BERKELEY, CA (UroToday.com) - “Utility” is a global measure of health-related quality of life, which ranges from 0 (for death) to 1 (for full health), although values less than 0 are possible for conditions considered to be worse than death.
Utilities can be elicited directly using standard techniques, such as the Rating Scale or Standard Gamble. More commonly, utilities are obtained indirectly, by administering questionnaire-type instruments which have associated community or patient-derived preference weights which are used to calculate utility scores. Examples of these instruments are the EQ-5D and the Health Utilities Index (HUI). Both are generic instruments which may not be responsive to changes in health status in specific diseases.
The Patient Oriented Prostate Utility Scale (PORPUS) is a 10-item health state classification system. Each item describes a prostate cancer-specific quality of life domain: pain, energy, social support, communication with doctor, emotional well-being, urinary frequency, urinary leakage, sexual function, sexual interest, and bowel function. Within each domain, there are four to six levels representing a range of symptom severity from no problems to severe problems. The PORPUS may be used as a profile, disease-specific, non-preference based instrument to compute a quality of life measure that we call the PORPUS-P. The PORPUS can also be used for direct utility elicitation with Rating Scale and Standard Gamble techniques, but it is time-consuming and requires a trained interviewer as well as visual props.
The main purpose of this study was to fit a multi-attribute utility function (MAUF) for the PORPUS using data elicited from patients. The PORPUS could then be used as an indirect utility instrument.
We proposed to derive the utility-based measure for the PORPUS by fitting a multi-attribute utility function (MAUF) to data on patient preferences for health states described by the PORPUS. The MAUF is a function that maps each possible health state to a utility on a scale anchored at zero for dead and one for full health. The approach we chose used a pre-specified function based in multi-attribute utility theory and a step-by-step approach to fitting this function. This approach, termed the “explicitly decomposed” method, was used in the development of the Health Utilities Index Mark II (HUI-2). The MAUF may take one of three main forms: multilinear, multiplicative, or additive. We chose the multiplicative form, also used for the HUI; it allows for interactions between the preferences for different attributes without the large number of parameters required by the full multilinear form (over 1000 with 10 attributes).
We decided to use the preferences of prostate cancer patients, rather than the general public, in the weighting function for the PORPUS. Many side effects of prostate cancer treatment, namely urinary, bowel, and sexual problems, may be valued differently in the context of prostate cancer. Through structured interviews with prostate cancer outpatients, we obtained patients’ rating scale values for the 2 to 4 intermediate levels within each of the 10 single attributes of the PORPUS. We also obtained rating scale values for 14 health states described by the PORPUS. In order to estimate a function mapping rating scale values to standard gamble utilities, we also obtained standard gamble utilities for 4 of these health states.
We tried fitting the data to a total of 15 different models to obtain the best-fitting multi-attribute utility function. We validated this function with data from 2 previous studies in which prostate cancer patients had provided standard gamble utilities for three health state scenarios (mild, moderate, and severe symptoms of prostate cancer described by the PORPUS). In these validation cohorts, the mean standard gamble scores were somewhat lower than the calculated PORPUS utility scores (PORPUS-U) for the mild symptoms health state but the mean standard gamble scores for the severe symptoms state were centered around the PORPUS-U scores for that health state.
The result of our study is an equation that calculates a utility score, the PORPUS-U, on a scale from 0=dead to 1=full health from a patient’s answers on the 10-item PORPUS questionnaire. This is a much easier method to obtain utilities than direct utility assessment using measures such as the standard gamble or Time Trade Off, and is suitable for administration by mail or in a busy clinic.
Written by:
George Tomlinson, Karen E. Bremner, Paul Ritvo, Gary Naglie, and Murray D. Krahn as part of Beyond the Abstract on UroToday.com. This initiative offers a method of publishing for the professional urology community. Authors are given an opportunity to expand on the circumstances, limitations etc... of their research by referencing the published abstract.
UroToday.com Prostate Cancer Section