B-149 Establishing Pediatric Reference Intervals in a US Population using Refine R
Bibliographic record
Abstract
Abstract Background Establishing reference intervals (RIs) is essential for accurately interpreting laboratory test results, as they help define the range of values expected in a healthy population. There are two primary methods for establishing RIs: the direct method, which involves prospectively recruiting healthy individuals, and the indirect method, which uses retrospective data from a presumed healthy population. Established studies such as CALIPER, a direct RI-derivation study conducted in Canada, aimed to address gaps in pediatric reference intervals using thousands of prospectively recruited healthy pediatric patients. However, such studies are very costly to perform and their RIs may not translate to US hospitals serving different or more diverse populations. In this study, our objective was to establish RIs using an indirect approach using RefineR and to compare our results to those of CALIPER RIs. RefineR uses advanced algorithms to adjust for biases and variables such as age, sex, and other factors, ensuring that RIs are more context-specific and accurate for clinical decision-making. A secondary objective was to also compare our findings to RIs derived per CLSI guidelines (non-parametric central 95%, with Tukey outlier elimination). Methods To evaluate, we retrieved data from 5,064 unique patients under 19 years of age who visited our outpatient clinic between 1/1/23 and 12/31/24. These data were based on results obtained on the Roche Cobas instruments for a comprehensive metabolite panel (CMP). We compared the CALIPER-derived RIs with those generated using RefineR and the non-parametric RI (NPRI) derived per CLSI guideline EP28-A3c. Results There was good agreement between CALIPER, Refine R and NPRI when the distribution of results were Gaussian, as was the case for sodium, calcium, albumin, BUN, total bilirubin, and CO2. However, discrepancies were observed for other tests. For example, our ALP levels were higher than those of CALIPER RIs for children aged 1-<10 years (CALIPER: 142-335 U/L, NPRI: 123-397 U/L, RefineR: 137-381 U/L). This discrepancy may be explained by factors such as ethnicity, geographic location, and socioeconomic status, which can all influence ALP levels, making comparisons challenging. For total protein, our population*s levels were approximately 0.5 g/dL higher on the upper end (NPRI: 5.8-7.8 g/dL, Refine R: 6.1-7.8 g/dL) compared to CALIPER RIs (5.9-7.3 g/dL). This difference is likely due to the use of both serum and plasma samples in our analysis, while CALIPER only used serum, which is known to have lower total protein. We also observed higher ALT levels in our population for the 13-19 age group (CLSI: 9-45 U/L, RefineR: 9-41 U/L) compared to CALIPER RIs (12-27 U/L). This difference may be due to unaccounted differences in important variables, like body mass index or alcohol consumption, affecting each population. Conclusion When implementing pediatric RIs, it is crucial to ensure alignment with the correct methodology and preanalytical factors (such as sample type). Discrepancies may arise if variations in patient populations or preanalytical factors are not carefully considered. It is appropriate to adopt CALIPER RIs for US pediatric population for most tests in a CMP, but not all.
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 | 0.010 | 0.027 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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; a candidate call from one teacher head, 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".