Abstract Background Synthetic data hold substantial potential to address practical challenges in epidemiology due to restricted data access and privacy concerns. However, many current methods suffer from limited quality, high computational demands, and complexity for non-experts. Furthermore, common evaluation strategies for synthetic data often fail to directly reflect statistical utility and measure privacy risks sufficiently. Against this background, a critical underexplored question is whether synthetic data can reliably reproduce key findings from epidemiological research while preserving privacy. Methods We propose adversarial random forests (ARF) as an efficient and convenient method for synthesizing tabular epidemiological data. To evaluate its performance, we replicated statistical analyses from six epidemiological publications covering blood pressure, anthropometry, myocardial infarction, accelerometry, loneliness, and diabetes, from the German National Cohort (NAKO Gesundheitsstudie), the Bremen STEMI Registry U45 Study, and the Guelph Family Health Study. We further assessed how dataset dimensionality and variable complexity affect the quality of synthetic data, and contextualized ARF’s performance by comparison with commonly used tabular data synthesizers in terms of utility, privacy, generalization, and runtime. Results Across all replicated studies, results on ARF-generated synthetic data consistently aligned with original findings. Even for datasets with relatively low sample size-to-dimensionality ratios, replication outcomes closely matched the original results across descriptive and inferential analyses. Reduced dimensionality and variable complexity further enhanced synthesis quality. ARF demonstrated favourable performance regarding utility, privacy preservation, and generalization relative to other synthesizers and superior computational efficiency. Conclusions In summary, ARF reliably generates high-quality synthetic data that replicate diverse epidemiological analyses while offering a competitive privacy–utility trade-off.
J. Kapar, Kathrin Günther, L. Vallis et al.· International Journal of Epi...· 1 citation
Phosphorylated-tau (p-tau217) is a promising blood-based biomarker for Alzheimer’s dementia (AD) in clinical settings. However, research from prospective cohort studies is sparse. We measured plasma p-tau217 levels in baseline blood samples of 779 participants in a population-based cohort of older adults followed over 17 years. Associations with AD were assessed and compared to those with previous measurements of p-tau181, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP). Comparisons to the amyloid beta (Aβ) misfolding biomarker were performed in a subgroup analysis. P-tau217, NfL and GFAP showed strong associations with AD risk, especially within the first 9 years of follow-up, outperforming p-tau181. Over the later years of follow-up, the predictive accuracy of p-tau217 was significantly reduced. In contrast, the Aβ misfolding biomarker demonstrated superior performance especially as a preclinical indicator of the risk of AD many years before diagnosis. The combination of p-tau217 with NfL, GFAP, basic demographic and genetic information, as well as the misfolding biomarker yielded an AUC 0.86 for AD diagnoses over the entire 17-year follow-up period. P-tau217 is a promising blood-based biomarker for Alzheimer’s dementia (AD) in the clinical stage. In this prospective cohort study, it showed better performance in the prodromal than in the preclinical stage up to 17 years before diagnosis. In a subgroup analysis, the Aß misfolding biomarker showed higher discriminative ability in the preclinical stage compared to p-tau217. A prediction model including p-tau217, NfL, GFAP, Aß misfolding as well as basic demographic and genetic information reached an AUC of 0.86 for AD over the entire follow-up period. In a subgroup analysis, the Aß misfolding biomarker showed higher discriminative ability in the preclinical stage compared to p-tau217. A prediction model including p-tau217, NfL, GFAP, Aß misfolding as well as basic demographic and genetic information reached an AUC of 0.86 for AD over the entire follow-up period. P-tau217 is a promising blood-based biomarker for Alzheimer’s dementia (AD) in the clinical stage. In this prospective cohort study, it showed better performance in the prodromal than in the preclinical stage up to 17 years before diagnosis.
K. Trares, D. Duman, L. Beyer et al.· EMBO Molecular Medicine· 0 citations
