A Comparison of Methods to Adjust Risk Models for Measurement Error in Dietary Exposure
Bibliographic record
Abstract
To compare approaches to adjust for measurement error when exploring the relationship between the sodium-to-potassium ratio and hypertension and blood pressure. Using National Health and Nutrition Examination Survey (NHANES) data from 2005–2010, Zhang et al. (2013) examined relationships between the sodium-to-potassium ratio (Na:K) and hypertension and blood pressure. They used the Iowa State University (ISU) Method to account for measurement error assuming the dietary intakes assessed with up to two 24-hour recalls (24 HR) per person were unbiased for long-term (usual) intake. We replicated their analytic dataset (n = 10,467 adults ³20 y). Systolic and diastolic blood pressure values were calculated as the mean of three readings; a binary variable for hypertension status was derived from blood pressure values and self-reported health information. Logistic (hypertension) and linear (blood pressure) models were adjusted for age, sex, and race/ethnicity. The primary exposure variable, Na:K, was obtained from calibration models that adjusted for random variation in the 24 HR. The National Cancer Institute (NCI) Method used regression calibration from a joint model - fit with two methods, maximum likelihood and Markov Chain Monte Carlo (MCMC) - for sodium and potassium. A third model used the ISU Method to univariately model each component. All calibration models adjusted for the same covariates as the regression models plus nuisance effects. Standard errors for estimates and differences (e.g., MCMC - ISU) were computed using Balanced Repeated Replication to account for NHANES’ survey design. Both NCI implementations gave nearly equivalent results that sometimes differed from the ISU results. A one-unit increase in Na:K was associated with a mean systolic blood pressure increase of 2.82 mmHg, se .57 (ISU) vs. 3.87 mmHg, se .87 (MCMC) (t-test pdiff = 0.005), and a mean diastolic blood pressure increase of 0.80 mmHg, se .43 (ISU) vs. 0.61, se .68 (MCMC) (pdiff = 0.495). A one-unit increase in Na:K was associated with an increase of 0.41, se .11 (ISU) vs. 0.63, se .18 (MCMC) (pdiff = 0.008) in the log odds of hypertension. The choice of modeling approaches for 24 HRs may affect estimated error-corrected relationships between health outcomes and dietary intake. Vanier Canada Graduate Scholarship.
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 | 0.352 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.010 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.006 | 0.005 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.005 | 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".