Dataset Information

On the Use of Regression Calibration in a Complex Sampling Design With Application to the Hispanic Community Health Study/Study of Latinos.

ABSTRACT: Regression calibration is the most widely used method to adjust regression parameter estimates for covariate measurement error. Yet its application in the context of a complex sampling design, for which the common bootstrap variance estimator can be less straightforward, has been less studied. We propose 2 variance estimators for a multistage probability-based sampling design, a parametric and a resampling-based multiple imputation approach, where a latent mean exposure needed for regression calibration is the target of imputation. This work was motivated by the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) data from 2008 to 2011, for which relationships between several outcomes and diet, an error-prone self-reported exposure, are of interest. We assessed the relative performance of these variance estimation strategies in an extensive simulation study built on the HCHS/SOL data. We further illustrate the proposed estimators with an analysis of the cross-sectional association of dietary sodium intake with hypertension-related outcomes in a subsample of the HCHS/SOL cohort. We have provided guidelines for the application of regression models with regression-calibrated exposures. Practical considerations for implementation of these 2 variance estimators in the setting of a large multicenter study are also discussed. Code to replicate the presented results is available online.

SUBMITTER: Baldoni PL

PROVIDER: S-EPMC8245895 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Rationale & objectiveRecent studies suggest that periodontal disease may be associated with incident chronic kidney disease (CKD). However, studies have focused on older populations, and US Hispanics/Latinos were not well represented.Study designObservational cohort.Setting & participantsWe analyzed data from the Hispanic Community Health Study/Study of Latinos who completed a baseline visit with a periodontal examination and a follow-up visit, and did not have CKD at baseline.PredictorsPredictors included ≥30% of sites with clinical attachment loss ≥3 mm, ≥30% of sites with probing depth ≥4 mm, percentage of sites with bleeding on probing, and absence of functional dentition (<21 permanent teeth present).OutcomesOutcomes were incident low estimated glomerular filtration rate (eGFR) (eGFR <60 mL/min/1.73 m2 and decline in eGFR ≥1 mL/min/year); incident albuminuria (urine albumin:creatinine ratio [ACR] ≥30 mg/g); and change in eGFR and ACR.Analytic approachPoisson and linear regression.ResultsFor the sample (n = 7.732), baseline mean age was 41.5 years, 45.2% were male, 11.7% had ≥30% of sites with clinical attachment loss ≥3 mm, 5.1% had ≥30% of sites with probing depth ≥4 mm, 30.7% had ≥50% of sites with bleeding on probing, and 16.2% had absent functional dentition. During a median follow-up of 5.9 years, 149 patients developed low eGFR and 415 patients developed albuminuria. On multivariable analysis, presence versus absence of ≥30% of sites with probing depth ≥4 mm and absence of functional dentition were each associated with increased risk for incident low eGFR (incident density ratio, 2.31; 95% CI, 1.14-4.65 and 1.65, 95% CI, 1.01-2.70, respectively). None of the other predictors were associated with outcomes.LimitationsOnly a single kidney function follow-up measure.ConclusionsIn this cohort of US Hispanics/Latinos, we found that select measures of periodontal disease were associated with incident low eGFR. Future work is needed to assess whether the treatment of periodontal disease may prevent CKD.

Dataset Information

On the Use of Regression Calibration in a Complex Sampling Design With Application to the Hispanic Community Health Study/Study of Latinos.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets