In several biomedical studies, one or more exposures of interest may be subject to non-random missingness due to the failure of the measurement assay at levels below its limit of detection. This issue is commonly encountered in studies of the metabolome employing tandem mass spectrometry-based technologies. Due to a large number of metabolites measured in these studies, preserving statistical power is of utmost interest. In these settings, the missing indicator model minimizes loss of information and thus provides an attractive alternative to the oft-used complete case analysis and other imputation approaches. In this paper, we evaluate the small sample properties of the missing indicator approach in logistic and conditional logistic regression models. We show that under a variety of settings, the missing indicator approach outperforms complete case analysis and a variety of imputation approaches with regard to bias, mean squared error and power. We compare the results from the missing indicator model to that from a complete case analysis using data from a Cardiovascular Disease Biomarker study employing metabolomic technologies.