Concept Architecture
Concept
Theoretically, the Exponential Model is a parametric survival model that assumes the hazard of an event remains constant throughout the observation period. It is founded on the exponential probability distribution and survival analysis and provides the simplest mathematical representation of time-to-event data. In health economics, the exponential model is widely used to estimate survival, derive transition probabilities and extrapolate long-term outcomes when a constant hazard assumption is considered appropriate.
Mathematically, the exponential model is characterised by a single hazard parameter that completely determines the probability density, survival function and cumulative hazard. Because the hazard remains constant, survival declines exponentially over time and the model possesses the memoryless property. Model parameters are typically estimated using maximum likelihood estimation.
In practice, the exponential model is fitted to clinical trial and observational survival data and compared with alternative parametric models such as the Weibull, Gompertz, log-normal and log-logistic distributions. It is routinely applied in Markov models, partitioned survival models and health technology assessments when observed event rates remain relatively stable over time.
Purpose
Used to model survival under a constant hazard assumption, estimate transition probabilities, extrapolate long-term outcomes and provide survival inputs for health economic decision models.
Mathematical Formulae
Primary Formula
S(t) = exp(??t)
Supporting Formulae
Hazard function:
h(t) = ?
Probability density function:
f(t) = ?exp(??t)
Cumulative hazard:
H(t) = ?t
Mean survival time:
E(T) = 1 � ?
Related Mathematical Methods
- Exponential distribution
- Maximum likelihood estimation
- Parametric survival modelling
- Kaplan?Meier estimation
- Cox proportional hazards model
- Model selection using AIC and BIC
Example
A survival model estimates a constant annual hazard of ? = 0.06.
The estimated five-year survival probability is:
S(5) = exp(?0.06 ? 5)
S(5) = exp(?0.30) = 0.741
This model predicts that approximately 74.1% of patients survive for at least five years.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| EXP | =EXP(-B2*A2) | Calculate survival probability using an exponential survival model. |
| LN | =-LN(B2)/A2 | Estimate the hazard rate from observed survival probabilities. |
| EXPON.DIST | =EXPON.DIST(A2,B2,TRUE) | Calculate cumulative probabilities under the exponential model. |
VBA (Optional)
Automate estimation of exponential survival models and generate projected survival curves for health economic analyses.
Sources
- Collett D. Modelling Survival Data in Medical Research. CRC Press.
- Lawless JF. Statistical Models and Methods for Lifetime Data. Wiley.
- Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. Springer.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press.
- NICE. Health Technology Evaluation Manual.
Related Concepts (2)
Library
Publications
1
NICE DSU Technical Support Document 15: Cost-effectiveness modelling using patient-level simulation — Davis, Stevenson, Tappenden & Wailoo, TSD 15 ed., 2014 (NICE Decision Support Unit (University of Sheffield))
Guidance on individual patient-level (microsimulation) cost-effectiveness modelling — when to use it in preference to cohort models, how to structure it, and how to handle the associated computational and uncertainty challenges.
Frequently Asked Questions (6)
What is an exponential model?
A survival model assuming time-to-event data follow an exponential distribution, implying a constant hazard rate throughout the modelled time horizon.
Source: Collett 2015
What makes the exponential model the simplest survival model?
The exponential model needs only a single parameter, the constant hazard rate, to describe the entire course of survival, which makes it the simplest of the parametric survival models to fit and interpret. That economy is bought at the cost of realism, since it forces risk to stay the same at every age and time since diagnosis. It suits situations where risk genuinely does not change, or serves as a rough baseline against which more flexible models are compared. For most chronic diseases it is too rigid. Latimer (2013) discusses its use.
Source: Latimer 2013
What does the exponential model assume?
The exponential model assumes a constant hazard, meaning the instantaneous risk of the event is the same at every point in time, so the process is memoryless and survival declines exponentially. This single assumption defines the model, requiring only one parameter, the constant hazard rate. Because it holds risk fixed over time, the exponential model cannot represent hazards that increase or decrease, so its use rests on the assumption that the hazard genuinely does not change over the modelled period.
Source: Collett 2015
When is an exponential model used?
An exponential model is used when the hazard can reasonably be treated as constant, such as over short horizons or for processes without ageing or progression effects, and where its simplicity is an advantage. It provides a parsimonious baseline. However, in survival extrapolation and where the hazard clearly changes over time, the exponential model is often too restrictive, so its fit is compared with more flexible distributions, and it is chosen only if the constant-hazard assumption is supported by the data and plausible for the projection.
Source: Latimer 2013
How is the fit of an exponential model checked?
The fit of an exponential model is checked by examining whether the hazard appears constant, for instance by plotting an estimate of the hazard over time or checking whether the cumulative hazard is roughly linear, and by comparing the model's fit with distributions allowing a changing hazard, using goodness-of-fit measures and visual comparison of the fitted and observed survival. If these show the hazard varies or the exponential fits worse than alternatives, the model is rejected in favour of a more flexible distribution.
Source: Collett 2015
What are the limitations of an exponential model for extrapolation?
For extrapolation, the exponential model's constant hazard is a strong limitation, since it projects survival declining at the same rate indefinitely, which is often implausible where risk changes with time, and it cannot capture increasing or decreasing hazards that affect long-term survival. Because extrapolated survival drives mean estimates, an inappropriate constant-hazard assumption can badly misestimate them. So the exponential model is used for extrapolation only where a constant hazard is genuinely plausible, and more flexible distributions are considered where the hazard varies.
Source: Latimer 2013
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 20 Oct 2025
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/exponential-model
- Term code
- HE-EM-SM-019
Stable URI · Machine-readable · Resolvable · CC BY 4.0