Pros and Cons of Event Based Modelling in Economic Evaluation
Bibliographic record
Abstract
Background: Event based models are driven by the occurrence of clinical events such as primary clinical endpoints in randomised trials or adverse events associated with treatment. Statistical analyses of individual-patient data are used to determine event rates and further statistical analyses are performed to estimate survival, costs and health-related quality of life conditional on an event having occurred. For economic evaluations of health-care programmes such an approach has several advantages as extrapolation is facilitated, it is possible to explore cost-effectiveness in different risk groups and it makes it possible to bring in relevant external evidence such as pooled treatment effect. However, event based modelling also poses methodological challenges concerning not only technical issues but also conceptual ones regarding the scientific method. The aim of this paper is to explore and discuss these methodological challenges. Methods: Published event based models were reviewed and examples from a recently developed event based model in acute coronary syndrome were used to discuss and exemplify several of the methodological issues involving event based modelling. Results: The event based modelling approach normally uses randomised evidence to determine rates of clinical events, but given that an event has occurred, life expectancy, costs and health-related quality of life are estimated conditional on the event rather than randomised treatment. Some would argue this is appropriate as treatment only affect costs and quality of life through the impact of events rates. However, others would argue that such an approach is inappropriate as it adds 'structure' to the randomised evidence in terms of costs, life-expectancy and quality of life assuming conditional independence. Regarding more technical aspects, several methodological issues need to be considered as relatively advanced statistical models are combined in a decision-analytic framework to determine cost-effectiveness. The most important advantage of event based modelling identified in this work is that it provides a tool to estimate cost-effectiveness in a way relevant for policy, i.e. enabling the estimation of lifetime costs and health outcomes in different subgroups utilising all relevant evidence. Some of the challenges identified in this work include the choice of covariates to be included in the statistical analyses and the presentation of the results and probabilistic sensitivity analyses as much of the inputs into the cost-effectiveness model will be based on risk equations which can potentially define a very large number of subgroups. Conclusion: Although there are still methodological issues that need addressing in this framework, event based modelling is a useful method for providing relevant cost-effectiveness evidence.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.173 | 0.310 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.005 | 0.007 |
| Bibliometrics | 0.005 | 0.009 |
| Science and technology studies | 0.001 | 0.007 |
| Scholarly communication | 0.011 | 0.013 |
| Open science | 0.006 | 0.007 |
| Research integrity | 0.009 | 0.010 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".