HLTH 6483 Module 4 Epidemiological Study Critique Example

Does El Nino Predict Vibriosis? A Critique of an Ecological Time-Series Study Linking Ocean Phenomena to Non-Cholera Vibrio Infections in the United States

Student Name

American College of Education

HLTH6483: Principles of Public Health and Epidemiology

Module 4 Assignment

Instructor Name

October 5, 2026

Introduction

The Galveston Bay outbreak analyzed in Module 3 occurred when coastal water was warmer and saltier than usual, and the county followed in this course has seen its worst vibriosis years during unusually warm summers. Whether warming water drives vibriosis is therefore the central causal question for prevention planning. This paper critiques a study that addressed it at the national level: Logar-Henderson et al. (2019), who used United States surveillance data to test whether El Nino conditions, a natural source of ocean warming, were followed by more vibriosis. The critique examines the research question, design, measurement, analysis, sources of bias and confounding, and the strength of causal inference, then considers what the study means for the county.

The Study in Brief

The authors obtained counts of non-cholera Vibrio infections for the United States from publicly available surveillance data. They used negative binomial regression to model trends and seasonality and distributed non-linear lag models to relate disease to the El Nino Southern Oscillation and other large-scale ocean phenomena over the following months. They also used meta-regression to test whether changes in risk varied with latitude. They reported clear seasonality, an annual increase in incidence of 7% from 1999 to 2014, an increased risk of vibriosis in the 12 months after El Nino conditions, with a relative risk of 1.940, and a relative increase in annual incidence that grew with latitude, 1.066 per 10 degrees. They concluded that vibriosis risk in the United States is affected by ocean warming (Logar-Henderson et al., 2019).

Research Question and Rationale

The question is important and well framed. Because El Nino periodically warms ocean waters, it offers what the authors called a natural experiment: a source of variation in ocean conditions that people do not choose, which avoids some of the confounding that affects comparisons of warmer and cooler places. Framing climate variability as a natural experiment is a creative way to approach a question that cannot be tested by randomization. The question also has clear public health relevance, since a link would support using ocean forecasts to warn people at risk.

Design

The study is an ecological time-series analysis: both exposure and outcome are measured for populations over time, not for individuals. This design suits the question, since ocean conditions affect whole coastlines, and it allows long periods to be studied cheaply with existing data. Its central weakness is the ecological fallacy risk: an association between El Nino and national case counts does not show that the individuals who became ill were exposed to warmer water. An ecological design can show that bad years for the ocean are bad years for vibriosis; it cannot show which swimmers or oyster eaters were harmed by the warmth.

Measurement of Exposure and Outcome

The exposure, an index of El Nino conditions, is a large-scale climate measure rather than local water temperature at the places where people were infected. El Nino's effects on coastal waters vary by region and season, so the index is an indirect proxy, and misclassification would likely weaken associations rather than create them. The outcome, reported vibriosis, comes from surveillance that has changed over time. Vibriosis became nationally notifiable in 2007, in the middle of the study period, and national surveillance reviews show that reported rates rose while prevention and detection evolved (Newton et al., 2012). Reporting completeness likely varies by state and year.

Analysis

The analytic choices are appropriate for count data over time. Negative binomial regression handles counts that vary more than a simple Poisson model assumes, and distributed lag models allow the effect of an exposure to be spread over the following months rather than forced into a single lag. The inclusion of seasonal terms separates the regular summer peak from year-to-year variation. The meta-regression by latitude is a useful test of a prediction that follows from the warming hypothesis: if warming drives risk, the relative increase should be larger in cooler, northern waters where temperatures are crossing thresholds for Vibrio growth.

Bias and Confounding

Several alternative explanations deserve attention. The steady 7% annual increase could reflect better testing, greater clinician awareness and the change to national notification as much as real growth in infections. Population growth in coastal areas and changes in raw oyster consumption or recreational water use could raise case counts independently of ocean temperature. The El Nino association is less vulnerable to these trends, because El Nino years come and go while reporting and population change gradually, which is a strength of the natural experiment approach. However, El Nino also changes rainfall, which alters coastal salinity, another factor that affects Vibrio growth, so the study may capture several ocean effects bundled together.

Causal Inference

Judged overall, the evidence supports a causal role for ocean warming, though not conclusively on its own. The association is plausible given Vibrio biology, consistent with the seasonal pattern, strengthened by the temporal sequence in which El Nino precedes increased illness and supported by the latitude gradient the hypothesis predicts. Independent evidence points the same way: using a 30-year database, Archer et al. (2023) found that V. vulnificus wound infections in the eastern United States increased eightfold between 1988 and 2018, from about 10 to 80 cases a year, and that the northern limit of cases moved northward about 48 kilometers a year. Consistency across different designs and data sources is one of the strongest supports for causal inference.

What a Stronger Study Would Add

A study designed to answer the causal question more directly would link individual cases to the water conditions each person actually experienced. A case-crossover design, for example, would compare water temperature and salinity at the place and time of each person's exposure with conditions at the same place on other days, so that each case serves as its own control and stable personal factors such as liver disease cannot confound the comparison. Such a study needs exposure location and date for every case, which state case report forms often collect, and water readings from nearby monitoring stations, which are increasingly available. Adding information on testing practices, for instance the number of stool and wound cultures performed each year, would also allow analysts to separate a real rise in infections from a rise in detection. None of this diminishes the value of the national study, which pointed to where more focused research should go.

Implications for the County

For a Gulf coast county already at high risk, the study's national findings suggest that El Nino years and unusually warm summers should prompt stronger warnings, especially to older adults, people with liver disease and water workers identified in Module 2. Because the exposure measure was national, the county should test the relationship locally by linking its own case counts to nearby water temperature and salinity readings, which would provide more specific evidence. The latitude finding also suggests that the county's summer warnings should become longer as warm seasons extend.

Conclusion

Logar-Henderson et al. (2019) asked an important question with a creative design and suitable analytic methods. Its ecological nature, proxy exposure and changing surveillance limit what it can prove on its own, but its findings fit Vibrio biology and agree with other studies using different data. The study strengthens the case that warming water is raising vibriosis risk and supports using ocean conditions to guide prevention. Module 5 turns to evaluating a program designed to reduce that risk.

References

Archer, E. J., Baker-Austin, C., Osborn, T. J., Jones, N. R., Martínez-Urtaza, J., Trinanes, J., Oliver, J. D., González, F. J. C., & Lake, I. R. (2023). Climate warming and increasing Vibrio vulnificus infections in North America. Scientific Reports, 13, Article 3893. https://doi.org/10.1038/s41598-023-28247-2

Logar-Henderson, C., Ling, R., Tuite, A. R., & Fisman, D. N. (2019). Effects of large-scale oceanic phenomena on non-cholera vibriosis incidence in the United States: Implications for climate change. Epidemiology and Infection, 147, Article e243. https://doi.org/10.1017/S0950268819001316

Newton, A., Kendall, M., Vugia, D. J., Henao, O. L., & Mahon, B. E. (2012). Increasing rates of vibriosis in the United States, 1996-2010: Review of surveillance data from 2 systems. Clinical Infectious Diseases, 54(Suppl. 5), S391-S395. https://doi.org/10.1093/cid/cis243