Books
‘Deadly River – Cholera and Cover-Up in Post-Earthquake Haiti’
After much epidemiologic investigation, it became evident that cholera, never before in Haiti, had originated near the interior Haitian city of Mirebalais. The city sits in Centre department, on the banks of the Artibonite River a few miles north of a United Nations (UN) military base housing peacekeeping soldiers on a six-month assignment from Nepal. They had arrived in three shifts in October 2010, shortly before the world’s greatest cholera outbreak was first announced on October 21 by PAHO, the Pan American Health Organization. Maps created by the UN’s OCHA, or Office for Coordination of Humanitarian Affairs, would soon tell a tale of the budding epidemic, but offered a view more convenient to the UN.
Maps are closely intertwined with epidemiology. They allow disease detectives to view illness or death patterns in space and time, showing gradual—or sometimes explosive—spread through a community, country, or even the world. If the mapmaker intends to focus on disease cases without taking into account the underlying population, spot maps are typically used. They show where each case resided or worked—suggesting a possible relation to a suspected source.
When the underlying population is known, maps are created that use colors, patterns, or shading to show varying levels of disease rates. Here the epidemiologist must first know the population of a geographic area—most often gathered from a recent census. The population is combined with the number of illness cases or deaths—usually obtained from health departments—to derive rates per time period when cases occurred. The colors, patterns, or shades are then assigned to the geographic areas of the map to tell stories of associations or possible causes.
Maps were used in a similar manner in Haiti. On the day after cholera was officially reported in the country, PAHO and the Haitian government published the first disease map of the outbreak.
The map showed cholera in two departments, Artibonite (closer to the Caribbean sea) and Centre (in the interior of the country). It focused attention on the Artibonite River that flows through the two areas, which made perfect sense given cholera’s water-loving nature. The report emphasized that reported cases were “mainly coming from communities along the Artibonite river basin that runs through the mentioned departments.”
Was the northwest-flowing Artibonite River the source of early transmission, or did cholera spread along roads with travelers or food and water products moving from one region to another? If cholera began in the coastal communities, how did the agent get to the interior towns of the Artibonite valley or even farther to the city of Mirebalais in Centre department? The map offered no answers but certainly posed interesting questions.
Shortly after the first PAHO/government cholera map was published, the UN’s OCHA began to issue a series of colored maps, sometimes in English and other times in French. They showed cholera by commune, the smallest administrative unit.
In the upper Artibonite River region, cholera appeared in two communes next to the river near Mirebalais, coded in light red for places affected in Centre department. Moving downriver to the northwest there was a gap, with no cholera in the next two communes. Continuing toward the coast, cholera was evident in many communes by the river, coded in dark green for outbreak areas in Artibonite department, as it flowed toward the ocean. By focusing on the communes in the two departments the UN map provided a much more useful level of detail than the first PAHO/government map.
Two days later, the UN’s OCHA published another map, this time showing cumulative deaths and hospitalizations for October 20 through 26. The map, this time in French, had a slight but notable change in its legend for cholera in the communes in Artibonite department near the river: what had been called an “outbreak” in English now was called an “epidemic” in French, but otherwise was similar to the earlier one, other than showing further spread of the disease. A day later, Al Jazeera reporter Sebastian Walker and Associated Press reporter Jonathan Katz, broke their stories about Nepalese peacekeepers bringing cholera to Haiti following an outbreak in their home country.
Another day passed before OCHA issued its next map in the evening of October 28, again in French. By then, the epidemic had continued to expand. The map presented the cumulative deaths and hospitalizations in Centre, Artibonite, and the northwestern region from October 20 through 27. The map looked similar to the earlier map, but there was a major change to the legend, telling of the source: now the communes by the river in Artibonite department, still in dark green shading, were identified as the “zone where cholera began.”
Light red shading for communes affected by cholera—but not “where cholera began”—was used for the two communes in Centre department, near the UN Nepalese peacekeeper camp visited by the two reporters. The same light red shade was also used for communes away from the river in Artibonite department and many in the northwestern department.
The UN’s OCHA maps continued to falsify where cholera began until November 10, 2010, when the last in the series was issued showing cumulative data from October 20 through November 8. Each time, the UN map showed a dark green area in the lower Artibonite valley surrounding the Artibonite River as the region where the epidemic had its origin. Yet upriver in the Mirebalais commune, surrounding the UN Nepalese peacekeeper base, the map legends stated the area was merely “cholera affected.”
With this erroneous legend, the maps of the UN agency skillfully shifted attention away from the human transmission hypothesis and the Nepalese peacekeepers as the source and toward the environmental hypothesis suggesting the disease came from the coastal estuaries of the Artibonite River delta. Had reporters’ dispatches influenced the UN mapmakers to change the legend? To French epidemiologist Renaud Piarroux, later reviewing the maps, it certainly seemed possible. It was not only epidemiological falsehood but also cartographic deception.
Excerpted from Deadly River – Cholera and Cover-up in Post-earthquake Haiti by Ralph R. Frerichs. Copyright © 2016 by Ralph R. Frerichs. Excerpted with permission by Cornell University Press.