Optimal multiwave sampling for regression modeling in two‐phase designs

Chen, Tong; Lumley, Thomas

doi:10.1002/sim.8760

Cited by 23 publications

(52 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, constructing multiphase sampling designs would be a fruitful avenue for future work. (See McIsaac & Cook, 2015, Chen & Lumley, 2020, and Han et al., 2020, for some initial work.) These authors considered designs where a pilot sample could initially be selected from the cohort to obtain information on the validated data that can be used to guide follow‐up sampling waves.…”

Section: Discussionmentioning

confidence: 99%

Improved generalized raking estimators to address dependent covariate and failure‐time outcome error

Shepherd

Lumley

et al. 2021

Biometrical J

Self Cite

View full text Add to dashboard Cite

Biomedical studies that use electronic health records (EHR) data for inference are often subject to bias due to measurement error. The measurement error present in EHR data is typically complex, consisting of errors of unknown functional form in covariates and the outcome, which can be dependent. To address the bias resulting from such errors, generalized raking has recently been proposed as a robust method that yields consistent estimates without the need to model the error structure. We provide rationale for why these previously proposed raking estimators can be expected to be inefficient in failure‐time outcome settings involving misclassification of the event indicator. We propose raking estimators that utilize multiple imputation, to impute either the target variables or auxiliary variables, to improve the efficiency. We also consider outcome‐dependent sampling designs and investigate their impact on the efficiency of the raking estimators, either with or without multiple imputation. We present an extensive numerical study to examine the performance of the proposed estimators across various measurement error settings. We then apply the proposed methods to our motivating setting, in which we seek to analyze HIV outcomes in an observational cohort with EHR data from the Vanderbilt Comprehensive Care Clinic.

show abstract

Section: Discussionmentioning

confidence: 99%

Improved generalized raking estimators to address dependent covariate and failure‐time outcome error

Shepherd

Lumley

et al. 2021

Biometrical J

Self Cite

View full text Add to dashboard Cite

show abstract

“…This research, for example, could build on adaptive strategies that have been proposed for other instances in which key parameters are unknown at the outset, such as that proposed in the context of sample size and power calculations, 36 or that presented for approximating optimal allocation in other settings. 13,29,37,38 Another potential way forward would be to use imputation, following the approaches taken in the analysis of longitudinal data. 24,39 Furthermore, with regard to optimizing for multiple parameters, our simulation study only evaluated the approach in which every parameter is given equal weight (w q = 1 for q = 1, … , p).…”

Section: Discussionmentioning

confidence: 99%

“…To resolve this, we introduce a small rounding threshold, , which can be increased until the sum of the rounded sample sizes is equal to K s . For example, if K s = 40 and (k 00 , k 01 , k 11 , k 10 ) = (20.18, 7.01, 6.49, 6.32), using the standard rounding threshold of 0.5 would yield (k Other approaches have been suggested that directly yield integer solutions, such as the algorithm propose by Wright, 28 and used with good results by Chen and Lumley; 29 we do not consider those here, as our simulations indicate good performance of the continuous allocation strategy combined with the rounding procedure described above.…”

Section: Practical Issuesmentioning

confidence: 99%

Optimal allocation in stratified cluster‐based outcome‐dependent sampling designs

Sauer

Hedt‐Gauthier

Haneuse

2021

Statistics in Medicine

View full text Add to dashboard Cite

In public health research, finite resources often require that decisions be made at the study design stage regarding which individuals to sample for detailed data collection. At the same time, when study units are naturally clustered, as patients are in clinics, it may be preferable to sample clusters rather than the study units, especially when the costs associated with travel between clusters are high. In this setting, aggregated data on the outcome and select covariates are sometimes routinely available through, for example, a country's Health Management Information System. If used wisely, this information can be used to guide decisions regarding which clusters to sample, and potentially obtain gains in efficiency over simple random sampling. In this article, we derive a series of formulas for optimal allocation of resources when a single-stage stratified cluster-based outcome-dependent sampling design is to be used and a marginal mean model is specified to answer the question of interest. Specifically, we consider two settings: (i) when a particular parameter in the mean model is of primary interest; and, (ii) when multiple parameters are of interest. We investigate the finite population performance of the optimal allocation framework through a comprehensive simulation study. Our results show that there are trade-offs that must be considered at the design stage: optimizing for one parameter yields efficiency gains over balanced and simple random sampling, while resulting in losses for the other parameters in the model. Optimizing for all parameters simultaneously yields smaller gains in efficiency, but mitigates the losses for the other parameters in the model.

show abstract

“…Recently, Chen et al. 28 have considered a method for incorporation of prior information into multi-wave sampling in a regression framework. Also, further extension to time-dynamic models that introduce time-varying covariates and their time-dependent effects will also provide flexible tools for the two-phase analysis of a broader class of survival models.…”

Section: Discussionmentioning

confidence: 99%

“…Incorporation of prior information regarding the sampling priorities for certain strata may be an additional way to improve the performance of the adaptive design, particularly in the case of a small phase two sample. Recently, Chen et al 28 have considered a method for incorporation of prior information into multi-wave sampling in a regression framework. Also, further extension to time-dynamic models that introduce time-varying covariates and their time-dependent effects will also provide flexible tools for the two-phase analysis of a broader class of survival models.…”

Section: Discussionmentioning

confidence: 99%

Two-phase analysis and study design for survival models with error-prone exposures

Han

Lumley

Shepherd

et al. 2020

Stat Methods Med Res

Self Cite

View full text Add to dashboard Cite

Increasingly, medical research is dependent on data collected for non-research purposes, such as electronic health records data. Health records data and other large databases can be prone to measurement error in key exposures, and unadjusted analyses of error-prone data can bias study results. Validating a subset of records is a cost-effective way of gaining information on the error structure, which in turn can be used to adjust analyses for this error and improve inference. We extend the mean score method for the two-phase analysis of discrete-time survival models, which uses the unvalidated covariates as auxiliary variables that act as surrogates for the unobserved true exposures. This method relies on a two-phase sampling design and an estimation approach that preserves the consistency of complete case regression parameter estimates in the validated subset, with increased precision leveraged from the auxiliary data. Furthermore, we develop optimal sampling strategies which minimize the variance of the mean score estimator for a target exposure under a fixed cost constraint. We consider the setting where an internal pilot is necessary for the optimal design so that the phase two sample is split into a pilot and an adaptive optimal sample. Through simulations and data example, we evaluate efficiency gains of the mean score estimator using the derived optimal validation design compared to balanced and simple random sampling for the phase two sample. We also empirically explore efficiency gains that the proposed discrete optimal design can provide for the Cox proportional hazards model in the setting of a continuous-time survival outcome.

show abstract

Optimal multiwave sampling for regression modeling in two‐phase designs

Cited by 23 publications

References 28 publications

Improved generalized raking estimators to address dependent covariate and failure‐time outcome error

Improved generalized raking estimators to address dependent covariate and failure‐time outcome error

Optimal allocation in stratified cluster‐based outcome‐dependent sampling designs

Two-phase analysis and study design for survival models with error-prone exposures

Contact Info

Product

Resources

About