Sample Article Review On Probability Analysis In Public Health Research

Probability is a statistical concept that tries to define the chances of certain event occurring based on existing information on the variables. Probabilistic analysis is a crucial part of most public health research studies and provide effective management tools to administrators and public health professionals alike. The present study titled “probabilistic uncertainty analysis of epidemiological modelling to guide public health intervention policy” was published in the peer-reviewed journal Epidemics in 2014(Gilbert, Meyers, Galvani, & Townsend, 2014). This study investigates how probabilistic uncertainty analysis can be used to control disease outbreaks more effectively.

Article Summary

Mathematical probabilistic modelling has been used to drive policy in public-health situations, however the suitability of the model is largely dependent on the factors included in the model. Therefore, some mathematical models may fail due to the non-inclusion of some uncertain factors that were not measured while creating the said model, which might result in a public-health disaster. The researchers Gilbert, Meyers, Galvani & Townsend used a probabilistic uncertainty analysis on a global dynamic model of influenza transmission to show that unknown factors can offset the public health goals by more than 50% and thereby give rise to a health crisis. The researchers showed that probabilistic uncertainty analysis can enhance the robustness of mathematical models of infectious diseases and help policy makers take better decisions that effectively curb pandemics.

Results

The figure 2 of the research article shows two graphs where the percentage probability of restricting the spread of influenza virus has been computed using a. the mechanistic model and b. using the probabilistic uncertainty analysis. Re is defined as the number of secondary infections following a primary infection in a non-naïve population. The existing model of influenza (fig a) transmission showed that the probability of global spread of the disease could be curbed by vaccination of 54% of susceptible individuals or treatment of 64% of affected individuals and the Re would fall below 1 (effectively eradicating the disease). However, when the probabilistic uncertainty analysis was included (fig b) it was found that more intensive treatment regimens were required to eradicate the disease. It was found that when 76% of the susceptible individuals were vaccinated then only the Re would drop below 1 with probability above 99%. When only treatment was used as a public health measure 87% of affected individuals had to be covered to reduce Re below 1 with a probability of 90%. The levels of treatment computed by the previous model (without uncertainty analysis) was found to be only half as effective as previously computed.
Summary

References

Gilbert, J. a., Meyers, L. A., Galvani, A. P., & Townsend, J. P. (2014). Probabilistic uncertainty analysis of epidemiological modeling to guide public health intervention policy. Epidemics, 6, 37–45. doi:10.1016/j.epidem.2013.11.002