Is the whole larger than the sum of its parts? Impact of missing data imputation in economic evaluation conducted alongside randomized controlled trials
Bibliographic record
Abstract
Outcomes in economic evaluations, such as health utilities and costs, are products of multiple variables, often requiring complete item responses to questionnaires. Therefore, missing data are very common in cost-effectiveness analyses. Multiple imputations (MI) are predominately recommended and could be made either for individual items or at the aggregate level. We, therefore, aimed to assess the precision of both MI approaches (the item imputation vs. aggregate imputation) on the cost-effectiveness results. The original data set came from a cluster-randomized, controlled trial and was used to describe the missing data pattern and compare the differences in the cost-effectiveness results between the two imputation approaches. A simulation study with different missing data scenarios generated based on a complete data set was used to assess the precision of both imputation approaches. For health utility and cost, patients more often had a partial (9% vs. 23%, respectively) rather than complete missing (4% vs. 0%). The imputation approaches differed in the cost-effectiveness results (the item imputation: - 61,079€/QALY vs. the aggregate imputation: 15,399€/QALY). Within the simulation study mean relative bias (< 5% vs. < 10%) and range of bias (< 38% vs. < 83%) to the true incremental cost and incremental QALYs were lower for the item imputation compared to the aggregate imputation. Even when 40% of data were missing, relative bias to true cost-effectiveness curves was less than 16% using the item imputation, but up to 39% for the aggregate imputation. Thus, the imputation strategies could have a significant impact on the cost-effectiveness conclusions when more than 20% of data are missing. The item imputation approach has better precision than the imputation at the aggregate level.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.358 | 0.026 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.006 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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; both teacher heads agree on what is shown here.
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".