Concept Architecture
Concept
Theoretically, Expected Net Benefit of Sampling (ENBS) is a value of information measure that quantifies the net economic value of conducting additional research after accounting for both the expected value of the information obtained and the cost of generating that information. It is founded on Bayesian decision theory and extends Expected Value of Sample Information by incorporating research costs into the decision-making process. In health economics, ENBS is used to determine whether a proposed study is economically worthwhile and to identify the optimal research design or sample size.
Mathematically, ENBS is calculated as the difference between the Expected Value of Sample Information and the total expected cost of conducting the proposed research. A positive ENBS indicates that the expected reduction in decision uncertainty exceeds the cost of the study, whereas a negative ENBS indicates that the proposed research is not economically justified.
In practice, ENBS is estimated after calculating EVSI and determining all relevant research costs, including study design, participant recruitment, follow-up, data management and analysis. ENBS is routinely applied in health technology assessment to compare competing study designs, optimise sample sizes and prioritise research investments that provide the greatest expected societal benefit.
Purpose
Used to determine whether additional research provides sufficient expected value to justify its cost and to optimise the design of future health economic studies.
Mathematical Formulae
Primary Formula
ENBS = EVSI ? C
where:
ENBS = Expected Net Benefit of Sampling
EVSI = Expected Value of Sample Information
C = total expected research cost
Supporting Formulae
Population ENBS:
Population ENBS = Population EVSI ? Population Research Cost
Decision rule:
If ENBS > 0, conduct the research.
If ENBS � 0, additional research is not economically justified.
Related Mathematical Methods
- Expected Value of Sample Information
- Expected Value of Perfect Information
- Expected Value of Partial Perfect Information
- Bayesian Decision Theory
- Value of Information Analysis
- Research Prioritisation
- Sample Size Optimisation
Example
A proposed randomised controlled trial has an estimated population EVSI of �11.5 million. The total projected cost of conducting the study is �8.2 million.
ENBS = �11.5 million ? �8.2 million
ENBS = �3.3 million
Because the ENBS is positive, the expected value of reducing decision uncertainty exceeds the research cost, indicating that the proposed study is economically worthwhile.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| SUM | =SUM(CostRange) | Calculate the total expected research cost. |
| AVERAGE | =AVERAGE(EVSIRange) | Estimate the expected value of sample information from simulation results. |
| IF | =IF(B2-C2>0,"Research justified","Research not justified") | Determine whether the expected net benefit of sampling is positive. |
| MAX | =MAX(ENBSRange) | Identify the study design or sample size with the greatest expected net benefit. |
VBA (Optional)
VBA can automate ENBS calculations across alternative study designs and sample sizes to identify the research strategy that maximises expected societal benefit.
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.
- Ades AE, Lu G, Claxton K. Expected value of sample information calculations in medical decision modelling. Medical Decision Making. 2004;24(2):207?227.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press.
- ISPOR Value of Information Good Practice Reports.
- NICE. NICE Health Technology Evaluations: The Manual.
Related Concepts (2)
Library
Publications
1
Value of Information Analytical Methods: Report 2 of the ISPOR Value of Information Analysis Emerging Good Practices Task Force — Rothery, Strong, Koffijberg, Basu, Ghabri, Knies, Murray, Sanders Schmidler, Steuten & Fenwick, Vol. 23, No. 3 ed., 2020 (Value in Health)
The methods companion to the ISPOR VOI series, giving detailed algorithms and software guidance for computing EVPI, EVPPI, EVSI and the expected net benefit of sampling, with recommendations for selecting methods by decision-problem features.
Journal ArticleView source →
Frequently Asked Questions (6)
What is the expected net benefit of sampling?
A research prioritisation measure calculated as the expected value of sample information from a study minus the expected cost of conducting it.
Source: Claxton & Posnett 1996
Why does the expected net benefit of sampling weigh value against cost?
Information from a study has value, but gathering it has a cost, and a study is only worth commissioning if the first exceeds the second. The expected net benefit of sampling captures this by setting the expected value of the sample information a study would provide against its expected cost, so a positive figure signals that the study is worthwhile. Comparing the figure across possible studies also shows which offers the best return. It turns research funding into a value-for-money question. Claxton and Sculpher (2006) describe this measure.
Source: Claxton & Sculpher 2006
How is the expected net benefit of sampling computed?
The expected net benefit of sampling is computed by calculating the expected value of sample information for a proposed study, from value-of-information analysis given its design and sample size, and subtracting the expected cost of the study, including research and associated costs. The result is the net expected worth of the study. Computing it for different designs and sample sizes allows the most efficient study, and the optimal sample size, to be identified as those maximising the expected net benefit of sampling.
Source: Raiffa & Schlaifer 1961
Why is the expected net benefit of sampling used?
The expected net benefit of sampling is used to prioritise and design research by judging whether a study is worth its cost and which design is most efficient, since it directly compares the value of a study's information with the cost of obtaining it. A positive value indicates a study expected to be worthwhile, and maximising it identifies the best design and sample size. This helps direct limited research resources to studies whose value exceeds their cost, making it a practical tool for research planning.
Source: Claxton & Posnett 1996
How does the expected net benefit of sampling identify optimal study design?
The expected net benefit of sampling identifies optimal study design by being computed across alternative designs and sample sizes: as sample size increases, the expected value of sample information rises while the cost also rises, so the expected net benefit of sampling typically peaks at an intermediate size. The design and sample size that maximise it are the most efficient, balancing the value of additional information against its cost. Comparing the measure across options thus points to the study offering the greatest net value.
Source: Raiffa & Schlaifer 1961
What are the limitations of the expected net benefit of sampling?
The expected net benefit of sampling depends on the expected value of sample information, which rests on the model, its assumed parameter distributions, and the population and decision relevance period, and on the assumed study costs, so uncertainty or error in these carries into the result. Estimating the expected value of sample information can be computationally demanding, and forecasting costs and the future decision context is uncertain. These limitations mean the measure is treated as an informative but approximate guide, with its sensitivity to key assumptions examined.
Source: Claxton & Posnett 1996
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
- Term code
- HE-EM-VI-012
Stable URI · Machine-readable · Resolvable · CC BY 4.0