EDITORIALS

Asian tsunami: death-toll addiction and its downside

Michel Thieren

Team Leader for Core Health Indicators, Department of Measurement and Health Information Systems, Evidence and Information for Policy, World Health Organization, 1211 Geneva 27, Switzerland (email: thierenm@who.int)

"No man is an island, entire of itself;
... any man's death diminishes me"
— John Donne, 1624

Numbers on the move in the media are always addictive and a week-long succession of news flashes about deaths in a disaster makes them even more addictive. The initial death toll for the Asian tsunami went from a four-digit to a six-digit figure within a week: at the time of going to press it approaches 220 000 deaths (1), a figure unprecedented in recorded history. This increase triggered an equally historic sky-rocketing of donations. A week after the earthquake, some donor countries pledged up to 100 times their initial offer in what was termed "a political auction" (2). Public generosity has been so overwhelming that aid agency Médecins Sans Frontières was forced to decline further contributions. But when a number-hungry press shifts its audience's attention from the survivors to the deaths and from the deaths to the toll, there are also negative consequences.

First, the focus on death triggers one of the most common disaster myths — that dead bodies pose an imminent risk of epidemics and must be disposed of immediately. This idea is scientifically unfounded and no epidemiological evidence exists to support it. Bypassing the proper identification and ordered disposal of cadavers is public health malpractice; furthermore, "misguided action, such as mass burials, ... can add to the burden of suffering already experienced by survivors" (3). The result is scarce resources being diverted from the real concern: people who are wounded, sick, dispossessed.

Second, large totals associated with events of historical dimensions become history in themselves and, there is an accompanying moral responsibility to fully document such occurrences. Despite broad consensus on the current total, the death-toll addiction masks the fact that this figure may be inaccurate. No explicit audit trail — the set of crucial data behind the final number — is available, and the figure's credibility requires answers. Who is accountable? How was it estimated and computed? What is its margin of error? How are "deaths" and "missing" defined and do they overlap in the final count? Three weeks into the disaster, this information is simply not available. Meanwhile, a quarter of a million may soon be adopted as the legitimate death toll associated with the 2004 tsunami, just as 800 000 was quickly accepted for the Rwandan genocide. There are two contentious points here: the historical dimension of these figures makes them resistant to any subsequent scientific readjustment, while the absence of an audit trail makes them easy prey to revisionists.

Third, a death toll of historic proportions inevitably "sets the bar extremely high for future concern" (4). Rapid grasping at the largest number, even from understandable motives of attracting emergency resources, alters the ability to appraise the magnitude of future catastrophes if they fall short of the current tally. A year ago, over 26 000 lost their lives in Bam, Islamic Republic of Iran, in the worst seismic event in its recent history. This death toll might seem less significant today. As record-breaking statistics are announced, public perception adjusts so that quasi-apocalyptic events become "acceptable". In July 1994, 1 000 000 refugees fleeing Rwanda into Goma, Zaire, made world headlines. A month later, the 300 000 who fled into nearby Bukavu generated far less concern and the humanitarian operation was seen reassuringly as "business as usual". For those of us on the ground, it was anything but.

So should we refrain from publishing death tolls in real time? Not at all — they are valuable indicators of the human extent of a catastrophe. Although emergency environments make the collection of information difficult, scientific techniques exist to overcome such constraints. The techniques should be agreed upon, applied in a transparent manner from the beginning, and incorporated into the data audit trail. In this way, those who generate disaster-related statistics can be held accountable to the media and the public — and the victims. Mortality numbers quantify units: each unit is a human being, and a single death is one too many. For every death we count, there is an uncountable amount of suffering.

If death-toll addiction has mobilized large amounts of resources, it is to be hoped that in a few months, once the beaches are cleaned and the tourists are back, no fisherman will have to beg for a new boat to restore his livelihood. With up to US$ 8000 million already pledged (5) and an at-risk population of five million (another broadly disseminated un-audited figure) this would be, indeed, a waste of historic proportions.

References

1. Asia's tsunami death toll soars. BBC News World Edition, 20 January 2005. Available from: http://news.bbc.co.uk/2/hi/asia- pacific/4189883.stm

2. Debt forgiveness and the tsunami. Financial Times 5 January 2004. Available from: http://news.ft.com/comment/editorial

3. In natural disasters, cadavers pose no threat of disease. Washington (DC): Pan American Health Organization; Press Release 23 September 2004. Available from: http://www.paho.org/English/DD/PIN/pr040923.htm

4. Abramowitz M, Power S. A broken system. Washington Post 13 September 2004: A21.

5. Huit milliards d'aide internationale. Radio France, à la une de l'info. 11 January 2005. Available from: http://www.radiofrance.fr/reportage/laune/?rid=100000027