Concept Architecture
Concept
Theoretically, the Exponential Distribution is a continuous probability distribution used to model the time until an event occurs when the hazard remains constant over time. It is founded on probability theory and survival analysis and is the only continuous distribution possessing the memoryless property. In health economics, the exponential distribution is widely used for survival modelling, transition probability estimation and extrapolation within decision-analytic models when a constant hazard assumption is appropriate.
Mathematically, the exponential distribution is fully characterised by a single rate parameter representing the constant hazard. The probability density, survival and cumulative distribution functions are derived directly from this parameter, providing a simple mathematical framework for modelling time-to-event outcomes. Because the hazard remains constant, survival declines exponentially with increasing time.
In practice, the exponential distribution is fitted to survival data using maximum likelihood estimation or Bayesian methods and compared with alternative parametric distributions such as the Weibull, Gompertz and log-normal models. It is routinely applied in Markov models, cost-effectiveness analyses and health technology assessments when observed event rates are approximately constant over time.
Purpose
Used to model time-to-event outcomes under a constant hazard assumption, estimate survival probabilities, derive transition probabilities and support health economic decision modelling.
Mathematical Formulae
Primary Formula
Probability density function:
f(t) = ?exp(??t)
Supporting Formulae
Survival function:
S(t) = exp(??t)
Hazard function:
h(t) = ?
Cumulative distribution function:
F(t) = 1 ? exp(??t)
Mean survival time:
E(T) = 1 � ?
Variance:
Var(T) = 1 � ?�
Related Mathematical Methods
- Maximum likelihood estimation
- Parametric survival modelling
- Constant hazard modelling
- Kaplan?Meier estimation
- Cox proportional hazards model
Example
A disease has a constant annual mortality hazard of ? = 0.08.
The probability of surviving five years is:
S(5) = exp(?0.08 ? 5)
S(5) = exp(?0.40) = 0.670
The expected survival time is:
E(T) = 1 � 0.08 = 12.5 years
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| EXP | =EXP(-B2*A2) | Calculate survival probability using the exponential distribution. |
| LN | =-LN(B2)/A2 | Estimate the hazard rate from observed survival probability. |
| EXPON.DIST | =EXPON.DIST(A2,B2,TRUE) | Calculate cumulative probabilities for exponentially distributed survival times. |
VBA (Optional)
Automate fitting of exponential survival models and generate survival projections for health economic evaluations.
Sources
- Collett D. Modelling Survival Data in Medical Research. CRC Press.
- Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. Springer.
- Lawless JF. Statistical Models and Methods for Lifetime Data. Wiley.
- 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 14: Survival analysis for economic evaluations alongside clinical trials – extrapolation with patient-level data — Nicholas R. Latimer, TSD 14 ed., 2013 (NICE Decision Support Unit (University of Sheffield))
The reference guidance on survival analysis for economic evaluation: fitting standard parametric models (exponential, Weibull, Gompertz, log-logistic, log-normal) to censored trial data and extrapolating to estimate lifetime survival benefit, with a process guide for model selection and justification.
Frequently Asked Questions (6)
What is the exponential distribution?
A probability distribution characterised by a constant hazard rate over time, used to model event timing when risk is assumed not to change.
Source: Collett 2015
What shape does the exponential distribution give a survival curve?
Because the exponential distribution assumes a constant hazard, the survival curve it produces declines smoothly and steadily, falling by the same proportion in each equal interval of time. Plotted on a logarithmic scale the curve becomes a straight line, a feature used to check whether data plausibly follow it. The steady proportional decline reflects a risk that never rises or falls with age or time since diagnosis. This single fixed shape is both its simplicity and its main limitation. Collett (2015) describes this form.
Source: Collett 2015
What is the key property of the exponential distribution?
The key property of the exponential distribution is its constant hazard, meaning the instantaneous risk of the event does not change over time, so the process is memoryless: the future risk is unaffected by how long an individual has already survived. This implies that survival declines exponentially and the expected remaining time is the same at any point. The memoryless, constant-hazard property makes the distribution mathematically simple but often unrealistic for survival data where risk changes with time.
Source: Collett 2015
When is the exponential distribution appropriate?
The exponential distribution is appropriate when the hazard can reasonably be treated as constant over the relevant period, which may hold over short intervals or for processes without ageing or progression effects. It serves as a simple model or approximation where risk is roughly stable. However, where the hazard clearly changes over time, rising with age or disease progression or falling after an initial period, the exponential distribution is inappropriate, so its constant-hazard assumption is checked and a more flexible distribution used if it fails.
Source: Kalbfleisch & Prentice 2002
What are the advantages of the exponential distribution?
The exponential distribution is the simplest survival distribution, defined by a single parameter, so it is easy to fit, interpret, and use, and its constant hazard gives tractable expressions for survival and mean time. This simplicity makes it a convenient baseline and a useful component in more complex models, and it can adequately represent data where the hazard is genuinely stable. Its parsimony is an advantage where a constant hazard is a reasonable approximation and a simple model suffices.
Source: Collett 2015
What are the limitations of the exponential distribution?
The exponential distribution's main limitation is its assumption of a constant hazard, which is often unrealistic, since real risks usually change over time, so the distribution can fit poorly and mislead when extrapolated where the hazard actually varies. Its single parameter cannot represent increasing or decreasing hazards. Where the constant-hazard assumption does not hold, more flexible distributions, such as the Weibull or generalised gamma, are needed. These limitations mean the exponential distribution is used only where a constant hazard is genuinely plausible.
Source: Collett 2015
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
- Term code
- HE-EM-SM-018
Stable URI · Machine-readable · Resolvable · CC BY 4.0