Interep: An R Package for High-Dimensional Interaction Analysis of the Repeated Measurement Data

Using bilateral data in controls and patients with bilateral and unilateral pathology requires increased scrutiny

In many aspects of human research, capturing multiple measures from the same participant is common due to the symmetric nature of the human body (e.g., two eyes, ten fingers, two legs, etc.). This has established a concerning paradox in biomedical and clinical research. When the same condition exist bilaterally (controls or bilateral pathology), researchers often blindly include both (or multiple) measures into the statistical analysis. This assumes that measures between the two sides are statistically independent (uncorrelated). However, there are certain inherent factors within an individual (e.g., age, sex, physical activity, gait pattern, tissue characteristics, hormonal status, pain thresholds, etc.) that would point to a statistical dependence between bilateral measures. Conversely, in unilateral pathology, it is common practice to use the contralateral side as the comparator. This assumes the exact opposite, that sans pathology, bilateral measures are perfectly correlated without bias. Both of these assumptions can lead to errors in the study conclusions. Few studies have explored the statistical dependence between multiple measures from the same participant. Thus, the purpose of this perspective is to explore the statistical considerations associated with analyzing multiple measures from the same participant and provide recommendations for navigating the use of multiple, non-temporal, data points from the same participant. To give context for these recommendations, an example dataset involving patellofemoral kinematics is provided. Due to the prevalent use of bilateral data in the current literature and the resulting potential for invalid study conclusions, we recommend that future research use caution when using multiple measures from the same participant and apply proper statistical analysis (e.g., generalized estimating equations) when these measures are not independent. If the contralateral limb is used as a comparator in unilateral pathology, strong evidence must exist that the underlying pathology has not altered the measures of interest in this contralateral limb.

Springer: An R package for bi-level variable selection of high-dimensional longitudinal data

In high-dimensional data analysis, the bi-level (or the sparse group) variable selection can simultaneously conduct penalization on the group level and within groups, which has been developed for continuous, binary, and survival responses in the literature. Zhou et al. (2022) (PMID: 35766061) has further extended it under the longitudinal response by proposing a quadratic inference function-based penalization method in gene-environment interaction studies. This study introduces "springer," an R package implementing the bi-level variable selection within the QIF framework developed in Zhou et al. (2022). In addition, R package "springer" has also implemented the generalized estimating equation-based sparse group penalization method. Alternative methods focusing only on the group level or individual level have also been provided by the package. In this study, we have systematically introduced the longitudinal penalization methods implemented in the "springer" package. We demonstrate the usage of the core and supporting functions, which is followed by the numerical examples and discussions. R package "springer" is available at https://cran.r-project.org/package=springer.