Social connection is increasingly recognized as a vital component of population health. It includes the structure of social networks, the support that flows through relationships, and the quality of interactions that contribute to well-being. Research across multiple disciplines has established that limited or poor social connection is linked to higher risks of premature mortality, mental health problems, cardiovascular disease, cognitive decline, lower educational attainment, and reduced economic participation. These findings highlight the need for rigorous and comparable metrics that can illuminate the distribution, determinants, and consequences of social disconnection at scale.

Despite a growing interest in social connection, population level measurement remains uneven. Studies often rely on single items, short scales, or constructs that capture only one aspect of the broader social landscape. Loneliness, social isolation, and social support are frequently treated as interchangeable, even though each reflects a different dimension of social life and may influence health through distinct pathways. In many settings, there is no routine surveillance of social connection. As a result, policymakers and practitioners lack the data needed to design effective interventions or monitor changes over time.

Conceptual clarity is a prerequisite for measurement. Social connection is multidimensional and includes the size and diversity of social networks, the availability of emotional or instrumental support, and the positive or negative qualities of relationships. Social isolation is the objective absence of social roles or interactions. Loneliness is the subjective experience of wanting more or better connection than one currently has. These concepts intersect but are not synonymous. Accurate assessment requires tools that reflect each dimension and that can be adapted or validated across cultures, languages, and life stages.

Research from low and middle income countries is particularly limited even though many of these settings experience social conditions that may heighten vulnerability to social disconnection. Large and rapid demographic transitions, urban migration, changing family structures, economic insecurity, and exposure to conflict or displacement can all shape patterns of social connection. Yet most validated instruments originate in high income countries, and many have not been adapted or tested for linguistic, cultural, and contextual relevance in LMIC populations.

The need for strong metrics extends beyond conceptual work. The use of digital technology, demographic change, shifting household structures, migration, and urbanization all influence how social connections form and function. These forces create new opportunities for measurement, such as digital interaction data or innovative mixed-method approaches, but they also introduce new challenges. Measures grounded in one cultural or linguistic context may not translate easily to another. Children, adolescents, older adults, and marginalized groups may require tailored instruments that reflect their social realities. Community and policy interventions also need evaluation frameworks that capture both intended and unintended effects on connection.

This special issue invites research that advances the science of measuring social connection for population health. Relevant contributions include but are not limited to:

1. New conceptual models and frameworks for assessing social connection at scale

2. Validation or adaptation of measurement instruments across cultures and population groups

3. Methods for distinguishing between loneliness, social isolation, and social support in surveillance systems

4. Epidemiologic analyses that examine levels, patterns, and trends in social connection

5. Studies linking social connection metrics with health, economic, or educational outcomes

6. Approaches to integrating social connection into existing population health surveys

7. Measurement of social connection in digital and hybrid social environments

8. Evaluation metrics for policies, community initiatives, and individual level interventions

9. Proposals for harmonized indicators that support international comparisons

All submissions in this collection undergo the journal’s standard peer review process. Similarly, all manuscripts authored by a Guest Editor(s) will be handled by the Editor-in-Chief. As an open access publication, this journal levies an article processing fee (details here). We recognize that many key stakeholders may not have access to such resources and are committed to supporting participation in this issue wherever resources are a barrier. For more information about what support may be available, please visit OA funding and support, or email OAfundingpolicy@springernature.com or the Editor-in-Chief.

Computational approaches are becoming indispensable in the field of translational pharmacology and systems medicine. In particular, the integration of computational modelling and network medicine is reshaping our understanding of drug toxicity, safety profiling, and clinical pharmacovigilance. With the rapid accumulation of multi-omics, electronic health records, and real-world pharmacovigilance data, there is an urgent need for advanced computational frameworks to identify hidden drug–target–phenotype associations and elucidate mechanisms underlying adverse drug reactions (ADRs) and toxicities. Drug toxicology and pharmacovigilance are key components of translational medicine, bridging molecular mechanisms with patient outcomes. However, predicting and preventing drug-induced toxicities remain challenging due to the complexity of biological systems and individual heterogeneity. Network-based and machine learning methods, such as graph neural networks, causal inference, and mechanistic modelling, are providing new opportunities to reveal system-level toxicity pathways, optimize drug safety assessment, and enhance early signal detection in clinical practice.

This collection aims to gather latest research and comprehensive reviews that explore computational, network-based, and AI-driven strategies in drug toxicology and pharmacovigilance. We seek contributions that demonstrate how integrative modelling, network pharmacology, and multi-scale data analysis can advance the prediction, understanding, and management of drug-related adverse effects, ultimately contributing to safer and more effective therapeutics.

Potential topics include, but are not limited to:

- Computational modelling for drug toxicity prediction

- Network medicine approaches to drug safety and toxicity mechanisms

-Machine learning and AI in pharmacovigilance signal detection

-Integrative modelling of drug–target–pathway interactions

-Systems pharmacology and multi-omics network analysis in toxicology

-Network-based biomarker discovery for adverse drug reactions

-Data-driven prediction of idiosyncratic and immune-related toxicities

-Translational applications of network and computational models in clinical drug safety assessment

All submissions in this collection undergo the journal’s standard peer review process. Similarly, all manuscripts authored by a Guest Editor(s) will be handled by the Editor-in-Chief. As an open access publication, this journal levies an article processing fee (details here). We recognize that many key stakeholders may not have access to such resources and are committed to supporting participation in this issue wherever resources are a barrier. For more information about what support may be available, please visit OA funding and support, or email OAfundingpolicy@springernature.com or the Editor-in-Chief.