Sunday, January 22, 2006

Disturbing Infection Patterns in Turkey

Based on the Timeline published in Recombinomics: What's New I created the following chart of H5N1 infections in Turkey. Unfortunately it is a bit hard to read, but it can referenced against the text in the Recombinomics timeline (click on the image for an enlargement).

Red denotes known durations, or the onset dates of an illness when the duration is not known. In those cases when the end date of an individual's illness was not reported, I've used pink lines. A letter "D" at the end of a line indicates the individual died, while "R" indicates they recovered. If an individual's death was not reported I assume that they have recovered if a reasonable time has passed, but add a "?". I just use a "?" by itself if the timeframe suggests they are probably still in the hospital. Green squares note the dates of an event that is a likely point for the family to have contracted the infection.

The first key point to note is that the Ozcam family ate Christmas dinner with the Kocyigit family while the first Kocyigit son was ill but before the other children in that family developed symptoms(the upper greens square). Five days later they began to develop symptoms, fairly consistent with H5N1's normal incubation period.

The second point to notice is that the Kocyigit family caught the disease from birds, and had a severe mortality rate of 75%, very much in line with what we've come to expect from H5N1. Similarly the second Ozcam family caught the disease by slaughtering sick birds and had a 50% mortality rate, also consistent with other H5N1 outbreaks.

The first Ozcam family cluster is unique in two ways. One is that it is more than twice as large as any previous cluster we know of. The other is that not a single member of the Ozcam family died. Epidemiologists have long predicted that as H5N1 became better adapted to infecting humans it would become less lethal. The Ozcam family seems to be demonstrating this.

It isn't terribly obvious from the timeline above, but the linked clusters between the Kocyigit and Oscan families exhibit a disturbing pattern shown more clearly in the following graph.

Here the green line is the graph of an exponential growth function with a 5.5 day doubling period. Notice how closely the graph of the actual cases matches that function.

If the cases were caused by humans contracting the disease from an external source, then the trend should be a straight line as humans stumble into situations which lead them to be infected. It maight even tend to level off as humans learn to avoid the situations that lead them to become infected--which would be expected considering the massive education campagin and culling operations used in Turkey. Instead the cluster exhibits the classic exponential growth curve characteristic of unfettered population growth: 1 person infects 2 people who infect 4 more people, who infect 8 more people, and so on. It looks like a doubling rate of a little more than 5.5 days.

This suggests that the disease was not contracted by birds but was starting to be passed human to human in the ever expanding numbers that we would expect to see during a pandemic. What is hopeful is that quick hospitalization may have kept the virus from spreading outside the family. However even if disaster was averted this time around, the danger has not passed. If the strain of H5N1 currently circulating in birds has leaped the gap to sustained human to human transmission once, it will do it again, and again. We have to be lucky every time. H5N1 only needs to be lucky once.

Tuesday, September 27, 2005

H5N1 Spreading in the Human Population

Most people know we are in danger, but these sites make it clear just how much danger. In addition to the possibility of H5N1 simply continuing to evolve into a major killer there is also a steadily increasing danger that the more it spreads the more chance it has of infecting a human who is infected with a human flu. The mixing of two flu strains within the cells of a single organism could cause H5N1 to suddenly gain the ability to become a panemic. The biggest danger would be if there are a large increase in H5N1 cases this flu season, which would seem likely considering there is a large increas in the prevalence of other flu strains at this time. Because there would be more H5N1 cases and more human flu cases, the chance of a single person getting both forms of the flu at the same time would be dramatically increased.

Though the danger of such an event occuring has been with us ever since H5N1 reappeared in 2001, the danger is particularly acute this year because contrary to what is publically reported by the WHO, human to human transmission of H5N1 Avian flu has become quite common. And although the WHO grudgingly admits that H5N1 can be transmitted from a person who has been infected by chickens or other foul to another person, but denies that it can be transmitted beyond the first person infected, this may also be false.

Why the lack of accuracy? The simplest explaination is simply lack of data. Not every case gets tested and even when tests are done our weak immune response causes a lot of false negatives. Then again there are positive tests that are not counted because a second corroborating test wasn't performed. For example the official data records no cases of H5N1 in Thailand during 2004, yet there were 12 cases with positive tests for H5N1 that didn't make it into the official numbers. That article also lists multiple cases which due to lack of laboratory testing are not included in the numbers and which appear to indicate human to human transmission that is being reported not to exist.

Then there are cases which were not even initially thought to be flu related. This was due to causing meningitis-like symptoms. Four people attending the wake of a fifth person all died within a span of three days aproximately 1 1/2 weeks later even though they had returned home after the wake to separate cities. At the time H5N1 was not known to cause meningitis-like symptoms without apparent respiratory symptoms. Now it is known to do so. Since none of the victims from the wake were tested we cannot know whether or not this was caused by H5N1. However there are basically only two possible culprits. H5N1 would only have to gain the ability to transmit as efficiently from person to person as it transmits from chicken to checken. Meningitis would have to gain the same ability (and is less likely to do so since it has never been an easily spread illness), and would also have to develop a much higher case fatality rate. Note that the 5 fatalities in this cluster was 2 1/2 times the entire number of cases of Meningitis in the country that year. It is very unlikely that this was caused by Meningitis, thus making a strong case that H5N1 caused the deaths, and thus indicating that it can transmit from human to human with much greater ease than health officials have been willing to admit.

The danger posed by the meningitis form of H5N1 is particularly accute because it is operating completely under the influenza monitoring system's radar. Everyone is looking for a respiratory flu, not meningitis. Furthermore because Meningitis has symptoms that are less obvious than the respiratory symptoms that we associate with flu, others nearby are less likely to avoid contact with infected persons, and people who are infected are less likely to make changes in their behavior that would reduce their contact with others, like calling in sick from work. Thus a meningitis form of flu would be likely to spread more easily than a rispiratory form.

Once sustained human to human transmission begins, it will quickly accelerate. If it is occuring now, it is quite sporadic. However before now H5N1 couldn't improve its ability to transmit from human to human much because the virus died with its victims, so every human to human transmission was an evolutionary dead end. In a concentrated population even if the ability to transmit from human to human multiple times were weak, it would still occur. Thus the viruses able to transmit only once would die with the first victim, those which could transmit only twice would die with their second vicetim, etc. Those viruses that could pass on to new victims indefinitely would be able to survive and possibly be communicated into the community at large. Furthermore those viruses that could transmit from human to human most easily would infect more victims, thus forming a larger portion of each generation than those that could only transmit from human to human with difficulty. This would continue until the most virulent strain was the dominant strain of the flu in the human population. If those viruses were somehow able to re-enter the avian population, for example by a scavenger bird eating flesh from a corpse that had died of H5N1, they might then be able to spread even further.

Changes like these occur because the genome of any Influenza strain is in a constant state of change. This is due to the process of recombination, which occurs continuously as flu viruses infect animals or people who are already infected by another flu virus, and pick up genes from other strains. Viruses also mutate, and then those mutations are added back into the mix. Again, as H5N1 continues to spread, the number of these recombinations will continue to increase. Also as human flus become more common during the flu season, the chance that this mixing will occur between H5N1 and a human flu increases.

Tuesday, September 20, 2005

Official Death Predictions Grossly Conservative

Predicting exactly how many deaths a pandemic could cause is tricky. There are a lot of variables to take into account, and guessing which ones would come into play, and how much impact they would have is beyond my ability. However rough estimates can be made based on readily available information. These estimates are not exact, and could be off by a lot. However expecting them to be off as far as the official estimates is an unrealistic streach.

Most of the numbers I have seen quoted look like rehashes of the so-called Spanish Flu of 1918, or are even more conservative than that. This seems preposterous considering that the world population today is almost 4 times what it was in 1918. So even if all we were dealing with was a re-enactment of 1918 we should be expecting quadruple the number of deaths. Then there is the little fact that the spanish flu killed less than 10% of those it infected while H5N1 kills at least 50%. Taking this into account as well a basline figure should be about 20 times the spanish flu's death toll world wide. Since the Spanish flu left 20-50 million dead in 1918, we could expect H5N1 to take 400 million to 1 billion lives, but recent data suggests that the mortality rate for H5N1 infections has increased, so my numbers could be conservative. All of this is assuming little or no vaccination or effective use of antiviral drugs on a global scale. Some will certainly be used by the wealthier nations, but many nations have no capability to produce vaccines, and the most effective antiviral drugs are in short supply. Even if antiviarals are used, they seem to be of only limited effectvieness, and the virus is likely to become resistant fairly quickly, as was seen with the Chinese use of antivirals to control the disease in their chicken populations.

In 1918, America saw between 500,000 and 675,000 people lose their lives to the Spanish flu. America's population is about 2.5 times what it was in 1918. Multiplied by 5 for the increase in mortality exhibited by H5N1 versus the Spanish flu, and we should expect 12.5 times what our loses were that time around, or between 6.3 million and 8.4 million deaths if a vaccine could not be produced before the end of the flu season. Another way of evaluating this number is that 10-20% of Americans contract the flu in a given season, between 27.5 million and 55 million people. Assuming that 1/3 contract the H5N1 strain, and half of those die would suggest between 4.6 and 9.2 million deaths. Agian if H5N1's mortality rates have indeed increased as some experts are claiming then these numbers could be very much on the conservative side.

Death tolls this high in America are extremely unlikely however. About the only chance of them occuring would be if H5N1 managed to break out into the global human population as a full-fledged pandemic virus between now and the start of the flu season. This is quite unlikely based on current trends. If it doesn't break out until spring, then even if the standard schedule for produceing a vaccine were followed with no notable accelleration most of those lives would be saved, since the greatest majority of cases would occur during the flu season. I estimate that each month leading up to the beginning of the flu season would see 0.2% of the total deaths that could be expected from the disease. When the flu season hit, this number would jump dramatically. The aim of the standard schedule for producing the vaccine is to have the vaccine ready before the flu season starts, thus preventing the worst effects of the jump in cases that could be expected at that time.

The problem with H5N1 is that (assuming the worst case without vaccination would be about 7 million people dead in the United States) we could expect about 14,000 Americans would die for each month before the vaccine was ready, compared with up to 2 million a month once the flu season hit. Depending on how long it took to get vaccine production up to speed we are talking 50,000-100,000 dead. If it were impossible to avoid those deaths that would be one thing. If a new strain emerged and we had production capacity ready but needed to wait until a vaccine could be created, that would be understandable. If the only reason those people had to die was because we didn't prepare ahead of time only because we lacked the will to do so, that would be unforgivable. For our lawmakers to be willing to waste billions on pork projects but be unable to justify the spending needed to prepare for this epidemic would be beyond contempt.

Each of us has a nearly equal chance of falling victim, though admittedly those of us in coastal cities have the greatest risk, both because the flu is more likely to reach us before a vaccine is ready, and because the density and mobility of a large urban population gives the virus an ideal area within which to spread. If the harm to our fellow citizens, and the cost to our nation in medical bills and lost work was not enough we should also consider that some of those killed will undoubtedly be friends, relatives, perhaps close family, or even ourselves.

Sunday, September 18, 2005

I am gratified to learn that Bush has aparently been woken up at least a bit. A $100 million contract to purchase vaccine is a good start, with a target of a 20 million dose stockpile. I'm wondering when the completion date for that stockpile is. I'm also wondering just how much good a 20 million dose stockpile of a vaccine that requires two doses for full protection will do in a nation of 275 million people.

First of all you need two doses because the entire world population is immunologically naive. That means we have no resistance to the disease because our immune systems have never fought it, or anything like it, ever before. If you get this year's flu your body has seen similar flus before so it isn't a big deal. In 1918 a really nasty version came out that was nicknamed the Spanish flu because Spain was the only nation in the western world that hadn't buried their mass media's head in the sand due to WWI, but that's another story. It was pretty nasty but it killed well under 10% of the people it infected versus H5N1's avarage of about 50%. Why the difference? The 1918 flu started in rural Kansas from a human virus that had traded genes with some other animal virus. Becuase of that we had a fair degree of immunity to it. When it hit indigenous populations that didn't have immunity to the common flu, the death rates were similar to H5N1.

This weak immune response may be partially to blame for the fact that most victims who are tested for H5N1 initially test negative, only after the disease progresses greatly has their immune sytem produced enough antibodies to test positive. If a second test isn't done, the case is officially considered a negative. For example in Indonesia a government official's daughter got sick, aparently with no known contact with poultry since they all lived in a wealthy suburb. Then about a week later her sister gets sick, and a week after that their father gets sick. All three die. Only the father tests positive. Only the father is considered an official H5N1 death. A rational explaination would be that the first daughter passed it to her sister, and either she passed it to her father or her sister passed it to their father, thus indicating they are all victims of H5N1, and that at the very least it infected one and possibly two people after the first. Most likely the first daughter caught it from another person since there are no poultry around for her to catch it from. But since only the father is a confirmed case officially there is no human-to-human transmission, and the official suggestion is that he probably ate contaminated chicken somewhere. This case is not unique. It is the pattern found in most family clusters even when contact with poultry is present. One family member gets it, then some time later another does, and then another. This suggests that the people are catching it from each other, not all catching it from the poultry. But officially this is not considered human to human transmission and the World Health Organization classifies the disease at stage 3 on the 6 stage trip to becomming a pandemic. In reality it is at least stage 4. It might even be stage 5, as the case in indonesia suggests since the first daughter seemed to get it from another person raising the possibility that it is circulating in the population but only sporadically infecting people.

Back to my concern with the 20 million dose plan. My main concern is that 20 million doses doesn't force the creation of a great deal of new production capacity. We cannot stop production of the seasonal vaccine until the pandemic strikes. This is simple reality based on the need to continue protecting our citizens from the seasonal flu that kills 36,000 Americans each year even with modern vaccination programs that usually reach about half of the people who should be taking the vaccine. If we stop normal vaccination we can expect at least 70,000 deaths. Not a good idea. Yet when the pandemic strikes, if it is not controlled by quarantines it could kill thousands each month until the vaccine is produced. 20 million doses would be able to do two things: protect medical and emergency officials and give some hope to those closest to an outbreak.

If a single dose could provide full immunity, the vaccine could be used to stop the disease or at least slow it greatly. By vaccinating the population in an area of an outbreak, we could deny the disease a population to spread in. With two doses needed, the vaccine probably would not be able to stop an outbreak because it doesn't provide full immunity. If a person gets H5N1 after being vaccinated once they will still get sick but will probably survive, but the disease would still be able to pass on to others. So ultimately until the entire population receives at least one dose we will have a major crisis on our hands. Once that has been done only the most vulnerable would be at grave risk of death, and an outbreak could be contained by giving a second dose to those in the vicinity.

I believe that we need to create enough production capacity to provide at least 275 million doses, and to do so as quickly as possible, to allow time to produce additional doses to double up the dosing for medical and emergency personnel, and people close to an outbreak. This would do two things. First it would create a stockpile large enough to vaccinate everyone once, thus allowing control of an outbreak with a second dose from even a small additonal stockpile. Second it would force the creation of a great deal of excess production capacity. This is important because it is entirely possible that H5N1 will change between now and the time it becomes a pandemic. Having that extra production capacity would mean that a new vaccine could be produced quickly if the need arose. At present it would take 6 months to produce a new vaccine. Antiviral drugs are only partialy effective against the disease, and even then only if used soon after infection. This means that thousands would be dying every month--perhaps even thousands every week--while we wait for a vaccine. Even if it does change, it is likely that a dose of the vaccine would still be useful. Even if it only helped lessen the enormous disadvantage our immune systems face it would help.

It is easy to become complacent, simple to imagine that the world will always be the way it is today. I remember first hearing of the highly pathogenic strain of avian influenza known as H5N1 in 1997, when it appeared to have been contained in Hong Kong. I was worried but not truely frightened. I heard about it again in 2001 when it was making a comback, but September 11 drowned out the news. I've largely ignored it for years but it has always been hanging in the back of my mind.

Similarly when Katrina hit Florida I didn't pay it much attention. It did some damage, but not that much compared to other hurricanes, and I assumed it was just going to peter out like most do. Then I heard it had built up to Level 5 with 175mph winds and a 20-28 foot storm surge. I heard that Nagin was evacuating New Orleans and Bush was... going on vacation? Good thing Nagin didn't wait for a call from the president before evacuating the city this time.

At the time I couldn't believe it. I was pulling up publically available websites detailing the damage that a level 3 hurricane would do if it hit the city directly, (and was pretty close to what Katrina's level 4 near miss did) along with comparisons of what a Level 4 or 5 hurricane would do on a direct hit. The lack of concern at the top levels of govenrment seemed mind boggling. I assumed there must be behind the scenes work going on. In the aftermath, I must conclude there was not.

Katrinia was a wakup call. I started wondering what was being done to prepare for the other disaster I had been worrying about for so long, and started researching the disease in earnest. I've learned quite a few disturbing things since then. One is that H5N1 has been growing a lot closer to becomming a pandemic over the last few years. The second is that the World Health Organization is severely downplaying the reality of the situation, at least publically. The third is that mass media has gotten into the habit of parroting what official sources tell them, instead of investigating for themselves.

This might not be a big deal for a dictatorship. A dictator might wake up one day and decide to start producing vaccine, provided his nation could afford to. He might also just decide to build a palace to hide in instead... However, America and most of the other wealthy nations of the world are democracies. And although our elected officials don't always do what we want them to, they do tend to worry about going out on a limb and getting cut off at the ballot box. As such they tend to be conservative in their actions regardless of their political affiliation. And they tend to go on being politicians until they absolutely have to step up the the bar as a leader, because being a politician is the safe thing. Leading means taking risks.

Because of this the nations of the world fall into three categories when it comes to responding to H5N1. Dictatorships which cannot afford to produce vaccine against the disease, democracies that haven't produced vaccine against the disease, and China which has produced 2.6 billion doses of vaccine.

Why is this? I can think of two possibilities. One is that the World Health Organization doesn't know what is going on and actually thinks that human-to-human transmission is very rare and only occurs in cases of prolonged contact. The other is that they know that human-to-human transmission is common but use unrealistic definitions of "confirmed" cases to keep the publically released death toll low and to justify denying that sustained human-to-human transmission is occuring, while communicating the truth to governments. This prevents a panic, but unfortunately for those of us living in rich democracies, that means that our governments are hamstrung by the lack of public concern. They are beginning to respond to either behind the scenes communication from the WHO, or to a dawning realization that the WHO is hindering the disemination of knowledge more than it is helping. Either way I fear it may be too little to late.

China on the other hand has been dealing with H5N1 longer than any other nation. Unofficial reports are that it surfaced in China in 1996, a year before showing up in Hong Kong. And they have been having just as many outbreaks as everyone else, despite the lack of publically released information. Unlike the rest of the world though, they have not controlled the disease by culling diseased chickens, but by feeding their flocks the antiviral drug amantadine. This has led many strains of H5N1 to become resitant to the drug. However unlike the rest of the world China has been on a crash course to preparing for a pandemic. They are now vaccinating their chickens instead of using amantadine, and the presumably have the capacity to vaccinate their population as well. It would be a sad testimony to the state of our national will if China was protecting their people with vaccine while our were dieing while we tried to produce enough vaccine.