Two ways to skin a cat: a comparison of two variants of standard gamble John Brazier and Paul Dolan Prepared for the CHEBS workshop on Elicitation, 9 October.

Slides:



Advertisements
Similar presentations
Elicitation methods Health care demands exceed resource supply Therefore, rationing is inevitable Many ways by which we can ration health care One is economic.
Advertisements

Developing a Questionnaire
Hypothesis Testing. To define a statistical Test we 1.Choose a statistic (called the test statistic) 2.Divide the range of possible values for the test.
Validity (cont.)/Control RMS – October 7. Validity Experimental validity – the soundness of the experimental design – Not the same as measurement validity.
The impact of Genital Herpes on Health Status Preferences Raj Patel University of Southampton.
Scaling Session Measurement implies “assigning numbers to objects or events…” Distinguish two levels: we can assign numbers to the response levels for.
Research methods – Deductive / quantitative
Using a discrete choice experiment with duration to estimate values for health states on the QALY scale Nick Bansback Assistant Professor School of Population.
Judgment and Decision Making How Rational Are We?.
Estimation Procedures Point Estimation Confidence Interval Estimation.
1 Modelling valuations for the EQ-5D health states: an alternative model using differences in valuations Jennifer Roberts and Paul Dolan Sheffield Health.
Valuing the SF-6D: a nonparametric approach using individual level preference data Part 1): The SF-6D and its valuation Samer A Kharroubi, Tony O’Hagan,
COST–EFFECTIVENESS ANALYSIS AND COST-UTILITY ANALYSIS
Modelling Cardinal Utilities from Ordinal Utility data: An exploratory analysis Peter Gilks, Chris McCabe, John Brazier, Aki Tsuchiya, Josh Solomon.
The Analysis of Variance
Schneider Institute for Health Policy Heller Graduate School Brandeis University September by Donald S. Shepard, Ph.D. Schneider Institute for Health.
Utilising rank and DCE data to value health status on the ‘QALY’ scale using conventional and Bayesian methods John Brazier and Theresa Cain with Aki Tsuchiya.
25 Sept 07 FF8 - Discrete Choice Data Introduction Tony O’Hagan.
Survey Research Slides Prepared by Alison L. O’Malley Passer Chapter 7.
Learning Strategies and Low- Literacy Adult Hmong Students Julia Reimer LESLLA Conference 2009.
© 2008 McGraw-Hill Higher Education The Statistical Imagination Chapter 9. Hypothesis Testing I: The Six Steps of Statistical Inference.
CHAPTER 21: Comparing Two Proportions
Why use the EQ-5D? What are the alternatives?. What are the alternatives for Direct valuation? Other VAS Time Trade-Off Standard Gamble Willingness to.
Comparing Two Population Means
Economic evaluation of health programmes Department of Epidemiology, Biostatistics and Occupational Health Class no. 9: Cost-utility analysis – Part 2.
BIS 360 – Lecture Five Ch. 7: Determining System Requirements.
317_L26, Mar J. Schaafsma 1 Review of the Last Lecture Are looking at program evaluation in healthcare Three methods: CBA, CEA, CUA discussed CBA,
Type author names here Social Research Methods Chapter 10: Self-completion questionnaires Alan Bryman Slides authored by Tom Owens.
Sampling is the other method of getting data, along with experimentation. It involves looking at a sample from a population with the hope of making inferences.
Qualitative Research 2 Dr Shona Bettany.
Validity RMS – May 28, Measurement Reliability The extent to which a measurement gives results that are consistent.
1 EQ-5D, HUI and SF-36 Of the shelf instruments…..
Overview of Health-Related Quality of Life Measures May 22, 2014 (1:00 – 2:00 PDT) Kaiser Methods Webinar Series 1 Ron D.Hays, Ph.D.
Section 5.1 Designing Samples AP Statistics
Introductory Statistics Options, Spring 2008 Stat 100: MWF, 11:00 Science Center C. Stat 100: MWF, 11:00 Science Center C. –General intro to statistical.
By: Dr. AWATIF ALAM ASSOCIATE PROFESSOR MEDICAL COLLEGE,KSU.
Sample Size Determination in Studies Where Health State Utility Assessments Are Compared Across Groups & Time Barbara H Hanusa 1,2 Christopher R H Hanusa.
AP STATISTICS LESSON AP STATISTICS LESSON DESIGNING DATA.
Chapter 10: Comparing Two Populations or Groups
11/18/2015 IENG 486 Statistical Quality & Process Control 1 IENG Lecture 07 Comparison of Location (Means)
Buying and Selling Prices under Risk, Ambiguity and Conflict Michael Smithson The Australian National University Paul D. Campbell Australian Bureau of.
Using a Discrete Choice Experiment to Value the EQ-5D-5L in Canada Nick Bansback Assistant Professor School of Population and Public Health, University.
© 2006 The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill Sampling Chapter Six.
SOCIAL SCIENCE INQUIRY MODEL
Experimental Research Methods in Language Learning Chapter 5 Validity in Experimental Research.
Scaling Session Measurement implies assigning numbers to objects or events. In our case, the numbers “weight” responses to questions, so that saying “Yes”
Producing Data Lab #3 Reading correlation table: Bottom of p. 52 and top of page 53 of Sorenson Reading regression output to construct your equation: Sorenson.
Chapter 13: Inferences about Comparing Two Populations Lecture 8b Date: 15 th November 2015 Instructor: Naveen Abedin.
Bangor Transfer Abroad Programme Marketing Research SAMPLING (Zikmund, Chapter 12)
Heidi L. Dempsey, David W. Dempsey, Tomesha Manora, Amanda Webster, Jody Thompson, Aaron Garrett, Iyanna Cammack, Yawa Dossou, Angel Johnston, & Michael.
Item Analysis: Classical and Beyond SCROLLA Symposium Measurement Theory and Item Analysis Heriot Watt University 12th February 2003.
© 2012 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or in part.
Remaining Challenges in Assessing Non-Inferiority Steven Snapinn DIA Statistics Community Virtual Journal Club December 16, 2014 Based on Paper with Qi.
Testing for a difference
Instrumentation.
Unit 4 – Inference from Data: Principles
Sampling and Experimentation
Social Research Methods
Preference Assessment 1 Measuring Utilities Directly
Understanding Results
SAMPLING (Zikmund, Chapter 12.
Creating an Effective Test or Survey Instrument LaShawnda Purdie EDU:652 Instructional Design and Delivery Instructor: Dr. Judith Marged 04/02/2018.
Health-Related Quality of Life Assessment in Outcome Studies
Stat 217 – Day 28 Review Stat 217.
Volume 66, Issue 1, Pages (July 2004)
Using statistics to evaluate your test Gerard Seinhorst
Variations on Aschs Research
SAMPLING (Zikmund, Chapter 12).
Social Research Methods
Elicitation methods Health care demands exceed resource supply
Presentation transcript:

Two ways to skin a cat: a comparison of two variants of standard gamble John Brazier and Paul Dolan Prepared for the CHEBS workshop on Elicitation, 9 October 2002

Standard Gamble:Chronic Health States full health dead state Hi p 1-p where full health  Hi  death vary ‘p’ until the respondent is indifferent between the alternatives

Variants of SG The ‘ping pong’ variant (PPV): interview administered with the aid of props using a ‘ping pong’ procedure The ‘titration’ variant (TV): self-completed; presents a list of probability of success values (from 100 in 100 down to 0 in 100) for ticking No theoretical arguments for choosing between them

Please put a Ö against all cases where you are CONFIDENT that you would CHOOSE the risky treatment (Choice B). Please put an X against all cases where you are CONFIDENT that you would REJECT the treatment (Choice B) and accept the certain health state (Choice A). Please put an = against all cases where you think it would be most difficult to choose between the treatment (Choice B) and accept the certain health state (Choice A). Chances of successChances of failure 100 in 100*0 in 100* 95 in 100*5 in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in in 100

Study questions Do the two variants produce the same values? If they produce different values, then why?

Design Convenience sample of staff and students at UoS Randomised into two groups: G1 and G2 Both groups completed the SF-6D descriptive system and then ranked and valued seven states defined by the SF-6D 1.G1 valued the first 3 states plus PITS by PPV and the remaining 3 states by TV 2.G2 valued the first 3 states plus PITS by TV and the remaining 3 states by PPV

Example of (SF-6D) health states Your health limits you a lot in moderate activities (such as moving a table pushing a vacuum cleaner, bowling or playing golf) You are limited in the kind of work or other activities as a result of your physical health Your health limits your social activities (like visiting friends, relatives etc.) most of the time You have pain that interferes with your normal work (both outside the home and housework) moderately You feel tense or downhearted and low most of the time. You have a lot of energy a little of the time

Design (2) The six states paired where each pair differed only in one level of one dimension For six states, full health and PITS (i.e. the worst state defined by the SF-6D) were the reference states and PITs was then valued against full health and death Both groups valued states in the same order At the end of the interview, respondents asked about difficulty and understanding and the reasons for any apparent inconsistencies

Results (1) G1 (n=28) and G2 (n=30) comparable in terms of gender, age and own health Rankings similar and VAS ratings of states virtually identical Mean values generated by TV higher than PPV for first four states (by.107, 0.38,.13 and.138), but little or no difference for the second set of states TV had a higher proportion of 1.0 responses than PPV (25 vs. 6)

Mean SG health state values

Results (2) Logical ordering (Z>X, Q>Y, T>R) VAS values consistent Inconsistency between variant: G1 Z(PPV) vs. X (TV) and G2 Q(TV) vs. Y(PPV) Inconsistency within variant: PPV - Z and X and TV – Q and Y

Respondents comments Inconsistencies due to confusion/forgetting; some found the PPV made gambling more attractive (‘yo’s yo’s’ around) Understanding and degree of difficulty favoured PPV Found PPV easier and liked the visual aid A small number commented that they preferred TV since PPV was confusing

Discussion (1) TV produced higher values than PPV This Difference only existed for the 4 states valued first by each group – this represents a ‘pure’ variant effect Second set of states also influenced by what went before – possible reference point effect whereby TV valuations in G1 dragged down by respondent memory of PPV valuation of first 4 states (and vice versa) – which offset variant effect

Explanations for results Anchoring effect: TV starts respondents at upper end whereas PPV iterates respondents between upper and lower end Confusion effect: the ping pong of PPV confuses respondents e.g. being iterated from 100% chance of success to 10% for mild states Heuristics: respondents may be using cognitive strategies or shortcuts that operate to simplify cognitive functioning requirements of these tasks

Implications No basis for choosing one variant over the other More research into the way people answer these questions and what influences their response Re-design tasks by combining different procedures, props and methods of administration (e.g. TV with props) Supports the idea that preferences are not merely articulated in these exercises, but are being constructed