Dataset Information

Predicting preference-based utility values using partial proportional odds models.

ABSTRACT: BACKGROUND: The majority of analyses on utility data have used ordinary least square (OLS) regressions to explore potential relationships. The aim of this paper is to explore the benefits of response mapping onto health dimension profiles to generate preference-based utility scores using partial proportional odds models (PPOM). METHODS: Models are estimated using EQ-5D data collected in the Health Survey for England and the predicted utility scores are compared with those obtained using OLS regressions. Explanatory variables include age, acute illness, educational level, general health, deprivation and survey year. The expected EQ-5D scores for the PPOMs are obtained by weighting the predicted probabilities of scoring one, two or three for the five health dimensions by the corresponding preference-weights. RESULTS: The EQ-5D scores obtained using the probabilities from the PPOMs characterise the actual distribution of EQ-5D preference-based utility scores more accurately than those obtained from the linear model. The mean absolute and mean squared errors in the individual predicted values are also reduced for the PPOM models. CONCLUSIONS: The PPOM models characterise the underlying distributions of the EQ-5D data better than models obtained using OLS regressions. Additional research exploring the effect of modelling conditional responses and two part models could potentially improve the results further.

SUBMITTER: Ara R

PROVIDER: S-EPMC4118278 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting preference-based utility values using partial proportional odds models.

Ara Roberta R Kearns Ben B vanHout Ben A BA Brazier John E JE

BMC research notes 20140708

<h4>Background</h4>The majority of analyses on utility data have used ordinary least square (OLS) regressions to explore potential relationships. The aim of this paper is to explore the benefits of response mapping onto health dimension profiles to generate preference-based utility scores using partial proportional odds models (PPOM).<h4>Methods</h4>Models are estimated using EQ-5D data collected in the Health Survey for England and the predicted utility scores are compared with those obtained u ...[more]

PMID: 25000846

Similar Datasets

Project description:BackgroundGiven that treatments for chronic lymphocytic leukaemia (CLL) are palliative rather than curative, evaluating the patient-perceived impacts of therapy is critical. To date, no utility (preference) studies from the general public or patient perspective have been conducted in CLL. The objective of this study was to measure preferences for health states associated with CLL treatment.MethodsThis was a cross-sectional study of 89 members of the general population in the UK (England and Scotland). Using standard gamble, each participant valued four health states describing response status, six describing treatment-related toxicities based on Common Toxicity Criteria, and two describing line of treatment. The health states incorporated standardized descriptions of treatment response (symptoms have "improved," "stabilized," or "gotten worse"), swollen glands, impact on daily activities, fatigue, appetite, and night sweats. Utility estimates ranged from 0.0, reflecting dead, to 1.0, reflecting full health.ResultsComplete response (CR) was the most preferred health state (mean utility, 0.91), followed by partial response (PR), 0.84; no change (NC), 0.78; and progressive disease (PD), 0.68. Among the toxicity states, grade I/II nausea and nausea/vomiting had the smallest utility decrements (both were -0.05), and grade III/IV pneumonia had the greatest decrement (-0.20). The utility decrements obtained for toxicity states can be subtracted from utilities for CR, PR, NC, and PD, as appropriate. The utilities for second- and third-line treatments, which are attempted when symptoms worsen, were 0.71 and 0.65, respectively. No significant differences in utilities were observed by age, sex, or knowledge/experience with leukaemia.ConclusionsThis study reports UK population utilities for a universal set of CLL health states that incorporate intended treatment response and unintended toxicities. These utilities can be applied in future cost-effectiveness analyses of CLL treatment.

Dataset Information

Predicting preference-based utility values using partial proportional odds models.

Publications

Predicting preference-based utility values using partial proportional odds models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets