Malaysia has made significant progress in collecting health data, particularly through national surveys like the National Health and Morbidity Survey (NHMS).
We know, for instance, that one in five Malaysian adults has diabetes, and nearly half are overweight or obese. These figures are often cited in health campaigns and policy documents.
Yet, for all this data, Malaysia continues to struggle with controlling the rising burden of non-communicable diseases (NCDs) and other chronic health threats.
The problem is not a lack of numbers, but a lack of understanding.
Our public health system places strong emphasis on descriptive epidemiology. We report prevalence, incidence, and mortality rates, and then respond with generalised behavioural campaigns—“eat healthy”, “exercise more”, “get screened”.
But we often stop short of asking the deeper questions: Why are these diseases so prevalent in certain communities? What upstream factors are driving these trends? And most importantly, how can we interrupt these pathways before disease occurs?
In short, we are not doing enough to uncover causality.
For example, while diabetes rates are high nationwide, we lack detailed studies that explore how early-life nutrition, urbanisation, socioeconomic pressures, or even school and work environments contribute to the disease burden.
We seldom conduct spatial epidemiology to identify disease clusters linked to specific neighbourhood conditions. We rarely look into policy-level determinants such as food pricing, urban planning, or advertising regulations.
The result is a cycle of reactive policymaking. We screen, diagnose, and medicate, but we miss the chance to prevent disease at its root. We manage the consequences instead of preventing the causes.
This is not unique to diabetes. Similar patterns are seen in obesity, mental health, substance abuse, chronic kidney disease, and even infectious diseases like dengue.
Despite recurring outbreaks, public health responses often focus on immediate containment rather than mapping long-term environmental and behavioural drivers that allow these diseases to persist.
Malaysia’s public health data infrastructure is solid, but it is not yet directed toward action. Our ministries and research bodies often operate in silos, with limited integration of health data with socioeconomic, educational, or environmental data.
There is little policy evaluation. Campaigns and programmes are launched, but their effectiveness is rarely studied in depth. Without feedback loops, our policies remain static, regardless of changing epidemiological patterns.
To move forward, Malaysia must begin to treat epidemiology not just as a statistical tool, but as a strategic asset for national health planning.
We need to invest in the training and deployment of field epidemiologists, not only for infectious disease surveillance but also for investigating the social and structural determinants of health.
Public health units should be empowered to conduct risk mapping and causal analysis as part of routine planning. Epidemiological reasoning must become the standard for policy design, not an afterthought.
Moreover, prevention must be approached with the same rigour as outbreak response. When faced with leptospirosis or COVID-19, we mapped the sources, identified exposure routes, and applied targeted interventions.
This same logic should apply to chronic disease. If obesity is concentrated in urban low-income areas, we should examine the built environment, food access, and economic stressors that shape health behaviours there.
Malaysia’s disease burden will not decline until our policies are informed by the true drivers of disease. Descriptive data is a starting point, but without understanding causality, we risk repeating the same patterns year after year. The future of public health in Malaysia depends on making that shift—from prevalence to prevention, from reaction to strategy.
The author is an internal medicine specialist.
- This is the personal opinion of the writer or publication and does not necessarily represent the views of CodeBlue.
