Analysis of H5N1 cases finds sinking fatality rate

Aug 12, 2011 (CIDRAP News) – An analysis of global human H5N1 avian influenza cases over the past 5 years shows that the overall case-fatality rate (CFR) declined somewhat and that older children and adults were far more likely to die than were children under age 10, says a new report by German scientists.

The study, conducted by researchers from the Robert Koch Institute, Germany's national disease-control agency, also showed that the risk of death increased sharply with each day hospitalization was delayed and that 57% of patients were women, among other findings.

The authors compiled and analyzed publicly available information on H5N1 cases reported from September 2006 through August 2010. Their report was published online yesterday in Eurosurveillance.

"With this study, we show that data from the public domain yield important epidemiological information on the global AI [avian influenza] situation," they write. They add that the creation of a public "line list" of human H5N1 cases might enhance understanding of the disease and facilitate rapid detection of changes in its epidemiology.

Building a line list
The researchers created their own line list of cases by gathering information from public sources: the World Health Organization (WHO), the European Centre for Disease Prevention and Control, ProMED, and Reuters AlertNet. They looked at WHO-confirmed, probable, and suspected cases. The cases were grouped by countries that had 10 or more—China, Egypt, Indonesia, and Vietnam—and a remaining group of eight other countries with fewer cases.

The team identified 294 cases over the 5 years, including 235 confirmed by the WHO, 35 regarded as probable, and 24 suspected. The confirmed cases included 18 in China, 98 in Egypt, 82 in Indonesia, 25 in Vietnam, and 12 in the eight other countries.

The gender balance among confirmed case-patients was 57% women and 43% men, but it varied considerably by country, with larger majorities of women in Indonesia and Egypt than in the other countries.

The median age of patients was 18 but was significantly higher in women than men: 21 versus 14 years. Also, the median age differed widely by country, ranging from 6 years in Egypt to 23 in China. All but a few (96%) of confirmed case-patients with available information had had contact with potentially infected poultry.

Overall, 56% (132 of 235) of the confirmed case-patients died of the illness. Consistent with previous reports, the CFR was lowest in Egypt at 28% (27 of 98 patients) and highest in Indonesia, 87% (71 of 82).

Both the cumulative CFR and a rolling 19-month CFR calculated by the researchers declined over the 5 years. They don't report precise figures, but a chart shows that both the cumulative CFR and the rolling 19-month CFR were above 70% in 2007. By August 2010 the rolling 19-month CFR had dropped to about 25%.

The rolling CFR declined especially steeply from April 2008 to April 2009, when a major burden of cases was shifting from Indonesia, with the highest CFR, to Egypt, with the lowest, the report says.

"Using a 19-month rolling CFR, we found a clear decrease in case fatality, which persisted when stratifying for Egypt and Indonesia," it states. "It could thus not simply be explained by a predominance of Egyptian cases since 2009."

CFR increased with age
Children under age 10 were far less likely to die of the illness than were older children and adults, the analysis shows. Compared with the under-10 group, patients aged 10 to 19 and 20 to 29 were more than six times as likely to succumb. Adults aged 30 and up had almost a fivefold increase in risk, compared with children under 10.

The authors also found that in Egypt, H5N1 survivors averaged just 4 years old, versus 25 years for patients who died—a much larger difference than in the other countries. Also, the CFR in Egypt was significantly higher for females than males: 39% versus 12%.

Despite these findings, the researchers say the low overall CFR in Egypt can't be explained entirely by sex-related differences in CFR and the high proportion of children among Egyptian cases. They say that after adjustment for differences in these factors and in time to hospitalization, Egyptians still had the lowest risk of dying. This might reflect different H5N1 clades circulating in Egypt and in Asia and various other factors, they suggest.

In other results, using multivariable logistic regression, the researchers found that the risk of death increased by 33% with each passing day between symptom onset and hospitalization. The median time from illness onset to hospitalization was 4 days, and the median time from onset to death was 9 days. All patients who were not hospitalized until 8 or more days after onset died.

In conclusion, the researchers recommend that an organization such as the WHO establish and continuously update a case-based database for avian flu in humans. "Open access to analysable data might accelerate the identification and implementation of research questions and surveillance priorities and thus enhance our understanding of—still mostly fatal—AI in humans," they write.

Fiebig L, Soyka J, Buda S, et al. Avian influenza A(H5N1) in humans: new insights from a line list of World Health Organization confirmed cases, September 2006 to August 2010. Eurosurveillance 2011 Aug 11;16(32) [Full text]

See also:

Mar 22 CIDRAP News story on H5N1 cases in Egypt

Newsletter Sign-up

Get CIDRAP news and other free newsletters.

Sign up now»


Unrestricted financial support provided by

Bentson Foundation Gilead 
Grant support for ASP provided by


  Become an underwriter»