MétaCan
Menu
Back to cohort
Record W2607857384 · doi:10.1186/s12882-017-0552-3

Diagnosis and outcomes of acute kidney injury using surrogate and imputation methods for missing preadmission creatinine values

2017· article· en· W2607857384 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueBMC Nephrology · 2017
Typearticle
Languageen
FieldMedicine
TopicAcute Kidney Injury Research
Canadian institutionsUniversité de MontréalHôpital du Sacré-Cœur de Montréal
FundersKidney Foundation of CanadaAmerican Society of Nephrology
KeywordsMedicineCreatinineAcute kidney injuryNephrologyInternal medicineImputation (statistics)Missing dataIntensive care medicineStatistics

Abstract

fetched live from OpenAlex

Missing preadmission serum creatinine (SCr) values are a common obstacle to assess acute kidney injury (AKI) diagnosis and outcomes. The Kidney Disease Improving Global Outcomes (KDIGO) guidelines suggest using a SCr computed from the Modification of Diet in Renal Disease (MDRD) with an estimated glomerular filtration rate of 75 ml/min/1.73 m 2 . We aimed to identify the best surrogate method for baseline SCr to assess AKI diagnosis and outcomes. We compared the use of 1) first SCr at hospital admission 2) minimal SCr over 2 weeks after intensive care unit admission 3) MDRD computed SCr and 4) Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) computed SCr to assess AKI diagnosis and outcomes. We then performed multilinear regression models to predict preadmission SCr and imputation strategies to assess AKI diagnosis. Our one-year retrospective cohort study included 1001 critically ill adults; 498 of them had preadmission SCr values. In these patients, AKI incidence was 25.1% using preadmission SCr. First SCr had the best agreement for AKI diagnosis (22.5%; kappa = 0.90) and staging (kappa = 0.81). MDRD, CKD-EPI and minimal SCr overestimated AKI diagnosis (26.7%, 27.1% and 43.2%;kappa = 0.86, 0.86 and 0.60, respectively). However, MDRD and CKD-EPI computed SCr had a better sensitivity than first SCr for AKI (93% and 94% vs. 87%). Eighty-eight percent of patients experienced renal recovery at least 3 months after hospital discharge. All methods except the first SCr significantly underestimated the percentage of renal recovery. In a multivariate model, age, male gender, hypertension, heart failure, undergoing surgery and log first SCr best predicted preadmission SCr (adjusted R 2 = 0.56). Imputation methods with first SCr increased AKI incidence to 23.9% (kappa = 0.92) but not with MDRD computed SCr (26.7%;kappa = 0.89). In our cohort, first SCr performed better for AKI diagnosis and staging, as well as for renal recovery after hospital discharge than MDRD, CKD-EPI or minimal SCr. However, MDRD SCr and CKD-EPI SCr improved AKI diagnosis sensitivity. Imputation methods minimally increased agreement for AKI diagnosis.

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.

Full frame distilled prediction

Teacher imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.005
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.671
Threshold uncertainty score0.557

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.005
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.086
GPT teacher head0.484
Teacher spread0.398 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it