In this overview of preference-based measures, utilities and willingness to pay (WTP) are discussed as measures relevant to dermatology for capturing the burden of skin diseases. An overview is provided of the concepts of utilities and WTP and their importance in decision making. Specific examples of elicitation methods for capturing utility and WTP measures are provided. Prior studies exploring utilities and WTP in dermatology are reviewed. Each of these measures has limitations and likely varying relevance to specific dermatologic diseases and to specific individuals.
Utilities: definitions and methods
Utilities are preference-based measures that aim to assess the desirability of certain health states. Such measures are important for decision analysis and cost-effectiveness analysis for which an assessment of the desirability of certain outcomes is crucial. Preference-based measures add the dimension of desirability to a simple assessment of a health status that is crucial for decision making. Well-known methods of assessing utilities empirically include the standard gamble method and the time trade-off method. There are also population-based methods to estimate utilities. No matter the method, the utility is one number; a utility of 1 indicates the health state is equivalent to perfect health, and a utility of 0 indicates a preference for death over the current health state.
Standard Gamble Method
The standard gamble method is used to assess the risk of death (without suffering) that individuals are willing to accept to be restored to perfect health or to rid them of a given disease. The respondent is asked to compare 2 outcomes: one is the best health state under consideration, and the other is the worst health state under consideration (which would be death without suffering). The outcomes are varied in a ping-pong fashion until the decision maker is indifferent between the gamble and the intermediate health state. For instance, if a subject is willing to take a 10% risk of death in exchange for a 90% chance of perfect health, then the utility of the intermediate state is 90% or 0.9.
Time Trade-Off Method
The time trade-off method is an assessment of a person’s willingness to live a shorter but healthier life without the current health condition. Patients are asked to select between 2 choices: living their normal life expectancy with their current skin disease or giving up a defined amount of time from their life in exchange for living disease free. Patients are asked about varying lengths of time they would be willing to give up, by varying from one extreme to the other in a ping-pong fashion. The upper and lower bounds of time traded are then altered gradually, lessening the extremes. The task concludes when subjects indicate that they are indifferent between the 2 choices. The utility is calculated as the ratio of the time in best health to the life expectancy with skin disease. Lower scores indicate increased disease burden. Some believe this method is more comprehensible and simpler than the standard gamble method, especially for health states in which very small risk (or small amounts of time) are to be considered. If the subject is willing to give up one-tenth of his or her life expectancy in return for perfect health, then the utility of the intermediate health state is 1 minus 0.10 or 0.90.
Initially, the standard gamble technique was considered the best method of measuring utilities. This is because it is based directly on the fundamental components of utility theory. However, the time trade-off method is now accepted as the better option because it is more feasible, has a higher discriminating power, and has better face validity.
Utility measurements have many challenges. Investigators need to make sure that the questions are explained clearly and patients are fully informed and comfortable and are in an environment conducive to answering such questions. Because the questions involve theoretical trading of time or taking hypothetical risks, patients may be unfamiliar and may feel that the process is awkward. The iterative process of eliciting these utilities also makes for a very time-consuming endeavor.
Population-based utility assessments
In contrast to the direct utility assessment methods described, utilities can also be assessed indirectly, through population-based measures, whereby health state classification systems are used to assign a utility value to a patient who meets criteria defined by a reference group. The reference group is created based on calculating mean responses from the general public rather than from subjects with the disease. These methods have been embraced because it is less time consuming than empirically assessing utilities by the standard gamble or time trade-off approaches. Population-based utility estimations have been used mostly in the nondermatology literature, but dermatology studies have started to use them. The most popular population-based estimate used in dermatology is the European Quality of Life (EuroQoL) survey (EQ-5D).
The EQ-5D is a paper questionnaire in which subjects classify their health state according to a generic comprehensive health status classification system based on 5 dimensions, including mobility, self-care, usual activity (work, study, housework, family, and leisure), pain/discomfort, and anxiety/depression. Each of the dimensions has 3 levels: no, some, and extreme problems. For each respondent, a profile of health status is formed, and a corresponding health status valuation is matched to a corresponding utility derived from a reference population composed of members of the general public, not subjects with the disease in question. Weighted valuations for each health state have been elicited from respondents in the general population for 8 different countries, including the United States.
Despite the relative ease of using the EQ-5D and other similar population-based utility estimates, very few studies have been performed to validate these methods in the population with dermatologic concerns. The EQ-5D has been found to have limitations when compared with utilities elicited empirically from patients. A study by Lidgren and colleagues assessed the impact of quality of life (QoL) of breast cancer by administering both the EQ-5D and a directly elicited time trade-off question to women with different stages of breast cancer. Across all stages, directly elicited time trade-off values were greater than EQ-5D values. Similarly, a study by Zethraeus and Johannesson found that patient time trade-off values were similar to social UK EQ-5D index values for mild health states but that patient time trade-off values were higher for more severe health states.
These studies suggest that population-based measures may have the capacity to assess health states mild in severity but not more severe health states. Because of this limitation, generic QoL instruments such as the EQ-5D may not accurately capture disease or condition-specific impact. In the context of cost-effectiveness analyses, using the EQ-5D may overestimate QoL impact, translating to overestimation of the cost-effectiveness of a new drug.
A pilot study was performed involving a population with pruritus to examine the potential applicability of population-based utility measures in dermatology. In this study, 58 subjects with chronic pruritus were recruited from the Emory Dermatology clinic. These subjects were interviewed via the time trade-off approach for empirically derived utilities and also given 3 paper-based questionnaires: a demographic survey, the EuroQoL, and a paper time trade-off survey. The paper-based survey consisted of 3 questions in which participants are asked to make hypothetical choices about improvement in their pruritic health state and duration of life. The 3 questions pose 3 different scenarios: (1) 100% relief from current itch, (2) 50% relief from current itch, and (3) never having itch. Rather than an iterative approach with the time trade-off method in the face-to-face interview, the subject is given 4 choices to each question: (1) give up 5 years of life; (2) give up 10 years of life; (3) give up more than 10 years, with a blank for the subject to fill in the number of years; and (4) none of the above, with a blank for the subject to complete. The last option allows for a choice less than 5 years.
The mean (standard deviation [SD]) age of the subjects was 56 (17) years; 38% were men and 64% were white. Of the subjects, 16 (28%) reported experiencing mild pruritus, with 28 (48%) and 13 (22%) experiencing moderate and severe pruritus, respectively. Paired t tests of the data revealed that mean (SD) EuroQoL utility scores 0.80 (0.13) were significantly lower (overestimating burden of disease) than face-to-face time trade-off derived scores (0.87 [0.27], P = .03). When utility scores were stratified by severity, a significant mean difference in valuation was observed for the moderate health state (0.14 [0.17], P <.01) between the 2 methods. Paper time trade-off utilities were overall greater than those derived by the face-to-face time trade-off method (mean [SD] difference −0.04 [0.17], P = .11).
These data indicate that population-based measures for eliciting utilities such as the EuroQoL may demonstrate a ceiling effect. The EQ-5D only includes the most basic measures of health status and does not account for emotional impact such as frustration, anger, helplessness, and stigmata—issues that significantly contribute to the overall QoL impact of cutaneous diseases. Inclusion of these factors may assist in raising the ceiling effects observed. However, the paper version of the time trade-off did not fare better and seemed to underestimate QoL burden of chronic pruritus. These findings provide preliminary evidence that current proxy measures for utilities in chronic pruritus may not be accurate. Further studies should investigate the potential role of other population-based utility measures as proxy measures for eliciting utilities.
Willingness to pay: definitions and methods
Willingness to pay (WTP) is also considered a preference-based health status measure, but, rather than trading time or taking risks, QoL is quantified in monetary terms. Respondents consider how much they would be willing to pay either as a lump sum or as small payments over time to get rid of an undesirable health state. The expectation is that a health state that is more bothersome to the patient will involve a greater burden of disease, and thus a patient will be able to be willing to make relatively more financial sacrifices to get rid of the disease. Information on annual income is also gathered so that WTP amounts can be expressed as percentages of income to help standardize these measures across income levels.
There have been some WTP studies that have varied from this method. Common variations include asking the payment amounts as predefined payment scale categories rather than in an open-ended format.
A variation from a prior study is inserted here as an example to illustrate the ways in which this method may slightly differ but may still capture this preference-based measure: “You are offered a one-time treatment for your nail fungus. The treatment has an 85% cure rate, almost no side effects, and consists of taking a pill for 3 months. Think about all the things for which you spend money—food, rent/mortgage, bills, etc. How much would you be willing to pay out-of-pocket for this?” In this study, patients selected their responses from predefined payment scale categories. Patients were also asked about their annual income, and they responded in predefined income range categories.
Willingness to pay: definitions and methods
Willingness to pay (WTP) is also considered a preference-based health status measure, but, rather than trading time or taking risks, QoL is quantified in monetary terms. Respondents consider how much they would be willing to pay either as a lump sum or as small payments over time to get rid of an undesirable health state. The expectation is that a health state that is more bothersome to the patient will involve a greater burden of disease, and thus a patient will be able to be willing to make relatively more financial sacrifices to get rid of the disease. Information on annual income is also gathered so that WTP amounts can be expressed as percentages of income to help standardize these measures across income levels.
There have been some WTP studies that have varied from this method. Common variations include asking the payment amounts as predefined payment scale categories rather than in an open-ended format.
A variation from a prior study is inserted here as an example to illustrate the ways in which this method may slightly differ but may still capture this preference-based measure: “You are offered a one-time treatment for your nail fungus. The treatment has an 85% cure rate, almost no side effects, and consists of taking a pill for 3 months. Think about all the things for which you spend money—food, rent/mortgage, bills, etc. How much would you be willing to pay out-of-pocket for this?” In this study, patients selected their responses from predefined payment scale categories. Patients were also asked about their annual income, and they responded in predefined income range categories.
Related posts:
Health-Related Quality of Life in Dermatology: Introduction and Overview
Overview of Health Status Quality-of-Life Measures
Melasma Quality of Life Measures
Quality-of-Life Instruments: Evaluation of the Impact of Psoriasis on Patients
Clinical Meaning in Skin-specific Quality of Life Instruments: A Comparison of the Dermatology Life Quality Index and Skindex Banding Systems
Future Directions in Dermatology Quality of Life Measures
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
