Synthetic data generation has emerged as a powerful strategy for sharing sensitive datasets without compromising privacy. However, despite the use of advanced generative models, synthetic data often retains patterns from the original dataset, posing potential privacy risks. In this paper, we propose PrivGen, a human-in-the-loop framework that enhances privacy in synthetic data generation by integrating expert-guided preprocessing. PrivGen introduces robust outlier detection and data cleaning steps prior to model training, effectively removing high-risk information while preserving data utility. We evaluate PrivGen across nine state-of-the-art generative models and three public datasets, using a suite of seventeen evaluation metrics covering statistical fidelity, sanity checks, and privacy. Results demonstrate that PrivGen consistently improves privacy protection, especially in sparse datasets, without severely compromising data quality. These findings underscore the importance of expert involvement and preprocessing in achieving both practical utility and enhanced privacy guarantees in synthetic data generation.

We present a domain-specific instance of the Dataverse Project tailored to support Age-it project. The platform addresses the growing need to enhance the value of information generated by scientific research, especially within complex, multidisciplinary ecosystems. It provides a centralized, open-access infrastructure to preserve, cite, and disseminate diverse research outputs produced by over 700 researchers engaged in the Age-It program. Organized into 10 thematic spokes and multiple Work Packages (WPs), the system enables granular curation of datasets, indicators, reports, images, and code repositories. The platform integrates advanced technologies and user-friendly applications to facilitate data discoverability, access, and reuse. Registered users benefit from full-text search and multicriteria filtering, while WP and spoke administrators manage and update content. This initiative reinforces the FAIR principles (Findable, Accessible, Interoperable, Reusable) and offers a replicable model for other large-scale interdisciplinary research infrastructures.

With global ageing accelerating and the World Health Organization defining healthy ageing (HA) as the sustained development of functional ability in later life, robust measurement frameworks are critical for effective policy design. Building on the systematic review by Piriu et al. (2025), which introduced the PIETHA framework – a multidimensional model organizing HA into three domains (intrinsic capacity, functional ability, and environment) nesting 15 sub-domains and 84 themes – this contribution outlines recent advances in operationalising HA. Testing the validity of the proposed framework, we develop and validate a Composite Healthy Ageing and Resilient Longevity Index (CHARLI) using longitudinal data from the German Socio-Economic Panel (GSOEP) and other longitudinal national and international studies. Exhibiting a structure of 40 rigorously selected variables – validated via principal component analysis, Lasso regression and structural equation modelling – CHARLI demonstrates strong construct, empirical and predictive validity, with significant associations with subjective well‐being, reduced chronic disease risk and increased mental health score. However, an analysis of international studies (e.g. HRS, ELSA, SHARE, CHARLS) reveals that Italy is lagging behind in terms of data interoperability and standardised measurement instruments. Our policy-oriented discussion emphasizes that integrated, longitudinal data – combined with harmonised operational frameworks – are essential for developing an Italian healthy ageing index. This would not only enhance cross-national policy comparability but also enable tailored interventions across Italian territories. Investment in methodological standardization, as suggested by the categorization of measurement instruments (functional/physiological, self-reported, and environmental), is a priority. This research provides both the rationale and quantitative evidence to advocate for a systematic, interoperable approach to measuring and monitoring HA in Italy.