Concept Architecture
Concept
Theoretically, the Weibull Distribution is a continuous probability distribution widely used to model time-to-event data because of its flexibility in representing increasing, decreasing and constant hazard rates. It generalises the exponential distribution through the inclusion of a shape parameter that determines how the hazard changes over time. In health economics, the Weibull distribution is one of the most frequently used parametric distributions for survival analysis, survival extrapolation and decision modelling.
Mathematically, the Weibull distribution is characterised by a scale parameter and a shape parameter that together determine the probability density, cumulative distribution, survival function and hazard function. The value of the shape parameter governs whether the hazard decreases, remains constant or increases over time. Model parameters are typically estimated using maximum likelihood estimation from censored time-to-event data.
In practice, the Weibull distribution is fitted to patient-level survival data from clinical trials or observational studies to estimate survival curves and extrapolate long-term outcomes beyond observed follow-up. It is extensively applied in health technology assessment, cost-effectiveness analysis and state-transition modelling because of its mathematical tractability and ability to represent a wide range of hazard patterns.
Purpose
Used to model time-to-event data, estimate survival and hazard functions, extrapolate long-term survival and generate transition probabilities for health economic decision models.
Mathematical Formulae
Primary Formula
Probability density function:
f(t) = (k/?)(t/?)??? exp[?(t/?)?]
Where:
t � 0
? = scale parameter
k = shape parameter
Supporting Formulae
Survival function:
S(t) = exp[?(t/?)?]
Hazard function:
h(t) = (k/?)(t/?)???
Cumulative hazard:
H(t) = (t/?)?
Mean survival time:
E(T) = ?�(1 + 1/k)
Related Mathematical Methods
- Maximum likelihood estimation
- Parametric survival modelling
- Weibull regression
- Cox proportional hazards regression
- Kaplan?Meier estimation
- Survival extrapolation
Example
A Weibull model is fitted to oncology trial data with a scale parameter ? = 30 months and a shape parameter k = 1.4. Because k > 1, the hazard of death increases over time. The fitted survival curve is then extrapolated beyond the observed trial period to estimate lifetime survival for cost-effectiveness analysis.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| WEIBULL.DIST | =WEIBULL.DIST(A2,Shape,Scale,FALSE) | Calculates Weibull probability density values. |
| WEIBULL.DIST | =WEIBULL.DIST(A2,Shape,Scale,TRUE) | Calculates cumulative Weibull probabilities. |
| EXP | =EXP(-(A2/Scale)^Shape) | Calculates Weibull survival probabilities. |
| GAMMA | =Scale*GAMMA(1+1/Shape) | Calculates mean survival time. |
VBA (Optional)
VBA can automate Weibull parameter estimation, survival curve generation and long-term survival extrapolation for health economic models.
Sources
Weibull W. A statistical distribution function of wide applicability. Journal of Applied Mechanics. 1951;18:293?297.
Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. 2nd ed. Springer; 2003.
Collett D. Modelling Survival Data in Medical Research. 3rd ed. Chapman & Hall/CRC; 2015.
Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation. Oxford University Press; 2006.
NICE. Health Technology Evaluations: The Manual. National Institute for Health and Care Excellence; 2022.
Related Concepts (2)
Library
Publications
1
NICE DSU Technical Support Document 21: Flexible methods for survival analysis — Rutherford, Lambert, Sweeting, Pennington, Crowther, Abrams & Latimer, TSD 21 ed., 2020 (NICE Decision Support Unit (University of Sheffield))
Guidance extending standard survival analysis to flexible parametric methods — spline-based models, fractional polynomials, mixture and cure models, and relative-survival approaches — for capturing complex hazard functions in economic evaluation.
Frequently Asked Questions (6)
What is the Weibull distribution?
A flexible probability distribution capable of a monotonically increasing or decreasing hazard rate depending on its shape parameter.
Source: Weibull 1951
What does the Weibull shape parameter control?
The Weibull distribution adds a shape parameter to the constant-risk exponential, and it is this parameter that governs how the hazard changes over time. A shape value of one gives a constant hazard, recovering the exponential, while values above one give a hazard that rises steadily and values below one a hazard that falls. This single parameter lets one distribution represent risk that increases or decreases with time, though always in one direction. Its monotonic flexibility explains its wide use. Collett (2015) describes it.
Source: Collett 2015
What hazard shapes does the Weibull distribution represent?
The Weibull distribution represents monotonic hazards: a shape parameter greater than one gives a hazard that increases over time, less than one a hazard that decreases, and equal to one a constant hazard, reducing to the exponential distribution. It cannot represent hazards that turn, such as unimodal or bathtub shapes, since its hazard is always monotonic. This flexibility to capture increasing or decreasing hazards, while remaining monotonic, defines the range of the Weibull and determines when it is appropriate.
Source: Weibull 1951
How does the Weibull generalise the exponential distribution?
The Weibull distribution generalises the exponential distribution by adding a shape parameter that allows the hazard to increase or decrease over time, whereas the exponential, a special case with shape parameter one, has a constant hazard. So the Weibull includes the exponential and extends it to monotonically changing hazards. This makes the Weibull more flexible than the exponential, able to fit data where risk rises or falls, while reducing to the exponential when the hazard is constant.
Source: Collett 2015
When is the Weibull distribution used?
The Weibull distribution is used in survival analysis and reliability when the hazard is expected to increase or decrease monotonically over time, and where its shape fits the data. It is a common candidate parametric survival distribution, chosen where a monotonic hazard is plausible, and it is used for extrapolation, projecting survival beyond the data according to its increasing or decreasing hazard. Its fit and the plausibility of its extrapolated monotonic hazard are compared with other distributions before selection.
Source: Weibull 1951
What are the limitations of the Weibull distribution?
The Weibull distribution's hazard is monotonic, so it cannot represent hazards that turn, such as unimodal hazards that rise then fall or bathtub shapes, limiting its fit where the hazard is not monotonic. When extrapolated, its increasing or decreasing hazard is projected indefinitely, which may be implausible far beyond the data. As with any parametric model, its extrapolation depends on the assumed form. These limitations mean the Weibull is used where a monotonic hazard is appropriate and compared with alternatives.
Source: Collett 2015
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 24 Oct 2025
Content version: 1.0.0
Canonical Identity
- Term code
- HE-EM-SM-090
Stable URI · Machine-readable · Resolvable · CC BY 4.0