I’ve spent the last day trying to answer this question, to add to a proposal I’m writing. Like my proposal will, the introduction of almost every paper on dengue includes an estimate of the annual incidence of dengue. Those estimates are usually in the range of 50 to 100 million cases. Only about half of those papers provide a citation, and the papers that are cited similarly make the statement in their introductions. Half of those statements are (again) uncited and the other half simply cite another paper that… Well, you see how it goes. Everyone says 50 to 100 million, but no one seems to know where the numbers come from. For the first few hours of my search, I felt like I was chasing my tail.

Eventually, I tracked down what I believe is the back-of-the-napkin estimate that started it all, the index case that started this epidemic of estimates. It’s a 1988 article written by Dr. Scott Halstead, whose introduction said: “Up to 100 million cases of dengue infection per year worldwide can be estimated from available data if one assumes there is an average annual infection rate of 10% for endemic areas, with most susceptible hosts being children” (1). In a letter to the editor a decade later, the same author wrote, “A number of this magnitude was first suggested in [the 1988 paper]” (2). Finally, in an interview two decades after the original paper, the author makes the same back-of-the-napkin estimate, but updated to “hundreds of millions” (3). Based on my research, I believe that this one author’s 25-year-old guesstimate that was created only to provide colour for his introduction is the most widespread estimate of the annual incidence of dengue.

But that estimate was for 100 million. Where’d the lower bound of 50 come from? I think I can answer that, too. There’s a WHO report on dengue prevention and control that has a single line buried deep inside: “Based on statistical modelling methods there are an estimated 51 million infections each year” (4). No methods. No discussion. It’s not important for the rest of the report and is never mentioned again. Just, simply, “51 million”. Since it’s an uncited and rather specific number, I suspect that someone actually came up with it for the report. I imagine a guy in an office shouting down the hall, “Hey Jim, that dengue report I’m writing needs a little flair. How many cases are there each year?” Jim would then grab a few numbers from Altavista, throw them in a haphazard linear regression, and come up with 51 million. Considering how little detail is provided, I doubt that the number was intended to do more than provide colour for the report.

Fifty to a hundred million. One is a guess and the other we’re supposed to take on faith. Both were created to spice up some drab text and were not intended to be real, scientific answers. And no one cites the original article or report where the numbers originated. I wonder, of the myriad papers that have been published on dengue in the past two decades, how many people tried to find the source of the estimate? How many people, like me, were left wanting? Why do these numbers persist, when they’re possibly wrong and definitely out of date?

Because the WHO says so on their “Media centre” website (5). (Uncited, of course.) If it’s wrong, researchers can say, “Blame the WHO copywriter’s sloppy research, not mine.” I’m tempted to do the same.

Epilogue

When discussing the global burden of disease, the obvious place to check is the WHO Global Burden of Disease report (6). I trust them. They aren’t throwing numbers around, and they aren’t just trying to spice up dull reports. What do they say, in their venerable tome? Nine million. Nine million cases of dengue a year, a number far, far lower than any in the literature. A number so low that even I have trouble believing that it’s accurate. Or maybe I’m just brainwashed, having seen “50 to 100 million” repeated so many times, and suddenly 9 million isn’t impressive enough. Maybe my proposal should say 50 to 100 million, to match the literature. That’s enough dengue that people will care, right?

Bibliography

1. Halstead, Scott B. Pathogenesis of Dengue: Challenges to Molecular Biology. Science. 1988 Jan 29;239(4839):476-81.
2. Halstead, Scott B. Is there an inapparent dengue explosion?. Lancet. 1999 Mar 27;353(9158):1100-1.
3. Interview with Scott B. Halstead, M.D.. Vector Borne Zoonotic Dis. 2007 Spring;7(1):99-105.
4. World Health Organization. Strengthening Implementation of the Global Strategy for Dengue Fever/ Dengue Haemorrhagic Fever Prevention and Control: Report of the Informal Consultation. Geneva: World Health Organization; 1999. Report WHO/CDS/(DEN)/IC/2000.1.
5. World Health Organization. Fact Sheet 117: Dengue and dengue haemorrhagic fever. World Health Organization; updated March 2009. Accessed 18 January 2012.
6. World Health Organization. The global burden of disease: 2004 update. Geneva: World Health Organization; 2008.