Concept Architecture
Concept
Theoretically, a Discrete Choice Experiment (DCE) is a stated preference method used to estimate preferences by observing choices between hypothetical alternatives described by multiple attributes. It is founded on random utility theory and consumer choice theory, and exists to quantify the trade-offs individuals are willing to make between different characteristics of healthcare interventions, services or policies.
Mathematically, a Discrete Choice Experiment is represented using random utility models in which the utility of each alternative comprises a systematic component explained by observed attributes and a random component representing unobserved influences on choice. Model parameters are estimated from repeated observed choices and are used to quantify attribute preferences, marginal rates of substitution and willingness to pay where a cost attribute is included.
In practice, Discrete Choice Experiments are implemented by developing an efficient experimental design, constructing choice sets, administering preference surveys and analysing responses using econometric choice models. They are widely used in health economics to value health services, estimate patient and public preferences, inform policy decisions and derive utility or willingness-to-pay estimates.
Purpose
Used to quantify preferences for healthcare interventions and services, estimate trade-offs between attributes, derive willingness to pay, inform health technology assessment, and support healthcare decision making.
Mathematical Formulae
Primary Formula
U?? = V?? + �??
where:
- U?? = total utility of alternative j for individual i
- V?? = systematic (observable) utility
- �?? = random utility component
Supporting Formulae
Systematic utility:
V?? = ??X? + ??X? + ? + ??X?
Conditional logit choice probability:
P?? = exp(V??) / ????? exp(V??)
Marginal rate of substitution:
MRS = ??? / ??
Willingness to pay (when cost is included):
WTP = ??? / ?_cost
Related Mathematical Methods
- Random utility theory
- Conditional logit model
- Mixed logit model
- Multinomial logit model
- D-efficient design
- Latent class model
- Maximum likelihood estimation
Example
Patients choose between two hypothetical medicines.
| Attribute | Medicine A | Medicine B |
|---|---|---|
| Effectiveness | 80% | 90% |
| Mild side effects | 10% | 20% |
| Monthly cost | �25 | �40 |
Responses from 1,000 participants are analysed using a mixed logit model. The estimated coefficients show that respondents strongly prefer higher effectiveness and lower cost, allowing willingness to pay for improved effectiveness to be calculated from the ratio of the estimated coefficients.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| EXP | =EXP(B2) | Calculates the exponential utility component for logit models. |
| SUM | =SUM(C2:C5) | Calculates the denominator of the choice probability equation. |
| Division | =C2/$C$6 | Calculates predicted choice probabilities. |
| SUMPRODUCT | =SUMPRODUCT(B2:E2,B10:E10) | Calculates systematic utility from attribute coefficients. |
VBA (Optional)
Automate construction of choice sets, calculation of predicted utilities and probabilities, and generation of summary outputs for discrete choice experiments.
Sources
- Louviere JJ, Hensher DA, Swait JD. Stated Choice Methods: Analysis and Applications. Cambridge University Press.
- Train KE. Discrete Choice Methods with Simulation. Cambridge University Press.
- Bridges JFP, Hauber AB, Marshall D, et al. Conjoint Analysis Applications in Health: A Checklist. Value in Health. 2011.
- Lancsar E, Louviere J. Conducting Discrete Choice Experiments to Inform Healthcare Decision Making. Pharmacoeconomics. 2008.
- Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. Oxford University Press.
Related Concepts (3)
Library
Publications
2
NICE DSU Technical Support Document 11: Alternatives to EQ-5D for Generating Health State Utility Values — Brazier, Rowen, TSD 11 ed., 2011 (NICE Decision Support Unit (University of Sheffield))
Guidance on alternatives to EQ-5D — including SF-6D, HUI, condition-specific preference-based measures, direct valuation and vignette methods — for generating health-state utility values.
Conjoint Analysis Applications in Health — A Checklist: A Report of the ISPOR Good Research Practices for Conjoint Analysis Task Force — Bridges, Hauber, Marshall, Lloyd, Prosser, Regier, Johnson & Mauskopf, Vol. 14, No. 4 ed., 2011 (Value in Health)
The ISPOR good-practice checklist for conjoint analysis and discrete-choice experiments in health — the stated-preference methods used to elicit patient and public preferences over treatment attributes for value assessment and priority-setting.
Journal ArticleView source →
Frequently Asked Questions (6)
What is a discrete choice experiment?
A quantitative technique for eliciting preferences in which respondents choose among hypothetical alternatives defined by varying attributes and levels.
Source: Louviere, Hensher & Swait 2000
How is a discrete choice experiment designed?
The good or service is described by a small number of attributes, each taking defined levels, and combinations of those levels form alternatives which are grouped into choice sets. The combinations shown are selected by a formal experimental design rather than at random, so that the effect of each attribute can be estimated separately and the maximum information extracted from a limited number of questions. Attributes are normally developed through qualitative work with the people who will answer, since those chosen by analysts routinely omit what respondents actually weigh. Including an option to decline allows the value of the service as a whole to be estimated rather than only the relative value of its features.
Source: Louviere, Hensher & Swait 2000
How is a discrete choice experiment analysed?
The analysis assumes each respondent attaches a value to each attribute level, selects the alternative with the highest total, and that an unobserved random component accounts for departures from that rule. Estimating the model yields a weight for each level, from which the relative importance of attributes can be compared. Where cost is an attribute, dividing another attribute's weight by the cost weight expresses it in money, and where a health measure is included the same operation expresses it in health. Specifications allowing weights to vary between respondents usually describe the data better than assuming everyone values things identically.
Source: Lancsar & Louviere 2008
What is a discrete choice experiment used for in health?
It values features of services that no clinical outcome measure captures, including waiting time, travel distance, continuity of clinician and consultation mode. It establishes how patients weigh effectiveness against side effects when choosing between treatments. It predicts how uptake would respond to a service being configured differently, which is its distinctive contribution, since the estimated weights apply to any combination of levels rather than only to those presented. It is also used to examine what influences where health professionals choose to work.
Source: de Bekker-Grob, Ryan & Gerard 2012
What threatens the validity of a discrete choice experiment?
The design bounds the answer, since any characteristic not included is valued at zero by construction and the levels chosen determine how important each attribute can appear. Respondents frequently attend to only some attributes, which the standard model assumes they do not. Choices carry no consequence, so stated willingness to pay exceeds what people actually pay. And evidence that the estimates predict real behaviour remains limited, which is why the method is on firmer ground establishing relative weights than producing absolute monetary values.
Source: Bridges et al. 2011
How does a discrete choice experiment differ from contingent valuation?
Contingent valuation asks directly what a respondent would pay for a described good, which produces a monetary value immediately and requires the respondent to name a figure. A discrete choice experiment infers values from repeated choices between alternatives, which avoids asking anyone to price anything and yields weights for every attribute rather than a single amount. The choice format also allows the good to be varied systematically, so sensitivity to quantity can be tested within the design rather than requiring separate subsamples.
Source: Ryan, Gerard & Amaya-Amaya 2008
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 31 Jul 2025
Content version: 1.0.0
Canonical Identity
- Term code
- HE-EE-CBA-016
Stable URI · Machine-readable · Resolvable · CC BY 4.0