Contingent valuation and public
health: eliciting values from
patients and the public
TO IDEP- GREQAM WORKSHOP
By Cam Donaldson
23RD-24TH JUNE 2003
“Political economy has to take as the measure of utility of an
object the maximum sacrifice which each consumer would
be willing to make in order to acquire the object....the only
real utility is that which people are willing to pay for.”
Jules Dupuit (1844)
What I am not doing today…
• Discrete choice/conjoint analysis
• Willingness to accept
• Conventional validity studies:
– e.g. payment vehicles
Mostly about enhancing the validity of WTP
studies with respect to relevance to
• Why WTP in health?
• Using the values in decision making
– Uptake of WTP studies
– Patients and the public
– Distributional issues
• Progress in eliciting values from patients
• Progress in eliciting values from the
What are we valuing?
inputs Healthy time UTILITY
How resources are
used, e.g. process,
WTP: why we need it
• strength of preference
• can compare different „goods‟
• pulls together several factors
• „beyond health‟
• more in line with theory? Less restrictive on:
– Mode of sacrifice
– Evaluation „space‟
Uptake of WTP
• Do hypothetical questions give hypothetical
– Comparisons to revealed preference
– Still developing (CV/SG)
• Unethical to use in decision making?
• Perceptions and guidelines
• General problem with economics?
Private (1) (2)
good Collectively (3) (4)
USING WTP DATA:
four main scenarios
(1) Patients; private good
If WTP > cost, implement
(2) Public; private good
If WTP > cost, implement
(3) Patients; collective good
(4) Public; collective good
Need WTP and cost data for alternative uses of
WTP AND DISTRIBUTION OF INCOME:
“..the validity of this approach rests upon some important Paretian
assumptions; one being that individuals are the best judges of their
own welfare, another being that the current distribution of income is
[O‟Brien B and Drummond MF (1994) Statistical versus quantitative significance in the
socioeconomic evaluation of medicines. PharmacoEconomics, 5: 389-398.]
“Cost-benefit analysis‟s primary valuation method is willingness to
pay(authors‟ emphasis) (WTP), an approach whose difficulty lies in its
intrinsic favouring of the programs and diseases of the affluent over
those of the poor”
[Gold M et al. (1996) Effectiveness in Health and Medicine. New York, Oxford Univ Press,
Distribution: what to do?
• Universal problem: QALYs have it too
• Leave values unweighted? (Pauly)
• Decision-making view (Currie et al., Little
• Sensitivity tests (Boardman et al.,
– Direction of preference
– Strength of preference within income groups
– Sensitivity of end results to different weights
„WTP for own care‟
• ask patients to value their own treatment
• example: pre-natal CF testing:
– both parents have to test positive
– disclosure versus non-disclosure
– 127 out of 176 in trial responded to WTP
RESULTS FROM CF SCREENING STUDY
(1) About £21!
(2) No difference between groups.
This result has been shown in other studies.
CF follow up: ‘WTP for each’
• make respondents aware of alternatives
• reference point theory
• example: CF follow up
– non-trial sample of 450 pregnant women asked about
WTP for each method
RESULTS FROM CF FOLLOW UP
(1) Lower response rate.
(2) 61% preferred disclosure; 27% preferred non-
(3) Problems with validity.
PREFERENCES AND WTP
Is WTPD > or < WTPND?
Pref Less Equal Greater Total
Disc 61 (17%) 99 (28%) 62 (17%) 222 (62%)
Non-disc 78 (22%) 16 (4%) 94 (26%)
None 21 (6%) 22 (6%) 1 44 (12%)
N.B. 90 people did not respond to this question.
Give the respondent details of each alternative (i.e. “existing” and
“alternative” care) and ask them:
- which they prefer;
- what is the maximum amount they would be willing to pay to
have their preferred rather than their less-preferred option.
This has been done in studies of:
- management of miscarriage
- intra-partum care (postal; open-ended)
- parents‟ views of services for children (interview; bidding).
STUDY OF CHILD HEALTH SERVICES
(1) 82 parents interviewed (out of 300 asked
(2) Each parent was asked one WTP question
about each of three pairwise choices:
• inpatient stay versus day case for tonsillectomy
• hospital-based versus local clinics for bedwetting
• school health service to focus on all children or those with
RESULTS FROM STUDY OF
CHILD HEALTH SERVICES
Day case I/p stay
Mean £56 £43
(95% CI) (£28, £84) (£22, £63)
Median £45 £27.50
(25th,75th %iles) (£0.125, £100) (£0, £50)
Total WTP £1570 £1119
WTP AND DISTRIBUTION
Data on preferences allow us to analyse whether richer people
tend to prefer one option and, therefore, whether mean (or
median) WTP for that option is a result of greater
“purchasing power” (i.e. ability to pay) as well as
willingness to pay.
Therefore, analyse preferences by:
- social class
Distribution of preferences [and mean
WTP] for surgery by income group
Income (£s per week):
<£100 £100-£150 £150-£230 £230-£350 >£350
Inpatient 4 2 8 5 5
[£25] [£75] [£44] [£56] [£25]
Day case 5 2 8 4 6
[£32] [£50] [£86] [£77] [£43]
USING THE RESULTS
(1)Assuming options are mutually exclusive, other things
(including cost) being equal, the option to be implemented
would be that preferred by whichever group could
compensate the other and still remain better off.
(2)Where other things (e.g. cost) are not equal, if the option
with greater WTP (and hence utility) also has a greater
cost, the decision maker then has to decide whether this
extra cost is worth incurring.
BROADER PRIORITY SETTING
150 in Northern Norway were asked their maximum WTP for three public
sector health care programmes:
- a helicopter ambulance
- 80 more elective heart operations per annum;
- 250 more hip replacements per annum.
WTP RESULTS (NOK; n=143)
Helicopter Hearts Hips
316 306 232
Why WTP is important
• The highest valued QALYs are for the
helicopter, which might reflect a preference
for the „rule of rescue‟.
• Life saving is more highly valued than the
same QALY gain from life extension and
• If the results are valid!
Problems with this method
• Rank orderings do not match WTP orderings
• „Size of the good‟ problems.
Hence, EuroWill, to examine:
• Can different countries do it?
• Various methodological issues (size of good,
closed-ended vs payment scale, informational
effects, marginal approach, econometric issues)
Mean WTP ($PPPGDP, 1999) using
Norway 100 operations = $44 300 procedures = $49
Portugal 150 operations = $24 300 procedures = $27
Denmark 3 to 1 month = $73 Not applicable
UK 100 operations = $78 150 procedures = $81
France 250 operations = $52 Improved outcome = $47
Ireland 100 operations = $59 200 procedures = $69
WHERE NOW FOR WTP?
Conduct of studies
• Do not use „WTP for own care‟
• Open-ended questions are problematic
• Use some form of “payment vehicle”
• Postal questionnaires require careful design
• Tests for effects of income/class are feasible
• „Marginal approach‟ has potential for
choices between close substitutes
WHERE NOW FOR WTP?
Recommendations for research
• Compare payment vehicles
• More testing of the „marginal approach‟
• Extend to broader priority setting contexts
• External validity/strategic bias
THE ECONOMIST OF OLD
“Knows the price of everything and the value of
THE MODERN ECONOMIST
“Knows the price, and is some way towards knowing
the value, of health care at the margin”.
...MAUREEN LIPMAN‟S MOTTO
“Life‟s like a questionnaire. You didn‟t ask to
be sent it, can‟t think what the answers are,
but might as well fill it in and hope it gets a