What Can Go Wrong
Total Page:16
File Type:pdf, Size:1020Kb
The Dangerous Business of Aid A Report on the Risks to Aid Workers in the Field October 2, 2009 Prepared by: Prepared for: Noel Bauer, MA Melissa Thomas, JD, PhD [email protected] [email protected] Associate Professor of International The author’s views expressed in this Development publication do not necessarily reflect the The Johns Hopkins University views of the Paul H. Nitze School of The Paul H. Nitze School of Advanced Advanced International Studies or Johns International Studies (SAIS) Hopkins University. 1740 Massachusetts Ave., N.W. Washington, D.C. 20036 Cover photo by: Pierre Holtz for OCHA, Creative Commons License 2.0 The Dangerous Business of Aid: A Report on the Risks to Aid Workers in the Field Introduction Over the past decades, a steady flow of articles have warned of the dangers to aid workers, but decried the lack of rigorous study. However, in the last several years various investigations have provided good data on a number of threats to aid workers. Although much remains to be done, there is growing evidence to prove that aid work is an increasingly dangerous business. Increased awareness of the risks to aid workers comes at a time when the field is growing rapidly. A study by the Overseas Development Institute (ODI) shows that the humanitarian worker population alone has more than doubled from 136,204 in 1997 to over 290,000 in 2008 (Stoddard, Harmer, and Haver 2006; 2009). The purpose of this report is to give the International Development Department of Johns Hopkins University’s Paul H. Nitze School of Advanced International Studies (SAIS) a broad overview of the risks to aid workers in the field, and in doing so allow them to better prepare their graduates for life in the field. The report first addresses trends in overall mortality and then focuses on specific risks including accidents, mental health risks, physical illness, sexual assaults, and violence. The violence against aid workers is further analyzed to identify environmental and individual factors of risk. Notes on the data Although recent studies have greatly expanded awareness of the risks to aid workers, there is still a scarcity of rigorous research. Many studies suffer from small sample size and a host of biases. With few exceptions, the studies do not provide convincing baseline data and thus likely suffer from significant selection bias. Investigation into the risks of aid work makes one thing clear, that the lack of concrete information makes any conclusion partially guesswork. Recent initiatives, such as work by the Overseas Development Institute to measure violence against aid workers, represent a promising step in the right direction. However, sustained investment in research and complementary initiatives to monitor other risk factors will be necessary before a complete and accurate picture emerges. Notes on terminology This report uses the term aid worker to refer to both humanitarian and development personnel. If a study addresses only one of these groups, they will be identified as such. The report wishes to recognize the important distinction between the two groups. However, it must also be noted that development workers are being increasingly asked to work in post-conflict or conflict settings. New political priorities and organizational structures, such as Provincial Reconstruction Teams in Iraq and Afghanistan, are blurring the traditional lines between humanitarian and development assistance. - 1 - The Dangerous Business of Aid: A Report on the Risks to Aid Workers in the Field Overall Mortality Several studies have examined mortality among aid workers. Their conclusions of overall risk tend to vary according to the population they study. While it seems some risks have increased over time, a definite picture of how overall mortality has changed in recent years remains elusive. In looking at fatalities among 20 to 29 year-old Peace Corps volunteers from 1977 to 1983, Stephen Hargarten and Susan Baker found that the overall mortality rates for volunteers was not significantly different than that expected for the general population. However, they admit that volunteers’ higher education and verified strong health should lead to lower death rates. (Hargarten and Baker 1985, 1328) Additionally, it should also be noted that Peace Corps avoids operating in countries perceived as dangerous or that suffer from high levels of violence. Of the seven countries responsible for three-quarters of all violence against aid workers as reported by Stoddard, Harmer and Haver (2009, 4), only Chad has had any volunteers in the last decade (Peace Corps 2008). In contrast, a 1995 study by Erik Schouten and Martien Borgdorff found elevated mortality rates when examining the records of three large Dutch development organizations from 1984 to 1994. Over this time more than 6,500 development workers and their spouses spent a combined 15,144 years abroad. Seventy-five percent of these personnel lived in sub-Saharan Africa. The study reveals that Dutch development workers had a mortality rate 1.9 times greater than the general Dutch population. In addition, they point out that the figure probably understates the increased risk of mortality for three reasons: 1) the “healthy cohort” effect, meaning only healthy development workers are sent overseas; 2) on average, development workers are more educated than the general population, meaning they should have a lower mortality rate; and 3) development workers may die at a later date due to risks incurred during field work, such as contracting malaria or AIDS. (Schouten and Borgdorff 1995, 1343) The first examination of mortality among humanitarian workers came from a retrospective study by Mani Sheik, Maria Isabel Gutierrez, Paul Bolton, Paul Spiegel, Michel Thieren, and Gilbert Burnham that covered 1985 to 1998. The team examined the record of various non- governmental organizations (NGOs), the Red Cross, UN programs, as well as UN peacekeepers; 32 organizations provided data, but many provided only partial information. The study was not able to establish a baseline for humanitarian workers in the field and so did not provide a figure for overall mortality. The sample of organizations in the study shows a clear increasing trend in humanitarian worker deaths from 1985 to 1998, largely driven by a sharp peak during the Rwandan genocide in 1994. However, without baseline data, definite conclusions cannot be made. (Sheik et al 2000) A more recent study by Elizabeth Rowley, Byron Crape, and Gilbert Burnham from Johns Hopkins University examined overall morbidity and mortality among aid workers between September 2002 and December 2005. The study surveyed 20 humanitarian organizations, of which 17 ran a combination of relief and development programming. Only 3 organizations had purely humanitarian objectives. The organizations were all sizeable, with only 5 employing less than 1,000 field staff. Of the 20 organizations, 18 submitted sufficient data to be included in the - 2 - The Dangerous Business of Aid: A Report on the Risks to Aid Workers in the Field final analysis and only 13 submitted sufficient baseline data on workers present in the field to permit rate calculations. Although the study boasts global coverage, within the 13 organizations that provided staffing information, 70% of expatriate staff and 67% of national staff were located in Africa. The study found a mortality rate of 60 per 100,000 (Rowley 2009). However, the authors stress that the figure is at best an estimate due to data inconsistencies and the small number of overall fatalities. (Rowley, Crape, and Burnham 2008) Unfortunately, the most exhaustive study thus far, performed by Abby Stoddard, Adele Harmer, and Katherine Haver, only examines violence against aid workers and neglects overall mortality. The study, which is explained in greater detail later in this report, determines the average mortality rate due to violence from 1997 to 2008 to be 28.3 per 100,000 aid workers. However, this figure masks recent upswings. The rate for 2008 alone is 42 per 100,000. The report demonstrates an exponential rise in violence against aid workers in the past three years. Thus it would seem that the rate and composition of mortality have recently changed drastically. However, there is currently no hard evidence to confirm or deny this assertion. (Stoddard, Harmer, Haver 2009) Comparative Occupational Mortality The scarcity of data prevents a direct comparison between aid work and other professions. However, the cursory comparison below between aid workers and the most dangerous jobs for Americans as measured by the U.S. Bureau of Labor Statistics suggests that aid work is among the most dangerous. The second table shows that aid workers also face a relatively high risk of meeting a violent end. - 3 - The Dangerous Business of Aid: A Report on the Risks to Aid Workers in the Field Composition of Aid Worker Mortality Accidents Once it was common wisdom to say that more aid workers died in car accidents than any other way. Recent studies suggest that this is no longer true, if indeed it was ever. Accidents of all types, and especially traffic accidents, remain a significant risk to aid workers in the field. However, the rise in violence-related death in recent years appears to have reduced the percentage of accident-related fatalities. In their study of fatalities in the Peace Corps from 1962 to 1983, Hargarten and Baker found unintentional injury accounted for 70% of all deaths. Traffic accidents were the number one cause of death, causing 36% of all reported fatalities and 52% of unintentional injuries (Hargerten and Baker 1985, 1327). Unfortunately more recent comprehensive Peace Corps data is not available. In examining deaths among Dutch development workers from 1984 to 1994, Schouten and Borgdorff found that traffic accidents made up more than a third of all reported deaths. Furthermore, the rate of traffic related deaths was 10 times greater than that of the general Dutch - 4 - The Dangerous Business of Aid: A Report on the Risks to Aid Workers in the Field population.