• Choose a category
• All
• Classic-Fiction

By Root 1026 0

what kind of stroke? Does the stroke cause her arm or leg to be paralyzed or both? Is the stroke on the dominant side of her brain or the nondominant, so that there are different degrees of impairment with regard to speech and memory? Will she be unable to find and articulate the words to express herself? Does she become so forgetful that she can’t remember the hymns she loves to sing each Sunday at church? How much improvement might occur after the stroke, after months of intensive physical or occupational therapy? According to Bernoulli’s formula of “rational” decision making, Susan would assign a utility to each of these multiple outcomes and then multiply that number by the chance the outcome would occur. Even “simple” choices like taking a statin medication can become complex once a person begins to consider how many different outcomes could occur and their level of severity.

Jansen et al. studied unstable preferences: Sylvia J. T. Jansen et al., “Unstable preferences: A shift valuation or an effect of the elicitation procedure?” Medical Decision Making 20 (2000), pp. 62–71. Fifty-five breast cancer patients were evaluated before, during, and after post-op therapy. They were initially presented with a hypothetical scenario regarding radiation treatment or chemotherapy. There was a disconnect between the experienced health state and the prediction of life from the hypothetical scenario. The researchers concluded that “utilities,” meaning the value that people place on different health states, are not stable but change with experience.

A similar finding, widely cited in the field of research on forecasting, was made by Peter Ubel et al., “Misimaging the unimaginable: The disability paradox in health care decision making,” Health Psychology 24 (2005), pp. S57–S62. The “disability paradox” is the observation that people mispredict the impact that circumstances will have on their well-being and quality of life. Across a wide range of health conditions, patients typically report greater happiness and quality of life than do healthy people imagining the patients’ circumstance.

Among patients with illness, misreports can be due to (1) scale recalibration, meaning that what 90 out of 100 means to one person can be something different to another; (2) conversational context, meaning patients respond differently when they know they are being surveyed as patients (this is certainly true with prostate cancer outcomes); (3) theory-driven recall bias: “I must have been less happy back then, because I think I am getting happier over time” (theory-driven recall bias comes in part from the work of Michael Ross, “Relation of implicit theories to the construction of personal histories,” Psychological Review 96 [1989], pp. 341–357; Ross demonstrated this phenomenon in many other contexts, showing that people typically remember being less happy five years ago than they are currently, despite experiencing stable levels of happiness); (4) global judgments versus momentary moods: Despite experiencing moods rated, for example, at 5, 6, 5, 6, 7, out of 10 over the past day, the patient reports experiencing an average mood of 7 out of 10, since “7” was the most recent mood state.

There are several reasons to believe that moment-to-moment measures are more accurate than global measures. For example, when making global assessments, people place disproportionate weight on their recent moods (Daniel Kahneman et al., “When more pain is preferred to less: Adding a better end,” Psychological Science 4 [1993], pp. 401–405). In other research, one study looked at forty-nine dialysis patients with kidney failure versus forty-nine healthy controls matched by age, race, gender, and education. Researchers gave each subject a PalmPilot program to beep at random intervals of the week and then asked them about their mood on a scale of minus 2 (very unpleasant) to 2 (very pleasant). Both groups reported experiencing positive moods significantly more often than negative moods. Despite the similarity in the moods of the kidney failure patient group versus the