Why study the treatment for a disease when you can study how to prevent it? This was the question asked by infectious disease specialists when developing the COVID-19 vaccine and is currently being asked by leaders in the field of preventative medicine. Within preventative medicine, there is a new breakthrough that predicts future health events such as demands for health services, known as health forecasting. This breakthrough is a valuable tool that helps “pre-inform health service providers to take appropriate mitigating actions to minimize risks and manage demand” . In other words, it forewarns healthcare systems to prepare for possible health situations. For example, if a hospital could predict the outbreak of a disease, then the hospital could allocate more resources ahead of time to meet the demands of an infected population. These resource allocations are known as interventions and they are used alongside forecasts for better patient outcomes. Other examples of interventions include anticipatory alerts to the general public such as the ones that were released during the second and third waves of the COVID-19 outbreaks.
As a newly developing field, there is no single approach to health forecasting. Most health forecasting is completed through adapted forms of statistical procedures. This requires health data to be collected, which was a difficult process until the emergence of electronic health records (EHR) or “digital health data that is stored in secured repositories and shared only among authorized users” . Using this data, healthcare providers are able to predict many health phenomena. For example, linear regression methods provide accurate results, are easy to interpret, and have wide applications in modeling trends. However, this method incorrectly ignores errors and requires very large amounts of data .
Since health forecasting is modeling future events based on available data, it is as much an art as it is a science. Various techniques are used to determine both the accuracy and fit of the models in a forecast. At its current state, no one model is suitable for all. Thus, further research is needed to refine health forecasts to better predict dangerous outbreaks.
One particular disease that has been studied with health forecasting is Chronic Obstructive Pulmonary Disease (COPD) because of convincing evidence that “COPD exacerbations and admissions can be reduced by predicting periods of cold weather coupled with patients’ alerts and education” . A study in 2011 observed 157 participants who received varying types of weather alerts. The patients that received anticipatory weather alerts did not see a reduction in hospital admissions from COPD exacerbations . Another study investigating COPD used a “rulebased model predicting risk based on environmental conditions with an anticipatory care intervention. It provides information on self-management and warnings via an interactive telephone call” . Scientists of this study served patients receiving these warnings and concluded that they “added to patients' understanding of their illness and promoted better self-management” . These different outcomes for the same disease further prove that health forecasting is not a “one size fits all” solution.
Therefore, the future of health forecasting remains unclear. Although there are promising outcomes from health forecasting, some argue that using health forecasting to effectively prevent disease will be difficult because “we live in a medicalized society in which it is easier to take an inhaler or tablet rather than it is to change personal behavior” . Additional research is needed for health forecasting to be standardized and to determine to what extent their predictions will be useful in preventative care in the future. Nonetheless, health forecasting offers valuable insight into health trends and should be carefully studied by healthcare systems.
1. Soyiri, I. N., & Reidpath, D. D. (2013). An overview of health forecasting. Environmental health and preventive medicine, 18(1), 1–9. https://doi.org/10.1007/s12199-012-0294-6
2. Soyiri, I. N, & Reidpath D. D. “Evolving forecasting classifications and applications in health forecasting.” International journal of general medicine vol. 5 (2012): 381-9. doi:10.2147/IJGM.S31079
3. Bakerly, N. D., Roberts, J. A., Thomson, A. R., & Dyer, M. (2011). The effect of COPD health forecasting on hospitalization and health care utilization in patients with mild-to-moderate COPD. Chronic Respiratory Disease, 5–9. https://doi.org/10.1177/1479972310388950
4. Marno, P., Chalder, M., Laing-Morton, T., Levy, M., Sachon, P., & Halpin, D. (2010). Can a health forecasting service offer COPD patients a novel way to manage their condition?. Journal of health services research & policy, 15(3), 150–155. https://doi.org/10.1258/jhsrp.2009.009035
5. Ward M. (2011). Health forecasting and COPD. Chronic respiratory disease, 8(1), 3–4. https://doi.org/10.1177/1479972311398882