The cold of winter and holiday gatherings push people indoors, causing a surge in influenza hospitalizations.
Years of above-normal temperatures in southern states bring a species of mosquito that carries malaria to the US. Declining childhood immunization rates threaten to allow previously eradicated diseases like measles to become endemic again. Aquatic birds travel from Europe to Iceland, then North America, bringing a new strain of avian influenza that causes the most prolonged and severe outbreak among domestic poultry in US history.
These are all different events with varying long–term and short–term outcomes, but they have one thing in common: They are predictable.
The meaning of “epidemiology” is the study of disease trends, patterns, and how they move within populations. Epidemiologists know that most diseases have extrinsic levers that, if monitored, can predict how and when a disease will affect a population.
Forecasting and modeling are becoming the cornerstone of public health work. Whether addressing drug overdoses or conducting flu surveillance, the strategic use of data is essential for prioritizing interventions that can safeguard public health. Here's what needs to happen throughout 2024 and beyond.
Disease modeling and forecasting will continue to evolve
The U.S. public health system has largely focused on modeling and forecasting efforts on strongly predictable diseases such as influenza for many years. The only variables typically used in influenza models are past disease patterns, type and strain, hospitalizations and deaths. Yet we know temperature plays a role in the timing of influenza spikes and is rarely included in influenza forecasts.
The same is true for disease vectors like mosquitoes, ticks, or even wild birds. Scientists know these insects and animals carry diseases that can directly or indirectly spread to people. However, public health needs more resources to proactively detect and track these vectors and subsequent diseases.
Recognizing this, the CDC introduced a new approach for advancing modeling and forecasting of disease. Instead of placing the burden of prediction on states and local public health jurisdictions, the CDC will use 13 regional hubs, primarily academic centers and one health system, to set up and run models and forecasts for infectious disease.
This approach is a positive one. Regional models will likely be accurate for multiple states with the ability to drill into state-level data. Academic and health care sectors have the knowledge and expertise to run advanced models and forecasts. They have experience incorporating various independent variables such as climate data, migratory bird patterning, mosquito surveillance, and other unique and influential data.
Of utmost importance, these hubs could provide models and forecasts for diseases that inform clinicians and the public, potentially preventing illness and death.
The need for an agile and expansive technological foundation
Importantly, the technological foundation supporting this initiative must be both expandable and agile. Data sources and ingestion processes should be similar across hubs. The analytic languages and models used should be accessible to people with varying preferences and experience levels. SAS® Viya® makes that happen. It bridges the gap between entry-level and highly experienced analysts, allowing seamless interaction between open source and traditional coders. SAS Viya can manage extensive data and may be modified for nearly any analytic scenario.
The value of working from a solid foundation of predictive data in public health must be recognized. Forecasts help predict when infections of diseases might change or overlap. Models help health care systems pre-position resources to ensure critical patients receive the help they need and vulnerable people are protected with medication or vaccination.
A new era of public health
In 2024, if these models and forecasts are extended beyond what we currently have for seasonal diseases, we will be in a new era of public health. One where we are no longer defeated because an outbreak became a pandemic before we detected the outbreak. Mitigation measures like increased animal biosecurity, mosquito abatement, or social distancing can save more lives than ever imagined.
