Lidia Engel1,2, Nick Bansback2,3,4, Stirling Bryan2,3, Mary M Doyle-Waters2, David G T Whitehurst1,2. 1. Faculty of Health Sciences, Simon Fraser University, Burnaby, British Columbia, Canada (LE, DGTW) 2. Centre for Clinical Epidemiology and Evaluation, Vancouver Coastal Health Research Institute, Vancouver, British Columbia, Canada (LE, NB, SB, MMDW, DGTW) 3. School of Population and Public Health, The University of British Columbia, Vancouver, British Columbia, Canada (NB, SB) 4. Centre for Health Evaluation and Outcome Sciences, St. Paul's Hospital, Vancouver, British Columbia, Canada (NB)
Abstract
BACKGROUND: Health state valuation data are often excluded from studies that aim to provide a nationally representative set of values for preference-based health-related quality of life (HRQoL) instruments. The purpose was to provide a systematic examination of exclusion criteria used in the derivation of societal scoring algorithms for preference-based HRQoL instruments. METHODS: Data sources included MEDLINE, official instrument websites, and publication reference lists. Analyses that used data from national valuation studies and reported a scoring algorithm for a generic preference-based HRQoL instrument were included. Data extraction included exclusion criteria and associated justifications, exclusion rates, the characteristics of excluded respondents, and analyses that explored consequential implications of exclusion criteria on the respective national tariff. RESULTS: Seventy-six analyses (from 70 papers) met the inclusion criteria. In addition to being excluded for logical inconsistencies, respondents were often excluded if they valued fewer than 3 health states or if they gave the same value to all health states. Numerous other exclusion criteria were identified, with varying degrees of justification, often based on an assumption that respondents did not understand the task or as a consequence of the chosen statistical modeling techniques. Rates of exclusion ranged from 0% to 65%, with excluded respondents more likely to be older, less educated, and less healthy. Limitations included that the database search was confined to MEDLINE; study selection focused on national valuation studies that used standard gamble, time tradeoff, and/or visual analog scale techniques; and only English-language studies were included. CONCLUSION: Exclusion criteria used in national valuation studies vary considerably. Further consideration is necessary in this important and influential area of research, from the design stage to the reporting of results.
BACKGROUND: Health state valuation data are often excluded from studies that aim to provide a nationally representative set of values for preference-based health-related quality of life (HRQoL) instruments. The purpose was to provide a systematic examination of exclusion criteria used in the derivation of societal scoring algorithms for preference-based HRQoL instruments. METHODS: Data sources included MEDLINE, official instrument websites, and publication reference lists. Analyses that used data from national valuation studies and reported a scoring algorithm for a generic preference-based HRQoL instrument were included. Data extraction included exclusion criteria and associated justifications, exclusion rates, the characteristics of excluded respondents, and analyses that explored consequential implications of exclusion criteria on the respective national tariff. RESULTS: Seventy-six analyses (from 70 papers) met the inclusion criteria. In addition to being excluded for logical inconsistencies, respondents were often excluded if they valued fewer than 3 health states or if they gave the same value to all health states. Numerous other exclusion criteria were identified, with varying degrees of justification, often based on an assumption that respondents did not understand the task or as a consequence of the chosen statistical modeling techniques. Rates of exclusion ranged from 0% to 65%, with excluded respondents more likely to be older, less educated, and less healthy. Limitations included that the database search was confined to MEDLINE; study selection focused on national valuation studies that used standard gamble, time tradeoff, and/or visual analog scale techniques; and only English-language studies were included. CONCLUSION: Exclusion criteria used in national valuation studies vary considerably. Further consideration is necessary in this important and influential area of research, from the design stage to the reporting of results.
Authors: Lidia Engel; Duncan Mortimer; Stirling Bryan; Scott A Lear; David G T Whitehurst Journal: Pharmacoeconomics Date: 2017-07 Impact factor: 4.981
Authors: Barry Dewitt; Baruch Fischhoff; Alexander L Davis; Stephen B Broomell; Mark S Roberts; Janel Hanmer Journal: Med Decis Making Date: 2019-08-28 Impact factor: 2.583
Authors: Pedro L Ferreira; Patrícia Antunes; Lara N Ferreira; Luís N Pereira; Juan M Ramos-Goñi Journal: Qual Life Res Date: 2019-06-14 Impact factor: 4.147
Authors: Barry Dewitt; Baruch Fischhoff; Alexander L Davis; Stephen B Broomell; Mark S Roberts; Janel Hanmer Journal: Med Decis Making Date: 2019-08-28 Impact factor: 2.583