Secondary Logo

Journal Logo

The Authors Respond

Sandberg, John; Santos-Burgoa, Carlos; Roess, Amira; Goldman-Hawes, Ann; Pérez, Prof Cynthia M; Garcia-Meza, Alejandra; Goldman, Lynn R.

doi: 10.1097/EDE.0000000000001098
Letters
Free

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC, jsandber@gwu.edu

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Department of Epidemiology and Biostatistics, Milken Institute School of Public Health, The George Washington University, Washington, DC

Department of Biostatistics and Epidemiology, Graduate School of Public Health, University of Puerto Rico, San Juan, Puerto Rico

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Department of Environmental and Occupational Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

The authors report no conflicts of interest.

Back to Top | Article Outline

To the Editor:

The first assertion in this letter1 is that we misrepresented the authors’ results concerning excess mortality. We did not. Their analysis, cited in our table as “Santos-Lozada 2”, presents separate estimates of excess mortality for September, October, and November.2 We presented their estimate for the period from September to October to enable comparison to their prior estimate and others covering the same period. We did not present their result for September to November as no other estimate for that period had been produced.

We have addressed the remaining critiques concerning displacement assumptions in our response to a previous letter in another journal by the same authors.3,4 We encourage readers to read that response but will summarize here.

Concerning the critique that displacement models should account for potential selectivity in migration, we agree. We would argue that the issue of how selectivity may impact estimates is not as straightforward as the authors suggest, however. While voluntary migration tends to select for those in better health, the same is not likely to be true of compelled migration as in the aftermath of a disaster such as Hurricane Maria. Refugees may equally be negatively selected on health status.5,6 Anecdotal evidence from our work in Puerto Rico for this project suggests that this type of negative selection may have been at play. In general, researchers should attempt to minimize potential selection biases. Unfortunately, relatively little data suitable for this are available in most registration systems. As noted both in our original research and prior response, our own displacement estimates are conditioned on age, sex, socioeconomic status of municipality of residence, and interactions between them.7 These are major factors stratifying the risk of mortality, and likely, selectivity in displacement. To the extent that this is so, selection on these factors is accounted for in our estimates.

Finally, the assertion that our displacement estimates assume a higher mortality rate among the displaced is erroneous, based on the authors’ simple tabular analysis of mortality rates and population size at the end of the period and cumulative predicted deaths. Their analysis ignores monthly variation both in displacement and mortality, which our estimates explicitly take into account. As we have shown both in our original article and prior response, those displaced by Hurricane Maria were disproportionately removed from the population at risk during the period of highest seasonal mortality. Considered over the entire period as the authors do, their implied mortality rate appears higher simply because they were differentially removed from the population when mortality rates were highest.

John Sandberg

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC, jsandber@gwu.edu

Carlos Santos-Burgoa

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Amira Roess

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Ann Goldman-Hawes

Department of Epidemiology and Biostatistics, Milken Institute School of Public Health, The George Washington University, Washington, DC

Prof Cynthia M Pérez

Department of Biostatistics and Epidemiology, Graduate School of Public Health, University of Puerto Rico, San Juan, Puerto Rico

Alejandra Garcia-Meza

Department of Global Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Lynn R. Goldman

Department of Environmental and Occupational Health, Milken Institute School of Public Health, The George Washington University, Washington, DC

Back to Top | Article Outline

REFERENCES

1. Santos-Lozada A, Howard JT. Re: excess death estimates in Puerto Rico have been consistent all along. Epidemiology. 2019;30:e35e36.
2. Santos-Lozada AR, Howard JT. Use of death counts from vital statistics to calculate excess deaths in Puerto Rico following Hurricane Maria. JAMA. 2018;320:1491–1493. Retrieved September 22, 2018, from https://jamanetwork.com/journals/jama/fullarticle/2696479
3. Howard JT, Santos-Lozada A. Mortality risk within counterfactual models: Puerto Rico and Hurricane Maria. Lancet Planetary Health. 2019;3:e207–e208.
4. Santos-Burgoa C, Sandberg J, Suárez E, et al. Mortality risk within counterfactual models: Puerto Rico and Hurricane Maria. Lancet Planetary Health. 2019;3:e209.
5. Akresh IR, Frank R. Health selection among new immigrants. Am J Public Health. 2008;98:2058–2064.
6. Jasso G, Massey DS, Rosenzweig MR, Smith JP; National Research Council, eds. Immigrant health: selectivity and acculturation. In: Critical Perspectives on Racial and Ethnic Differences in Health in Late Life. 2004:Washington, DC: The National Academies Press; 227–266.
7. Santos-Burgoa C, Sandberg J, Suárez E, et al. Differential and persistent risk of excess mortality from Hurricane Maria in Puerto Rico: a time-series analysis. Lancet Planet Health. 2018;2:e478–e488. Retrieved October 16, 2018, from https://www.thelancet.com/journals/lanplh/article/PIIS2542-5196(18)30209-2/abstract
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.