Hepatitis B and China’s H7N9 Crisis

Posted on by Laurie Garrett

The H7N9 flu outbreak in China has offered a long list of puzzles, many of which are the key to understanding whether this Chinese epidemic will erupt into a global pandemic. Among them is the very odd age distribution of the flu cases: It’s an overwhelmingly elderly male phenomenon in eastern China.


This is not normal for influenza – particularly varieties of flu that spread from birds-to-people. The H5N1 avian flu, which has spread among birds and sometimes humans since the mid-1990s, has an opposite case distribution. Most of the 622 cases – 371 deaths – caused by H5N1 since 2003 have been in children. Why? Because the youngsters play around family chickens, ducks and other poultry that may be infected with the virus.

Even “normal flu” – the sort that circulates every year – will afflict and kill people in a different pattern, skewing towards over-70-year old adults, roughly equally by gender.

There simply is nothing normal about this new Chinese H7N9 influenza.

There are two very different ways to interpret this. First – and most likely – the skewing towards older men may offer a clue to what animal or bird species is the host of the virus, from which the men are acquiring their infection (if not from other people).  Or secondly – the older men may be especially vulnerable for some reason. As WHO put it: "It may be due to an imbalance in exposure of the elderly [to H7N9], or physiological factors related to aging, such as decreased immune function."


Analysis of the first 82 cases in China offers some vital clues – at least for the 77 individuals the Chinese CDC was able to interview before they passed away.  A handful of the individuals had professional contact with animals, as poultry workers, butchers or chefs. The remainder had some contact with chickens or chicken meat, and a few other avian species.  The numbers point to chickens.

But women prepare meals in China, pluck the chicken feathers and handle food in the home – if household chicken exposure is the key to infection, the gender distribution of H7N9 cases makes little sense.

 Source:  Li, et al.

Source: Li, et al.

One behavior that I have observed among older, especially retired, men in China involves pet birds, typically of exotic breeds. Many types of birds are thought to bring good luck to their owners, and older, more traditionally-oriented men take great pride in possessing beautiful ones that sing lovely tweets. A common sight in parks and plazas are clusters of men that gather to chat or play mah jongg, and hang their caged birds side-by-side from trees or fences.  The sound from these dangling bird cages can be enchanting.  Though the new New England Journal of Medicine study found just one patient acknowledged having such a pet birds, it is possible indirect exposure in social settings could have put men in contact with a previously obscure virus, carried by an exotic bird species. It’s a long shot. But when the obvious answers fail, more obscure explanations merit attention.

 Source: Reuters

Source: Reuters

Because the men often compete to have especially exotic varieties that sing complex melodies, there is a vibrant trade across Asia in the capture and sale of wild tropical birds.  Along the largely black market route from capture in the wild to eventual “retail” sale, there are hundreds of contact points between smugglers, traders, the birds, and other smuggled wild animals.

The urban delight in caged birds is an overwhelmingly older male pursuit. Well do I recall in 2003 seeing groups of men sitting in parks around Beijing, their faces masked and hands gloved to prevent spread of SARS between them, while their caged birds happily chirped from nearby trees.

According to the Beijing CDC, some type of bird contact has been identified for about 60 percent of the H7N9 patients, mostly in the form of poultry processing or purchase. That leaves 40 percent unexplained. Moreover, none of the patients that had some form of contact with chickens, ducks, quail or pigeons seems to have been engaged in activities that vary much from those followed equally by women.


China has the highest hepatitis B infection rate in the world; 65 percent of the global burden of the viral disease is in China. It is estimated that 350 million Chinese have hepatitis B, and 170 million harbor hepatitis C. And hepatitis is a leading cause of death in China. According to the World Hepatitis Alliance the majority of these infections were the result of use of unclean needles and equipment in Chinese medical settings, including traditional clinics that practice acupuncture and other invasive procedures.  There is no hepatitis C vaccine, but the hepatitis B immunization costs less than $1 for three doses, and is highly effective.  China has a national campaign underway to vaccinate babies, but adult immunization remains rare in the country.


Though clinical details on all of the H7N9 patients have not been released, there are indications that most of the deceased and those in serious condition have underlying chronic conditions, including hepatitis infection.  

A recently published survey of hepatitis B prevalence in southeastern China found infection strongly skewed towards older males, particularly aged 40-70 years. For example, nearly 15 percent of all men surveyed aged 60-69 years were hepatitis B-positive, compared to 10 percent of females in the same age group. Neither men nor women were likely to be infected if they were under 20 years old, but even in that group male prevalence was more than double the females (3 percent versus less than 1 percent of females).  A 2004-2010 survey of all types of hepatitis in Hubei Province found male incidence in all age groups was two-to-three times greater than female.

There are at least four reasons why Chinese males are more likely to be infected with hepatitis. First, in a culture that both prizes male children above females and until recently heavily penalized families that had more than one child, parents treasure little boys, and spare no expense to keep them healthy.  Boys are more likely to see doctors and traditional healers during their childhoods, and therefore be exposed to contaminated needles.

As the boys grow to be men, some will engage in anal intercourse with other men – a behavior strongly correlated with hepatitis transmission.  Third, desperately poor Chinese sell their blood for cash in procedures that are notoriously non-hygienic, and spawned the country’s shameful 1990s HIV epidemic in blood donors. Men are far more likely to sell their blood, and to acquire hepatitis through the non-sterile procedures. And finally, men are far more likely in China to inject narcotics and other illicit drugs, using shared needles and syringes – a behavior highly associated with acquisition of hepatitis infection.

Chronic hepatitis infection (of either B or C) increases individual’s vulnerability to a long list of diseases, including cirrhosis and cancer of the liver, and overall lowers life expectancy. The deleterious impact of chronic hepatitis increases over the length of infection, so that older individuals are more likely to suffer various symptoms, including declining immune system function and cancer.

During the 2009 swine flu pandemic there was concern that individuals suffering chronic hepatitis infection were at greater risk for severe flu. Damage to the liver – which may not always produce symptoms the patient recognizes – can weaken the body’s ability to fend off diseases, overall.

If either the pet bird behavior of older Chinese men, or their high prevalence of hepatitis infection turns out to correlate with risk for H7N9 infection and death it could suggest that the epidemic will remain in Asia, at least for the near future. Hepatitis prevalence is high in much of Asia in men – far higher than anything seen in Europe or the Americas. And while many in the West may possess caged birds, the social ritual of men’s gatherings and exposure to the avian exotica is unique to China and a few other Asian regions.

The University of Toronto's David N. Fisman posted on April 25 to ProMED his statistical analysis of H7N9 cases reported as of the previous day, concluding, "It can be seen that there's a marked increase in risk in males over 54, and in females over 64, relative to younger individuals, while female risk is substantially lower than risk in males."

He argues -- and I agree -- that the elder age distribution could reflect disease severity in older populations, and large numbers of younger, more mildly affected men and women are getting missed by Chinese investigators. I have argued for weeks that the specificity of the high-tech RT-PCR methods deployed by Chinese investigators is leading to underestimates of popuation infection because the tests find fragments of H7N9 RNA. Finding RNA is terrific -- irrefutable evidence of infection, to be sure. But if a healthy individual was infected in the past, and cleared the virus from his or her body by mounting an effective immune response, no viral RNA will be present in blood or sputum samples. To find those cases, cruder antibody serology tests need to be executed on a fairly massive basis.

As Fisman noted in his posting, the so-called CFR (case fatality rate) for H7N9 now appears to be a startling 21%, but that is based on the current odd pool of roughly 130 patients, most of them elderly and severely ill. The Fisman chart shows what happens to the assumptions about fatality rates for this virus if large numbers of asymptomatic or mildly infected individuals are discovered:

 Source:  Fisman Blog

Source: Fisman Blog

If with continued and expanded epidemiology it turns out the strange prevalence of H7N9 infection and death is, genuinely, skewed towards elderly men, and the numbers of identified cases are truly representative of the scale of the epidemic, the above options, and others, must be considered to determine why old men are so vulnerable. As a recent Chinese study shows, it is premature to judge the age, gender, geographic, or CFR distribution of H7N9, even focused on a single province of China, and the implications of a broad shift in the demographic picture are quite dramatic. Wider seroepidemiology and antibody tests may reveal, however, that the age and gender distribution are canards, early epidemiological artifacts.

We await further data.