Much of healthcare relies on statistics: with one in every 13 women expected to develop breast cancer in their lifetime, the Singapore Ministry of Health (MOH) recommends that women over 50 years of age undergo mammograms once every two years. Such population-based health screenings enable better predictions of disease risk or earlier detection of its onset, helping to prevent its clinical progression and mitigate worse outcomes.
Averages don’t always tell the full story, however. The interactions between each person’s unique biology and the spaces they live in can critically shift how disease manifests in two different people, affecting the nature and extent of their symptoms, as well as their responses to treatment.
As such, the framework of population health has evolved to also incorporate principles of precision medicine, tapping into a range of health information—genetics, medical records, behavioural trends, environmental factors and more—to develop what’s known today as Precision Preventive Population Health (P3H).
“P3H is about the integration of multimodal data to understand the health risk statuses of individuals at a population level,” explained Meijia Ng, Director of the AI for Health and Preparedness Department at A*STAR’s Biomedical Research Council (BMRC).
“The goal is to target high-risk individuals and subgroups based on their unique health profiles and tailor early interventions accordingly, as opposed to only following broad-based guidelines.”
Ng noted that a precise approach to breast cancer screening, for example, would go beyond age-based guidelines to account for an individual’s family history, the presence of cancer-linked variant genes, and other factors on their risk profile. This would then guide a tailored screening routine or initiation of prophylactic measures.
To make an impact in policy and practice, A*STAR’s P3H arm is advancing strong interdisciplinary linkages among fundamental biological studies, multimodal data collection and integration, artificial intelligence (AI), industrial innovation, and clinical translation.
Cohorts of and for change
Within Singapore, child and maternal health is a focus area for P3H initiatives, especially in light of the reported rising average age of first-time mothers and associated higher risks of pregnancy-related complications. Since 2009, the Growing Up in Singapore Towards healthy Outcomes (GUSTO) birth cohort study has been following over 1,000 then-pregnant women and their children, building up a rich multidimensional dataset that includes lifestyle factors, biological samples, longitudinal body and brain imaging, and clinical assessments from pre-birth onward.
“GUSTO’s primary goal is to understand how early-life exposures shape developmental trajectories, metabolic health, neurodevelopment, and the later risks of noncommunicable diseases (NCDs) in the Asian context,” said Johan Eriksson, Executive Director of the A*STAR Institute for Human Development and Potential (A*STAR IHDP) and GUSTO Principal Investigator.
A major collaboration between A*STAR IHDP, the National University of Singapore (NUS), KK Women’s and Children’s Hospital and other academic partners, GUSTO is driving preventive health policy shifts. Eriksson highlighted how prior to the study, only pregnant women considered at high risk for gestational diabetes mellitus (GDM) based on MOH guidelines were advised to undergo screening. However, one in five pregnant women screened in GUSTO were diagnosed with GDM—with over half missed by the original guidelines.
“On follow-up, 43 percent of these women developed prediabetes or type 2 diabetes (T2D) five to six years after delivery, which also emphasises the importance of regular screening postpartum,” said Eriksson, adding that GUSTO’s findings have since influenced changes to MOH GDM screening guidelines.
Following in GUSTO’s tracks, another ongoing longitudinal study—the Singapore PREconception Study of long-Term maternal and child Outcomes (S-PRESTO)—was launched in 2015, recruiting women before pregnancy and following their health trajectories through childbirth and beyond. Through both cohorts, researchers worldwide hope to find links between prenatal exposures—such as maternal characteristics, maternal lifestyle and maternal health—with child development, well-being and eventual risk for chronic disorders.
Shifting gears to a later developmental period, A*STAR IHDP Principal Scientist Evelyn Loo, Senior Scientist Desiree Phua and colleagues are investigating the intertwined effects of biology and environment on adolescent well-being. In partnership with the A*STAR Institute of High Performance Computing (A*STAR IHPC), Singapore’s Ministry of Education, and the National Institute of Education, the iAdoRe study seeks to identify key factors that promote adolescent vulnerability to or resistance against stressors.
“iAdoRe integrates measures of physical and mental health, socioemotional competency and support, and biomarker data from around 1,191 Secondary 2 students in Singapore,” said Phua.
While data collection is still underway, Phua noted that iAdoRe has revealed interesting differences in how younger and older adolescents in Singapore express their experiences and emotions. Compared to 2010, today’s adolescents have different perceptions of resilience and associated attributes, pointing to evolving forms of mental health-related expressions influenced by maturation and surrounding societal circumstances.
By gaining insight into how Singapore’s adolescents approach their own psychological condition, the iAdoRe team hopes to identify protective elements and potential interventions for their improved welfare. “An effective response system for enhancing adolescent resilience and well-being must account for sociocultural context and maturation in an increasingly unpredictable world,” said Phua.
The iAdoRe sub-cohort is part of the Integrative Adolescence Research Programme (IARP), a multidisciplinary research programme which aims to examine individual, familial and community-level determinants of identity formation, resilience building, and overall physical and mental well-being in adolescents.
“Building on data from GUSTO, S-PRESTO and other local cohort studies, IARP also seeks to identify levers for early interventions in adolescent well-being, and to develop public health recommendations that establish a resilient cohort of young, productive working adults in future,” Loo added.
Exposing disease determinants
Lessons from the Singapore Chinese Health Study
Large, long-term population studies are vital not only to monitor trajectories of human development, but also of disease. The Singapore Chinese Health Study (SCHS), which involved 63,257 older adults in Singapore of Chinese descent, led to a wealth of insights on the dietary and environmental determinants of chronic diseases in Asian populations, said Woon-Puay Koh, a Professor at the NUS Yong Loo Lin School of Medicine, an A*STAR IHDP Senior Principal Investigator and study leader.
With over 520 resulting publications, SCHS has shown that higher midlife diet quality—which includes increased intake of plant-based foods, nuts, dietary antioxidants and B vitamins—is associated with reduced risk of cancers, cardiovascular diseases, and cognitive and physical frailty, as well as better late-life ageing outcomes. SCHS findings have also elucidated the roles of smoking, alcohol, sleep disturbances, obesity and chronic diseases such as diabetes and gout in accelerating mortality, frailty, and cognitive or functional decline.
“SCHS highlighted the importance of studying our own Asian population to understand the influence of genetics and lifestyle on health outcomes, and to develop culturally appropriate prevention strategies,” Koh added. “Beyond establishing associations between exposures and risk of disease, our research uses molecular and genetic tools to identify potential modifiable factors for disease prevention, or potential biomarkers for early detection or screening of disease.”
Precision nutrition for Asian palates
Dietary habits are among the strongest drivers of chronic disease. To address this, researchers such as Germaine Yong, a Scientist at the A*STAR Singapore Institute of Food and Biotechnology Innovation (A*STAR SIFBI), are working towards a future of precision nutrition, where individuals can receive personalised and real-time dietary and lifestyle recommendations.
Together with local biotech company AMILI—host of the world’s largest Asian gut microbiome database—Yong and colleagues embarked on Project REMEDY (Researching the Effects on Microbiome from Ethnic Diets and Natural Product Yields) to gather real-world evidence on the links between diet, sleep, blood sugar levels, multi-omics data and metabolic health.
“As research on Asian nutrition remains limited, REMEDY aims to dissect the impact of the tremendous diversity of Asian cuisine on the gut microbiome in multi-ethnic Asian populations,” said Yong.
As an example of differences between Asian gut microbial profiles and their Western counterparts—the latter often tied to high animal protein intake—Yong pointed to the greater dominance of Segetella bacteria in Asian gut microbiomes, driven by high-fibre and carbohydrate-based diets.
“Studying these differences allows us to build actionable, localised insights that help identify Asian-specific nutraceuticals and functional foods for better health outcomes,” said Yong. “Through REMEDY, we also hope to eventually advance personalised nutrition models that help prevent and manage metabolic disorders.”
Besides the gut microbiome, physiological factors, such as body fat distribution and metabolism, also appear to shape Asian health trajectories in markedly different ways from Western populations, said Melvin Leow, a Professor at the Lee Kong Chian School of Medicine, Nanyang Technological University, and Deputy Head of A*STAR SIFBI’s Nutrition and Digestive Health Division.
“For a given body mass index (BMI), Asian populations tend to have greater amounts of body fat versus Western populations, especially unhealthy visceral fats that add to insulin resistance and T2D risk,” Leow explained.
Not all fat tissue is necessarily unhealthy, however. Brown fat, known to be extremely metabolically active, can contribute to 20 percent of total daily energy expenditures, playing a critical role in staving off weight gain and obesity.
“However, because the natural stimulus for brown fat is cold exposure, it’s not always activated in countries with warm climates,” said Leow. “In Singapore, brown fat activation may only account for about five percent of one’s basal metabolic rate.”
Leow and A*STAR SIFBI colleagues are conducting studies to uncover how functional foods might help boost brown fat activation and the browning of white fat. A variety of compounds, such as capsinoids in sweet peppers and maslinic acids from several fruits and vegetables, have already shown promising results in both aspects, according to Leow.
“Compared to drug-based approaches, a ‘food as medicine’-based approach to preventing T2D and other metabolic diseases is much more feasible to implement at a population scale,” Leow highlighted.
Crunching population numbers
Between obvious skin rashes and less-obvious surges of inflammation-linked cells, allergic reactions are marked by varied physical and molecular symptoms. However, such complex biomarker profiles and immune responses are still poorly defined in many allergic conditions, according to Anand Andiappan, a joint A*STAR Skin Research Labs (A*STAR SRL) Principal Scientist and A*STAR Singapore Immunology Network (A*STAR SIgN) Principal Investigator.
“Allergy care today remains suboptimal or insufficient for 90 percent of patients, leading to significant productivity losses and hampered quality of life,” said Andiappan. “By identifying how allergic responses are triggered, clinicians can avoid trial-and-error and choose treatments that target underlying pathways, leading to better symptom control and fewer medication failures.”
For example, Andiappan noted that antihistamines—a common standby drug for allergic rhinitis—work most effectively for patients experiencing inflammation driven by immunoglobulin E (IgE) antibodies, while patients with mixed or non-allergic inflammation driven by eosinophils may respond better to intranasal corticosteroids and other treatments.
Comprehensive allergy profiles are therefore key to improved treatment decisions, requiring a mix of traditional markers such as IgE and eosinophils, as well as next-generation molecular profiling, microbiome and digital data.
To transform patient outcomes, Andiappan and A*STAR SIgN colleagues are working in collaboration with NUS, the National University Health System (NUHS), and hospitals in Singapore and the US on Project ENTenna. Led by Ng Teng Fong General Hospital, Singapore, the project combines Asia’s first population allergy database with continuous digital engagement via a data-driven AI-supported chatbot.
“Project ENTenna’s allergy database will constitute a scalable reference population for the development of predictive analytics, digital twins and therapeutic targets, enabling personalised patient-centred care and new pathways for research and policy,” said Andiappan.
Digital tools and AI-enabled workflows are also turning the tide of population-level disease screening and health promotion initiatives, noted Pavitra Krishnaswamy, Head of the Healthcare and MedTech Division at the A*STAR Institute for Infocomm Research (A*STAR I2R).
As part of Singapore’s National Precision Medicine programme, coordinated by Precision Health Research Singapore (PRECISE), Krishnaswamy and A*STAR I2R colleagues are working with diverse ecosystem partners to leverage big data and AI for precision screening of genetic disorders. These include the A*STAR Genome Institute of Singapore (A*STAR GIS), A*STAR Bioinformatics Institute (A*STAR BII), A*STAR IHPC, Diagnostics Development Hub (DxD Hub), NUS, Khoo Teck Puat Hospital and NUHS.
The group’s first target is an AI approach to identify individuals with potential familial hypercholesterolaemia (FH): a genetic condition found in one of every 140 individuals that increases risk of premature heart disease.
“FH patients are often under-diagnosed and under-treated, leading to poor outcomes,” said Krishnaswamy. “We anticipate that this project will enhance FH screening efficiency in our wider population through targeted and cascading follow-ups. This will enable FH patients and their family members to receive genetic testing and improve disease management.”
On the preventive front, A*STAR I2R has collaborated with Singapore’s Health Promotion Board (HPB) to support personalised health promotion and wellness interventions within their national programmes. A*STAR I2R researchers have developed scalable big data and AI approaches to uncover meaningful behavioural patterns for robust population-level segmentation, enabling HPB to deliver targeted, timely interventions tailored to the needs of individual Singaporeans.
“By leveraging large-scale population health datasets spanning genomic, clinical, lifestyle and behavioural facets, these projects support national P3H priorities towards a healthier Singapore,” Krishnaswamy added.
Meanwhile, teams at the A*STAR-EVYD Joint Laboratory are rapidly pushing innovations in wearables, diagnostics and medical records across Southeast Asia.
“A*STAR brings world-class research in AI agents, multimodal data science, and behavioural and social research for health, while EVYD contributes production-grade engineering, cloud and security architecture, and importantly, access to real-world deployment environments across the region,” said Zhao Li, Product Director at EVYD Technology.
One major project by the joint lab has been the launch of a national population health platform in Brunei with a holistic array of healthcare services, including risk assessments, personal disease management plans, access to screening programmes and gamified incentives for participation in wellness initiatives. The joint lab is further looking to add AI models for classifying patients to different risk groups and integrating digital biomarkers from wearable devices, all while keeping patient privacy and data security top-of-mind.
“What excites me most is that such platforms are not pilots—they are live systems serving real populations,” said Li. “The goal is to evolve what is already a nationwide platform into a truly intelligent population-health system that promotes healthier lifestyles, earlier detection and more efficient healthcare resource use.”
From data to action
Whether from microbial profiles, allergy biomarkers or brain scans, A*STAR is working towards a deeper understanding of what constitutes ‘Asian biology’ to guide public health policies and enhance population health in the region, Ng commented. Besides physiological factors, environmental and social variables are also at the forefront of A*STAR research priorities in designing heterogenous cohorts that reflect Singapore’s multi-ethnic population.
“A*STAR also aims to build data bridges by investing in and developing integrative, multimodal, AI-enabled analytics platforms, with many already being deployed and tested within current healthcare processes,” Ng added. “The agency is also working with healthcare and research partners to improve data interoperability and linkages for advanced analytics, as well as to streamline data access.”
Backed not just by innovative research but also strong clinical and government partnerships, Ng noted that P3H approaches can lead to better risk prediction models tailored to local groups, and intervention strategies aligned with on-the-ground needs. These large-scale collaborative efforts are influencing policy decisions and transforming healthcare practices with a predictive and preventive focus, ultimately shifting the bar of ‘a healthy population’ for the better.
