information presented in the interface so that it allows individuals to more easily engage with their systematic processing that ordinarily would require a great deal of mental effort.13,14 In doing so, individuals will be less susceptible to errors caused by the cognitive biases and heuristics described above.15 DCIDA is based on the weighted additive model that underlies much research in economics and decision-making. For treatment decisions, this assumes that the preferred option is based on the sum of the importance or weight (on a 0-100 scale) of each feature, say benefit or harm, multiplied by each options score (on a 0-100 scale) for that feature, with a higher weighted score being the better.

Briefly, the app takes the user through a series of pages similar to a conventional DA. However, the unique feature of the DCIDA is that the value clarification task, which is normally at the end of conventional DA is moved to the beginning of the application. This: 1) helps the patient clarify their preferences surrounding each potential consequence and 2) generates the ‘weights’ for each consequence. The detailed information for each consequence is then described in turn. For each, the patient is asked to rate in comparison to no treatment how important the benefit or harm, of each option. This is used to create the score for each consequence. Finally, the summary information for all consequences is displayed. However in contrast to a conventional DA and consistent with economic principles discussed earlier:

  1. Consequences are ordered in accordance with the size of individual’s weights– obtained from the value clarification exercise. This helps individuals overcome the framing effects by focussing on the information that will most influence their decision.
  2. Rows are sized in proportion to the weights of each consequence, with the most important consequences being presented in wide rows and less important consequences in narrow rows. Attention is therefore taken away from rare events for the majority of people who rate these types of events of low importance when considered in value clarification exercise.
  3. For each consequence, the colour of each cell for each option is based on the score, with a lighter colour indicating a more preferred option. It has been shown that simple colouring can help individuals process multiple pieces of information to make informed decisions
  4. Finally, the sum of the weights and scores are used to determine which option would be preferred based on normative weighted additive model. This becomes the default choice, helping overcome ambiguity aversion. Users may of course select alternative options, but having a default option selected can help this process.

References
1. Kahneman, D., Tversky, A. & Foundation, R. S. Choices, values, and frames. (Cambridge Univ Pr: 2000).

2. Tversky, A. & Kahneman, D. Judgment under uncertainty: Heuristics and biases. science 185, 1124 (1974).

3. Chapman, G. B. & Elstein, A. S. Cognitive processes and biases in medical decision making. Decision making in health care: Theory, psychology, and applications 183–210 (2000).

4. Baron, J. Thinking and deciding. (Cambridge University Press: 2008).

5. Scott, A. & Vick, S. Patients, Doctors and Contracts: An Application of Principal‐Agent Theory to the Doctor‐Patient Relationship. Scottish Journal of Political Economy 46, 111–134 (2003).

6. Stewart, J. M., O’Shea, E., Donaldson, C. & Shackley, P. Do ordering effects matter in willingness-to-pay studies of health care? Journal of health economics 21, 585–599 (2002).

7. Carney DR, Banaji MR. First is best. PLoS One. 2012;7(6):e35088.

8. Schkade David, A. & Kleinmuntz Don, N. Information displays and choice processes: Differential effects of organization, form, and sequence. Organizational Behavior and Human Decision Processes 57, 319–337 (1994).

9. Amsterlaw, J., Zikmund-Fisher, B. J., Fagerlin, A. & Ubel, P. A. Can avoidance of complications lead to biased healthcare decisions. Judgment and Decision Making 1, 64–75 (2006).

10. Ubel, P. A. Is information always a good thing? Helping patients make‘ good’ decisions. Medical care 40, V39 (2002).

11. Epstein, L. G. A Definition of Uncertainty Aversion. Review of Economic Studies 66, 579–608 (1999).

12.Thaler, R. H. & Sunstein, C. R. Libertarian Paternalism. American Economic Review 93, 175–179 (2003).

13. Payne, J. W., Bettman, J. R. & Johnson, E. J. The adaptive decision maker. (Cambridge Univ Pr: 1993)

14. Todd, P. & Benbasat, I. Inducing compensatory information processing through decision aids that facilitate effort reduction: An experimental assessment. Journal of Behavioral Decision Making 13, 91–106 (2000).

15. Holmes-Rovner, M. & Wills, C. E. Improving informed consent: insights from behavioral decision research. Medical care 40, V (2002).