Concept Architecture
Concept
Theoretically, Population Expected Value of Perfect Information (Population EVPI) is the total expected economic value of eliminating all decision uncertainty across the entire population expected to benefit from improved healthcare decisions. It extends the Expected Value of Perfect Information from an individual patient to the relevant decision population and is founded on Bayesian decision theory and value of information analysis. In health economics, Population EVPI is used to determine whether the societal benefits of further research justify its cost.
Mathematically, Population EVPI is calculated by multiplying the per-person Expected Value of Perfect Information by the effective population expected to be affected during the decision relevance period. The effective population may be adjusted for technology uptake, disease incidence, prevalence, discounting and the time horizon over which the decision remains relevant. Population EVPI therefore represents the maximum amount society should be willing to invest in eliminating all uncertainty.
In practice, Population EVPI is estimated after probabilistic sensitivity analysis has produced the per-person EVPI. Health economists estimate the eligible patient population over the decision relevance period and apply appropriate adjustments for implementation and discounting. Population EVPI is routinely used by health technology assessment agencies to prioritise research funding, compare competing research proposals and determine whether additional evidence is economically worthwhile.
Purpose
Used to estimate the total societal value of eliminating decision uncertainty across the affected population, supporting research prioritisation and funding decisions.
Mathematical Formulae
Primary Formula
Population EVPI = EVPI ? N
where:
EVPI = Expected Value of Perfect Information per person
N = effective population affected during the decision relevance period
Supporting Formulae
Effective population:
N = Eligible Population ? Uptake Rate
Discounted Population EVPI:
Population EVPI = EVPI ? ?? (N? � (1 + r)?)
where:
N? = eligible population in year t
r = annual discount rate
t = time period
Related Mathematical Methods
- Expected Value of Perfect Information
- Expected Value of Partial Perfect Information
- Expected Value of Sample Information
- Expected Net Benefit of Sampling
- Value of Information Analysis
- Probabilistic Sensitivity Analysis
- Discounting
Example
A probabilistic sensitivity analysis estimates a per-person EVPI of �420. The intervention is expected to affect 18,000 patients annually for five years. After accounting for expected uptake and discounting, the effective decision population is estimated to be 75,000 patients.
Population EVPI = �420 ? 75,000
Population EVPI = �31.5 million
This indicates that society should be willing to invest up to �31.5 million to eliminate all uncertainty surrounding the reimbursement decision.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| SUMPRODUCT | =SUMPRODUCT(B2:B6,C2:C6) | Calculate the discounted effective population across multiple years. |
| PV | =PV(DiscountRate,Years,0,-AnnualPopulation) | Estimate the present value of the eligible population when appropriate. |
| PRODUCT | =EVPI_per_person*EffectivePopulation | Calculate Population EVPI. |
| IF | =IF(B2>ResearchCost,"Research justified","Research not justified") | Compare Population EVPI with the expected cost of additional research. |
VBA (Optional)
VBA can automate discounted population calculations and estimate Population EVPI for alternative implementation scenarios and decision time horizons.
Sources
- Claxton K. The irrelevance of inference: a decision-making approach to the stochastic evaluation of health care technologies. Journal of Health Economics. 1999;18(3):341?364.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press.
- Fenwick E, Claxton K, Sculpher M. Representing uncertainty: the role of cost-effectiveness acceptability curves. Health Economics. 2001;10(8):779?787.
- ISPOR Value of Information Good Practice Reports.
- NICE. NICE Health Technology Evaluations: The Manual.
Related Concepts (2)
Library
Publications
1
Value of Information Analysis for Research Decisions — An Introduction: Report 1 of the ISPOR Value of Information Analysis Emerging Good Practices Task Force — Fenwick, Steuten, Knies, Ghabri, Basu, Murray, Koffijberg, Strong, Sanders Schmidler & Rothery, Vol. 23, No. 2 ed., 2020 (Value in Health)
The introductory ISPOR good-practice report on value-of-information (VOI) analysis, explaining how VOI quantifies the value of reducing decision uncertainty through further research and where it fits in resource-allocation decisions.
Journal ArticleView source →
Frequently Asked Questions (6)
What is population EVPI?
The expected value of perfect information calculated for an entire affected patient population over time, rather than for a single patient.
Source: Claxton & Posnett 1996
What makes population EVPI larger than individual EVPI?
The value of resolving uncertainty is not confined to one patient, since a better decision benefits everyone treated under it while it remains current. Population EVPI scales the per-patient value up by the number of future patients the decision will affect over its relevant lifetime, so it can be far larger than the individual figure. A common decision affecting many patients over years can carry a population value large enough to justify substantial research, even when the per-patient value is small. Size and duration of the population drive it. Claxton and Sculpher (2006) describe this.
Source: Claxton & Sculpher 2006
How is population EVPI calculated?
Population EVPI is calculated by taking the per-patient expected value of perfect information and multiplying it by the number of patients who will be affected by the decision over its relevance period, sometimes discounting future patients' value. The per-patient EVPI comes from comparing expected outcomes under perfect and current information, and the population size and relevance period scale it to everyone the decision affects. The result is the total expected value of resolving all uncertainty for the whole population over the time the decision applies.
Source: Raiffa & Schlaifer 1961
Why is population EVPI used?
Population EVPI is used because research provides information that benefits not just one patient but all those affected by the decision over its lifetime, so the value of information must be scaled to the population to be compared meaningfully with the cost of research. A per-patient value would understate the worth of resolving uncertainty. Population EVPI gives the total upper bound on the value of further research, so if it falls below likely research costs, no study could be worthwhile, making it a screening measure.
Source: Claxton & Posnett 1996
What determines the size of population EVPI?
The size of population EVPI is determined by the per-patient EVPI, reflecting the decision uncertainty and the value at stake per patient, and by the number of patients affected over the decision's relevance period, which depends on the incidence or prevalence and how long the decision remains in force. A larger affected population or a longer relevance period raises population EVPI, as does greater per-patient decision uncertainty. So both the intensity of uncertainty and the scale of the population and time horizon govern its magnitude.
Source: Claxton & Posnett 1996
What are the limitations of population EVPI?
Population EVPI assumes all uncertainty could be eliminated, giving an upper bound rather than the value of a feasible study, and it depends on the model and its assumed parameter distributions, and on the estimated population size and decision relevance period, both of which require forecasting the future and are uncertain. Its magnitude scales directly with these estimates. These limitations mean population EVPI is used as a screening bound whose sensitivity to the population and relevance-period assumptions is examined, complemented by partial and sample information measures for feasible research.
Source: Raiffa & Schlaifer 1961
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 31 Oct 2025
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/population-evpi
- Term code
- HE-EM-VI-038
Stable URI · Machine-readable · Resolvable · CC BY 4.0