EP20.10: A clinical prediction model to estimate the risk of near‐term perinatal death in a low‐resource setting
Bibliographic record
Abstract
As part of our efforts to develop screening strategies that effectively identify vulnerable fetuses at risk of poor birth outcomes in high-burden settings, we examined the performance of a multivariable model in predicting perinatal death in women undergoing routine antenatal care in Uganda. Prospective cohort study of women with non-anomalous singleton pregnancies attending a rural Hospital in Uganda between 2018 and 2020. All participants underwent an early dating scan followed by detailed growth and Doppler evaluation once between 32 and 40 weeks. Missing data were imputed and multivariable binary logistic regression used to develop prediction models. We reported predictive performance of the model using measures of discrimination (area under the receiver-operating characteristics curve [AUC]) and calibration (slope and intercept). We included 995 pregnancies and there were 31 (3.1%) perinatal deaths of which 18 (1.8%) were stillbirths. In a model combining maternal characteristics with middle cerebral artery pulsatility index (PI), the AUC for predicting perinatal death was 0.78 (95% CI: 0.67–0.87); it was similar for cerebroplacental ratio (0.78, 0.65–0.87). A bootstrap-corrected AUC was 0.71, with a slope of 0.70. Uterine and umbilical artery PIs had minimal impact on the prediction of perinatal death in this near-term cohort. Near-term perinatal death in a low-resource obstetric population is best predicted by combining maternal characteristics and fetal cerebral Doppler. Even so, predictive performance is only moderate and addition of novel biochemical and maternal hemodynamic markers needs to be investigated in future studies. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
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 | 0.005 | 0.010 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".