Econ 6038: Lecture Eight QALY and Cost Utility Analysis

1
Econ 6038 Lecture Eight QALY and Cost Utility
Analysis

• Raymond Yeung
• http//web.hku.hk/rytyeung
• 14 November 2005

2
Contents

• The basic of utility in health economics
• Definition of QALY
• Methods of estimation Standard gamble,
  time-tradeoff etc
• Applications

3
Utility in health economics

• Utility maximization is the fundamental goal in
  modern economics analysis
• Cardinal approach to measuring preference under
  uncertainty, where both the sign and magnitude
  counts, is the basic premises in quantitative
  economics
• Incorporating utility measurement marks a new
  development in conventional CEA where consumer
  preference was not included into consideration
  CEA is only a financial analysis, not economic
  analysis

4
Axioms of von Neumann-Mortgenstern utility theory

• Let gt symbolize preferred to
• Preference exists and are transitive if there
  are three prospect y, y, y with respect to
  three risk profile, people can rank their
  preference among one another and when ygty,
  ygty, then ygty
• Independence If y p1,x1,x2 and
  ypp2,x1,x2, then Yp,y,y p p1(1-p)
  p2, x1, x2
• Continuity If x1gtx2gtx3 in full certainty, there
  exists a level of p where x2 p x1 (1-p) x3

5
Cost utility analysis

• In Cost Effectiveness Analysis (CEA), benefits of
  an intervention, I.e. health outcomes, are
  measured in natural unit e.g. mortality, life
  year saved etc
• In Cost Utility Analysis (CUA), the raw health
  benefits are adjusted to reflect the change in
  quality Quality adjusted life year is used

6
Quality adjusted life year (QALY)

• A single measurement that combines morbidity gain
  (quality) and mortality gain (life year)
• QALY Q life year gained
• Quality weight Q is needed to operationalize the
  QALY concepts
• The weight should be (a) anchored on perfect
  health and dead (b) based on preference (c)
  measured on an interval scale

7
Advantages of using QALY

• Intermediate benefits (e.g. blood pressure
  reduction) can be expressed in the ultimate
  objective of healthcare
• Common measurement unit enables cross-programme
  comparison QALY league table
• Quality of life is an equivalent terms in
  economics the use of QALY provides strong
  economic foundation for medical decision making

8
Application

• A new hypertension control medicine can extend
  the life expectancy of patients aged 60 from 10
  to 15 years, same as the conventional formulae.
  It has less undesirable side effects that
  represent an improvement of quality of life from
  0.4 by a factor of 0.8
• The cost of prescribing this drug is US150 per
  patient, comparing with the conventional drug of
  US100
• Do you recommend Hospital Authority to use this
  drug?

9
How do you measure preference?
Source Drummond et al (1999) Table 6.1
10
Measuring health state preference

• The methodology keeps evolving. Different schools
  (e.g. McMaster University, University of
  Sheffield, San Diego etc) are inventing different
  approaches
• Measurement of health states / measurement of
  health state preference

11
Measuring health state

• In typical public health questionnaire,
  respondents are asked with the following
  question
• We call the result self-perceived health status
• Compare with people in the same age, how do you
  describe your present health condition
• 1) Excellent
• 2) Very good
• 3) Good
• 4) Fair
• 5) Poor

12
Measuring health state preference

• Utility is a preference measurement
• Simply assign the categories into a quantitative
  scale from 0 to 100 does not represent a
  measurement of health state preference
• People choices or will towards different states
  (perfectly healthy or dead inclusive) should be
  accounted for
• The preference should be expressed by the
  individual, not physicians

13
Ratio, rating, VSA

• Respondents are asked to assign the health
  outcomes among the following scale, e.g.

Cannot walk
Hospitalized 7 days
Die
Perfect
100
0
Chronic illness
Rest at home
Blind
14
Standard gamble

• Direct application of VM utility theory
• The goal is to find the equilibrium p

15
Time tradeoff for chronic condition

• Respondents are asked to provide x to match the
  state of chronic condition from present age to
  the life expectancy
• For chronic condition, preference score is
  computed using the formula h x/t

16
Time tradeoff temporary condition

• For temporary condition, state i are compared
  with the perfect state 1.0 and the worst case j
  (e.g. flu episode)
• The utility score is computed using the formula
  h 1 (1-h) x/t

Utility
1.0
Option 2
hi
Option 1
hj
Time
0.0
x
t
17
Other methods

• Person tradeoff How many patients in the
  designated state of health should have their
  lives extended by one year in order to be
  equivalent to extending the lives of 100 healthy
  patient by one year
• Rosser Index (outdated) e.g. people are asked
  whether blind is twice as bad as limb is.

18
Multi-attribute health status classification
systems

• Health status is not simply summarized by one
  single number but a set of numbers
• Each number represents the health status of a
  particular dimension e.g. mobility, emotion,
  cognition, self-care, pain etc.
• Each attribute consists of different levels
• Axioms of VM utility theory were expanded
  (particularly independence) to allow additivity,
  multiplicity of various dimension so that we can
  derive an ultimate summary score

19
Popular multi-attribute systems

• Quality of Well Being (US)
• EuroQol (Europe)
• SF12, SF36 (UK, HK)
• HUI (Canada)

20
Healthy-year equivalents (HYE)

• It is a less popular alternative to QALY
• HYE measures not simply a single point of health
  condition but an entire health profile
• It is estimated using two stage standard gamble

21
Task 8

• This exercise is taken from Drummond et al (1999)
  Box 6.7
• Drummond et al (1999) is the authoritative
  reference for economic evaluation. Another
  authority for CEA specifically is Gold MR, Siegel
  JE, Russell LB and Weinstein MC (1996)
  Cost-effectiveness in health and medicine, Oxford
  University Press