Skip to main content

Risk Factors for Injury and Death from a Tornado in Rural Bangladesh


Tornadoes are well studied in the United States, but the impact of these natural disasters is poorly understood in developing countries. Researchers from the Johns Hopkins Bloomberg School of Public Health and their local counterparts had a unique opportunity to prospectively document the devastation caused by a deadly tornado that struck rural northwestern Bangladesh on March 20, 2005. Affected communities were located within the 450 sq km region maintained by the JiVitA Project, a community research activity of Johns Hopkins Bangladesh (Ltd) that has been evaluating effects of nutrition and health interventions on maternal, infant and child survival, involving more than 100,000 pregnant women and 70,000 infants, in Gaibandha and Rangpur Districts since 2001. The study is available online in advance of publication in the journal Disasters.

On the evening of March 20, 2005, a tornado carved a path through the northern part of the JiVitA population site, destroying over 9,000 homes and impacting the lives of over 35,000 of the 650,000 people living in the area. According to the study, 10 percent of the population within the tornado-affected area was injured, and 56 deaths recorded. The most common causes of death were head injuries, bleeding or sepsis following wounds sustained. Most of the 3,710 injuries involved lacerations, bruises and minor cuts.

The elderly were almost 9 times more likely to die and 58 percent more likely to be injured than 15 to 24 year olds and women were 24 percent more likely to get injured than men. As preschool children were one quarter as likely to die and less than half as likely to be injured, it is plausible that mothers shielded their youngest children from the ravages of the storm.  Being indoors during the tornado was protective, but damage to the house caused injury to those indoors. Although homes made of thatch or bamboo were more likely to be blown away and destroyed, people living in tin or brick houses sustained more injuries if their homes were damaged. 

“Our study might be the most detailed investigation of mortality caused by a tornado in Bangladesh. These findings provide guidance for targeting the most vulnerable groups in an emergency response. Tin sheets, which many homes are made of, were shredded by the wind, resulting in extremely dangerous flying debris. Another concern was the high rate of secondary infections—20 percent of deaths were caused by post-traumatic infection—likely exacerbated by inappropriate first aid or by not seeking appropriate care for injuries,” said Alain Labrique, PhD, MHS, a lead investigator for the study and assistant professor in the Bloomberg School’s departments of International Health and Epidemiology.

Authors of the Epidemiology of tornado destruction in rural northern Bangladesh: risk factors for death and injury include Jonathan D. Sugimoto, a graduate of the Bloomberg School, now with the University of Washington, and Salahuddin Ahmad of the Royal Melbourne Institute of Technology University, and  Mahbubur Rashid, Abu Ahmed Shamim, Barkat Ullah, Rolf D.W. Klemm, Parul Christian and Keith P. West Jr. of the Bloomberg School of Public Health.

This research was supported by the Bill and Melinda Gates Foundation and the U.S. Agency for International Development with additional support from the Sight and Life Research Institute, the Canadian Agency for International Development (CIDA), the Nutrilite Health Institute and the Center for Human Nutrition.

Media contact: Tim Parsons, director of Public Affairs, 410-955-7619 or