Signature
C_AE = sum_(k=1)^K [p_k * c_k]
| Inputs | Definition | Unit |
|---|---|---|
p_k | Probability of an episode of adverse event type k over the period costed | probability from 0 to 1 |
c_k | Cost of managing one episode of event type k, built up from the admissions, visits and drugs it triggers, often by micro-costing | currency per episode |
C_AE | Expected cost of managing adverse events per patient over the stated period | currency per patient, in a stated price year |
|---|
KNumber of adverse event types included in the costing, equal to the length of the p_k and c_k lists (count)
Function
Adverse event rate to model input function
Maps the adverse event counts in a trial safety table, with their denominators of patients or person-time, to the probability of an event in one model cycle, and maps those probabilities to the expected adverse event cost and QALY loss per patient. The general rate and probability conversions are on the Transition Probability page (HE-FN-TP-001); the records here apply them to adverse event data and add the cost and QALY weighting.
Try this function
Implementations
Excel
Expected adverse event cost with SUMPRODUCT
With the event probabilities in a range named AEProbs and the episode costs in a range of the same size named AEEpisodeCosts, Excel returns the expected cost per patient.
=SUMPRODUCT(AEProbs,AEEpisodeCosts)
Assumptions
One cost per episode of each adverse event type
Every episode of event type k costs the same, whatever the arm and whenever it occurs. Where later episodes or episodes in one arm are managed differently, the event type is split.
Stated rule for the adverse events costed
The events included are stated with a reason. Restricting them to a severity grade, a frequency cut-off or events that differ between arms removes costs, and the restriction is a structural choice to test in scenario analysis.
Worked examples
Expected diarrhoea cost on the new drug over 24 weeks
Illustrative figures: with a probability of 0.18 and £1,800 to manage each episode, the expected adverse event cost is £324 per patient over 24 weeks.
p_k = [0.18]; c_k = [1800]; K = 1; C_AE = 324
Expected diarrhoea cost on the comparator over 24 weeks
With a probability of 0.08 and the same episode cost, the expected cost is £144 per patient, so the new drug adds £180 per patient in adverse event costs.
p_k = [0.08]; c_k = [1800]; K = 1; C_AE = 144
Common errors
Costing patients with an event when adverse events recur
A model that costs every episode needs the event count, not the number of patients with at least one event. Illustrative figures: if the 36 patients with diarrhoea have 60 episodes between them, 0.30 episodes per patient replace the probability of 0.18, and using 0.18 understates the 24-week cost by £216 per patient.
Sources
Review of adverse effect cost parameters in HTA decision models
Craig D, McDaid C, Fonseca T, Stock C, Duffy S, Woolacott N. Are adverse effects incorporated in economic models? An initial review of current practice. Health Technology Assessment. 2009;13(62). Review of 80 NIHR HTA reports from 2004 to 2007: of the models that included adverse effects, 79% used a cost of adverse effects parameter. The authors recommend clearer reporting of how adverse effects are included or why they are left out.
NICE table of disaggregated costs including adverse events
National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). 2022, updated March 2026. Section 4.10.4, which asks for a table including disaggregated costs and the decrements associated with further interventions and adverse events.
Methods for estimating adverse event costs in economic evaluation
Ghabri S, Dawoud D, Drummond M. Methods for including adverse events in economic evaluations: suggestions for improvement. Value in Health. 2024;27(7):936-942. Discusses how adverse events are incorporated in decision models and how their consequences for costs and quality of life loss are estimated.
Canonical Identity
Stable URI · Machine-readable · Resolvable · CC BY 4.0