Dataset Information

Statistical biases due to anonymization evaluated in an open clinical dataset from COVID-19 patients.

ABSTRACT: Anonymization has the potential to foster the sharing of medical data. State-of-the-art methods use mathematical models to modify data to reduce privacy risks. However, the degree of protection must be balanced against the impact on statistical properties. We studied an extreme case of this trade-off: the statistical validity of an open medical dataset based on the German National Pandemic Cohort Network (NAPKON), which was prepared for publication using a strong anonymization procedure. Descriptive statistics and results of regression analyses were compared before and after anonymization of multiple variants of the original dataset. Despite significant differences in value distributions, the statistical bias was found to be small in all cases. In the regression analyses, the median absolute deviations of the estimated adjusted odds ratios for different sample sizes ranged from 0.01 [minimum = 0, maximum = 0.58] to 0.52 [minimum = 0.25, maximum = 0.91]. Disproportionate impact on the statistical properties of data is a common argument against the use of anonymization. Our analysis demonstrates that anonymization can actually preserve validity of statistical results in relatively low-dimensional data.

SUBMITTER: Koll CEM

PROVIDER: S-EPMC9769467 | biostudies-literature | 2022 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Statistical biases due to anonymization evaluated in an open clinical dataset from COVID-19 patients.

Koll Carolin E M CEM Hopff Sina M SM Meurers Thierry T Lee Chin Huang CH Kohls Mirjam M Stellbrink Christoph C Thibeault Charlotte C Reinke Lennart L Steinbrecher Sarah S Schreiber Stefan S Mitrov Lazar L Frank Sandra S Miljukov Olga O Erber Johanna J Hellmuth Johannes C JC Reese Jens-Peter JP Steinbeis Fridolin F Bahmer Thomas T Hagen Marina M Meybohm Patrick P Hansch Stefan S Vadász István I Krist Lilian L Jiru-Hillmann Steffi S Prasser Fabian F Vehreschild Jörg Janne JJ

Scientific data 20221221 1

Anonymization has the potential to foster the sharing of medical data. State-of-the-art methods use mathematical models to modify data to reduce privacy risks. However, the degree of protection must be balanced against the impact on statistical properties. We studied an extreme case of this trade-off: the statistical validity of an open medical dataset based on the German National Pandemic Cohort Network (NAPKON), which was prepared for publication using a strong anonymization procedure. Descrip ...[more]

PMID: 36543828

Similar Datasets

Project description:The COVID-19 pandemic has significantly influenced various aspects of society, including environmental factors such as methane emissions. This study investigates the changes in methane concentrations in Seoul, South Korea, from 2019 to 2023, using TROPOMI satellite data and rigorous statistical analyses. The normality of the sample data is first assessed using the Shapiro-Wilk (S-W) and Kolmogorov-Smirnov (K-S) tests, indicating that the data can be considered to come from a normal distribution. The S-W test demonstrated superior discriminative power (highest statistical power: 0.8668) compared to the K-S test (highest statistical power: 0.4002), confirming the validity of parametric tests for our data. The S-W test shows better discriminative power than the K-S test in terms of sensitivity to departures from normality, particularly for small sample sizes. Based on this confirmation, parametric tests such as analysis of variance (ANOVA) and post-hoc tests (Bonferroni correction, Tukey's HSD, Scheffe's method) are employed to identify significant differences in methane levels across different years. The ANOVA results show a statistically significant difference in methane concentrations across years (p-value: 2.02×10-13 , F-value: 26.572). Post-hoc analyses reveal no significant difference in methane concentrations between 2019 and 2020 (p-values: Bonferroni - 0.1045, Tukey's HSD - 0.397, Scheffe's - 0.1045), and no significant difference between 2020 and 2021 (p-values: Bonferroni - 0.917, Tukey's HSD - 0.840, Scheffe's - 0.917). However, a significant increase in methane levels is observed from 2022 to 2023 (p-values: Bonferroni - 0.0001, Tukey's HSD - 0.0002, Scheffe's - 0.0001), correlating with the "new normal" policy implemented in South Korea starting in November 2021 and effectively from the beginning of 2022. This suggests that changes in industrial activities and transportation patterns due to the "new normal" have contributed to higher methane emissions. Student's t-test and Welch's t-test were used to validate the ANOVA results. Permutation tests showed no significant difference between 2019 and 2020 (test statistic: -0.0096, p-values: 0.1191 for Student's and 0.1156 for Welch's). However, a significant difference was found between 2022 and 2023 (test statistic: -0.0172, p-value: 0.0001), confirming ANOVA results that indicated increased methane levels post-pandemic. This study provides a robust quantitative assessment of the pandemic's impact on methane levels and sets a methodological statistical approach for future research in the environmental research community.

Project description:On the behest of the Office of Science and Technology Policy in the White House, six institutions, including ours, have created an open research dataset called COVID-19 Research Dataset (CORD-19) to facilitate the development of question-answering systems that can assist researchers in finding relevant research on COVID-19. As of May 27, 2020, CORD-19 includes more than 100,000 open access publications from major publishers and PubMed as well as preprint articles deposited into medRxiv, bioRxiv, and arXiv. Recent years, however, have also seen question-answering and other machine learning systems exhibit harmful behaviors to humans due to biases in the training data. It is imperative and only ethical for modern scientists to be vigilant in inspecting and be prepared to mitigate the potential biases when working with any datasets. This article describes a framework to examine biases in scientific document collections like CORD-19 by comparing their properties with those derived from the citation behaviors of the entire scientific community. In total, three expanded sets are created for the analyses: 1) the enclosure set CORD-19E composed of CORD-19 articles and their references and citations, mirroring the methodology used in the renowned "A Century of Physics" analysis; 2) the full closure graph CORD-19C that recursively includes references starting with CORD-19; and 3) the inflection closure CORD-19I, that is, a much smaller subset of CORD-19C but already appropriate for statistical analysis based on the theory of the scale-free nature of the citation network. Taken together, all these expanded datasets show much smoother trends when used to analyze global COVID-19 research. The results suggest that while CORD-19 exhibits a strong tilt toward recent and topically focused articles, the knowledge being explored to attack the pandemic encompasses a much longer time span and is very interdisciplinary. A question-answering system with such expanded scope of knowledge may perform better in understanding the literature and answering related questions. However, while CORD-19 appears to have topical coverage biases compared to the expanded sets, the collaboration patterns, especially in terms of team sizes and geographical distributions, are captured very well already in CORD-19 as the raw statistics and trends agree with those from larger datasets.

Dataset Information

Statistical biases due to anonymization evaluated in an open clinical dataset from COVID-19 patients.

Publications

Statistical biases due to anonymization evaluated in an open clinical dataset from COVID-19 patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets