Case‐control data analysis for randomly pooled biomarkers

Abstract

Pooled study designs, where individual biospecimens are combined prior to measurement via a laboratory assay, can reduce lab costs while maintaining statistical efficiency. Analysis of the resulting pooled measurements, however, often requires specialized techniques. Existing methods can effectively estimate the relation between a binary outcome and a continuous pooled exposure when pools are matched on disease status. When pools are of mixed disease status, however, the existing methods may not be applicable. By exploiting characteristics of the gamma distribution, we propose a flexible method for estimating odds ratios from pooled measurements of mixed and matched status. We use simulation studies to compare consistency and efficiency of risk effect estimates from our proposed methods to existing methods. We then demonstrate the efficacy of our method applied to an analysis of pregnancy outcomes and pooled cytokine concentrations. Our proposed approach contributes to the toolkit of available methods for analyzing odds ratios of a pooled exposure, without restricting pools to be matched on a specific outcome.

Exposure to Persistent Organic Pollutants and Birth Characteristics

“…Methods for analysis of pooled data have been developed and are further described in the statistical methods section. 34 As previously described 26,30 , DDE had a limit of detection of 0.02 ng/ml, PBDE-47 of 0.003 ng/ml, PCBs 138, 153, and 180 of 0.002 ng/ml, PCBs 18 and 28 of 0.004 ng/ml, and all remaining PCBs (8, 15, 52, 70, 95, 101) of 0.008 ng/ml. Further details of the analytical methods have been reported elsewhere.…”

“…38 The gamma distribution is similar in shape to lognormal but has been assumed in the pooling literature because of the convenient property where if the individual concentrations follow a gamma distribution, then the pools also follow a gamma distribution which lognormal lacks. 34,[38][39][40][41][42] After employing multiple imputation (100 imputations) to obtain individual level pollutant exposures from pooled samples, we used linear and logistic regression to examine the associations between a 0.1-ng/mL increase in each POP and continuous outcomes and binary birth characteristics, respectively. 34,40 Values of pollutants that were below the limits of quantification were replaced by a near-zero constant, as consistent with EPA guidance.…”

“…34,[38][39][40][41][42] After employing multiple imputation (100 imputations) to obtain individual level pollutant exposures from pooled samples, we used linear and logistic regression to examine the associations between a 0.1-ng/mL increase in each POP and continuous outcomes and binary birth characteristics, respectively. 34,40 Values of pollutants that were below the limits of quantification were replaced by a near-zero constant, as consistent with EPA guidance. Models failed to converge for PCB-180 (likely due to higher levels of near-zero data for this pollutant) and the pollutant was dropped from the analysis.…”

“…symmetric or skewed). 34,[38][39][40]42 Of note, most existing methods require pools be formed based on outcome status. However, we utilized and adapted methods to analyze our pools of mixed outcome and covariate status (wherein members of a pool may have different outcomes statuses).…”

Logistic regression with a continuous exposure measured in pools and subject to errors

“…The discriminant function approach could still be used, however; fitting it separately for cases and controls would produce two odds ratio estimates, whereas fitting it with case status as a covariate would be akin to adjusting for the original outcome variable. Other methods compatible with heterogeneous pools could also be used, although most cannot correct for errors in the pooled biomarker.…”