Statistics can save lives

Statistics can save lives
© Getty Images

While much of Washington, D,C. and the nation is focused on the U.S. Supreme Court nomination hearings, the storm water in the Carolinas from the recent Hurricane Florence is still dangerously high. The hard recovery work is just beginning. Understanding the scope of Hurricane Florence’s impact on communities, businesses, families, and sadly, lost lives will take time, effort and funding to rebuild.

And, recently, we learned the extent of the lives lost in Puerto Rico following Hurricane Maria. One year ago, Hurricane Maria devastated a broad swath of the Caribbean, hitting Puerto Rico especially hard. The infrastructure was devastated, depriving the vast majority of the population of power, clean water, other utilities and access to medical care. The initial death toll of 64 people jumped into the thousands after two independent analyses raised the estimated number of storm-attributable deaths.

With the current administration still questioning the figures used to measure storm deaths from Maria, we must better understand how and why the number of storm attributed deaths increased so dramatically and to clarify how the reported death counts were calculated. The evidenced-based approach to estimating the deaths from natural disasters illustrates how mortality may occur. Deaths occur not only from the disaster itself, but also from breakdowns in infrastructure in the post-disaster period. These complete measurements of storm impacts are critically important for future disaster planning.

As the chairs of the North American biostatistics departments, we conducted a review of the two scientific reports on estimated deaths attributable to Hurricane Maria. We found the methodology used in both studies to be well-established and statistically rigorous. Both estimates well exceed the initial report of 64 deaths after considering the limits of precision and data quality. The available evidence supports the number of storm-attributable deaths estimated by the recent independent analyses.                                                                                      

The first study, conducted by the Harvard Chan School of Public Health and published in the New England Journal of Medicine, was independently reviewed by scientists not involved in the work. The second study was commissioned by the government of Puerto Rico to help them better understand the scope and causes of the excess death. This study was performed by researchers at George Washington University in collaboration with the University of Puerto Rico Graduate School of Public Health.

Both studies began by estimating the death rate over time in Puerto Rico before, during and after the hurricane. Prior mortality rates came from government records collected yearly by the Puerto Rican government, irrespective of a major hurricane or other natural disaster.  The analysis essentially subtracted the number of deaths observed during and after the hurricane from the number that one would expect based on the numbers of deaths occurring during the same months in the previous year(s).

The Harvard researchers estimated that 4,645 deaths were attributable to the storm, and the GWU researchers, 2,975. In each case, the analysis estimated the number of deaths beyond what one would have expected in Puerto Rico if the hurricane had not occurred. It’s important to understand that they did not count every death that occurred during the post-hurricane period.

The difference in the two numbers reflect differences in the ways the two groups counted deaths during the hurricane and its aftermath and the specific analyses the two groups performed. They also reflect that each estimate has a degree of imprecision for characterizing the actual death toll. Characterizing the extent of this imprecision appropriately is the central focus of statistics.

Verifiable, trustworthy data on a storm’s impact helps inform the government’s immediate and long-term response. This information may also help people in the path of the next storm as they decide to shelter in place or leave for drier ground. Knowing how many people died from storm-related issues could be the tipping point for folks who need an extra push to get to safety.

Statistics — grounded in rigorous, objective data collection — can save lives. For example, data connecting specific types of infrastructure failures with mortality patterns from Hurricane Maria and other past storms could be analyzed to identify preventive actions, which could save lives. Hurricane Florence, in turn, can serve as a springboard to improve the response to further disasters.

Our nation’s approach to public health surveillance must be based on the collection of reliable, representative vital statistics that do not reflect any premeditated political agenda. Full transparency, which the authors of both Hurricane Maria studies provide, allows for open debate of assumptions that underlay the resulting calculations and resulting differences in death toll estimation.

As we wait to learn the full impact of Hurricane Florence, it’s crucial that we learn from the controversy over the death toll from Hurricane Maria so that the data can speak for itself and help inform the response to future storms. Safeguarding the integrity of this public health paradigm is vital to maintaining the health of our nation’s citizens.

Debashis Ghosh is a professor and chair at the department of Biostatistics and Informatics, Colorado School of Public Health. Karen Bandeen-Roche, Hurley Dorrier Professor and Chair, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health. Jason Roy, Professor and Chair, Department of Biostatistics and Epidemiology, Rutgers School of Public Health.