Computational Epidemiology

Hmmm.... well, I may have at least found the general term for the field that I am interested in. That would be Computational Epidemiology. Suffice it to say, when someone stands up in front of a group of medical school faculty and basically say, "I want to do academic research on infectious diseases, but I want to use a computer as my laboratory, and I could really care less if my research ever involves any direct patient contact; but ... um... I don't really know what that field of research is called... and, no I have no prior training in any of this (I went to medical school, duh!)but never mind that, give me some time and I'm gonna figure this out!", you get some strange looks.

Some Thoughts on Deficiencies in our Understanding of Epidemics

I'm sitting here trying to think of good research ideas to focus on; the areas of interest to me are (1) Infectious Disease, but specifically areas of epidemiology related to those infections that cause massive morbidity and mortality over short periods of time. So, this covers what is typically called "Emerging" and "Re-Emerging" infections as well as agents typically considered agents of biowarfare/bioterrorism in addition to a few infections that we constantly risk the evolution of a repeated outbreak (such as pandemic influenza).

Given the areas of specific interest I have, I also like to consider the sorts of methodologies which I either (1) have some basic skill in, or (2) would very much like to learn much more about. Of course, as a physician, this includes the clinical acumen to some degree, but I find myself more drawn to an interest in information technology and computational technologies to solve old problems which we still struggle with.

So, then, what sorts of problems do we find that we still need to address? And what sorts of questions or problems could one solve that is likely to make a significant impact to the world at large?

Well, what I am feeling that I would like to learn much more about is the computational modeling and simulation of epidemics. For one, I wonder if our understanding of the basic modeling of epidemics is still insufficient. For this, I will need to learn much more about the current field and then try to identify areas which are still lacking.

In addition, I believe that in the past, we have not had sufficient technology to fully incorporate the multitude of factors (human, societal, environmental, microbiological, political, etc...) that come in to play. For this, of course I need to learn much more, but I think it is much safer to say that this is likely to be the case. There is much technology that is only in its infancy which is applicable to these problems, (i.e. better computational power, graphical information systems, etc.)

Well, just some thoughts; more of me putting "ink on paper" to form my thoughts.

More to come...

Technorati Tags: Epidemiology , GIS , Computational Biology

Researchers Find Mutation in Influenza A H5N1 Which May Increase Infectivity in Humans

I've discussed before a little about Influenza A H5N1 which is more commonly referred to in the media as "avian influenza". I may have to post some other time about what this nomenclature means, and how H5N1 comes about into a human population from the original avian source (or, even better, I might be able to directly link to a review article I recently wrote if it will be made available publicly by the publisher). So, I am sorry, but the following couple paragraphs will assume some background knowledge of influenza genetics and epidemiology:

However, suffice it to say that when H5N1 is referred to by the press as being seen in a human case; this is referring to a human case of disease where the human directly contracted this virus from an avian source. The classical case involves people who are in direct contact with poultry such as is seen in parts of Asia. This is important to point out because more evidence is pointing to this direct spread of H1N1 to humans prior to the 1918 pandemic.

This is surprising because until H5N1 became noticed in 2003 (and the more recent work on a recovered strain of 1918 H1N1); it had been conventional thinking that the appearance of entirely new strains of influenza occurred through other means. Specifically, this "Antigenic Shift" was thought to occur from the mixing of strains in animals such as pigs who are able to be infected by both avian and humanized strains of influenza A. This thinking mostly is based upon the study of the H2N2 and H3N2 pandemics which were less devastating than the H1N1 pandemic and are more likely to have been started by a partially humanized, yet new, virus. However, the implications of a fully avian pandemic are strikingly different.

One key point that is missing in the investigation of why and how do these avain viruses jump straight to humans lies in the specific interactions between the virus and it's host. This is fairly basic stuff; viruses are made up of DNA (or RNA in the case of influenza) just like humans are, and any changes to this genetic code has a direct impact on the structure and function of the virus. This is particularly explicit with viruses because they are such simple organisms that their genetic code is so much more analagous to a computer program. In a virus, a single base mutation can make the difference between infectivity in an entirely new host species or significant attenuation of virulence. In addition, because of the extremely rapid growth rates of many viruses, it is almost certain that in a population of virus, there exists (on average) at least one mutation per base position across the entire genome.

In the article cited below, the researchers discover a single change in amino acid to a Lysine at position 627 of the PB2 gene which confers a phenotype much more amenable to infection in a mouse model of human infection. Of course, the implications of this are significant in the study of how this gene, and it's related pathophysiology, is important to infection in humans. And, I have no doubt that there may exist other positions across the influenza genome which may also serve as 'switches' to turn on a more human-specific virus.

However, as I love to point out, this also serves as a target for genetic engineering of a pandemic strain of influenza which could be designed for dispersal into the human population. This reminds me of the report of mutations mousepox related to IL-2 regulation from Australia which profusely increased lethality (mousepox is related to smallpox, btw). So, thus is the consequencs of studying infectious agents for good cause, we need to be aware that this same information could be used for nefarious means.

• Hatta M, Hatta Y, Kim JH, Watanabe S, Shinya K, et al. Growth of H5N1 Influenza A Viruses in the Upper Respiratory Tracts of Mice. PLoS Pathogens Vol. 3, No. 10, e133 doi:10.1371/journal.ppat.0030133

2DTV - Saddam Hussain - Germ Warfare

That 'ol Saddam... he sure was a "pain in the neck"...

Sample Cover Letter for Journal Manuscript Resubmissions

This is a Sample Cover Letter for Journal Manuscript Resubmissions written by an obviously frustrated author. The author starts with "here is the re-re-re-revised revision of our paper" and proceeds to vent his frustrations of the peer reviewed process of journal writing. Very funny.

Extension of Smallpox Vaccine License - Not a Big Deal Unless we Refocus our Attention

Wyeth, the makers of the Dryvax® vaccine, received extension for their eight lots of remaining vaccine.

Dryvax® is a smallpox vaccine. For all practical purposes, if you are old enough to have a small scar on your upper arm (or maybe your parents did), then you've received this vaccine. Problem is that we stopped vaccinating people in the United States in 1972 and then the last known case of (natural) smallpox was in Africa in 1977.

Smallpox vaccination resumed, shortly after 9/11, due to fears of biological attacks and the necessity to vaccinate military personnel and first responders. This resumption of vaccination was marked by significant adverse events, which have previously been known from the old days of vaccination and there was a recent report of secondary inoculation of a close contact. Of note, a new complication has been recognized with cases of pericarditis/myocarditis (inflammation of the heart).

Considering the fact that there is absolutely no naturally occuring smallpox in the world and there has been no "known" (at least known to the public) dissemination of smallpox to terrorist groups or rogue nations; then it is reasonable to avoid mass vaccination at this point in time.

However, I believe that the threat is still present and we do need to maintain preparation. And, thankfully, there are options. I will simply refer to Imvamune®, which is produced by Bavarian Nordic. This is an attenuated vaccinia virus (due to multiple serial passages) which does not replicate in humans. There is sufficient research to suggest that this vaccine is nearly as effective as the older smallpox vaccines and has even been tested in immunosuppressed patients and in HIV.

So, we have eight lots of Dryvax? Big deal, in my opinion. I haven't looked up yet to see how many doses are in 8 lots (and keep in mind that we would be diluting this further to give 1/2 or 1/4 doses to make this supply stretch). But, even with ring vaccination techniques, how are we supposed to control a mass release of Smallpox with this small amount of vaccine?

I think that people invision one dude out in the New York City trying to release smallpox to unwary passers by. And, if that were the case, then the proposed plans to do contact tracting (and see my prior post to see how well we are at doing that currently) and then ring vaccinate in order to contain a biological event. But that's absolute nonsense!!!

Check out Dark Winter. And then some peer-reviewed responses to this exercise: CID Paper (freely available) and Med Anthro paper (not free). This table top exercise looked at mass dissemination at three separate shopping malls in Oklahoma City. But these guys STILL don't get it.

OK, I'm not a psycho, but I just like to think about this stuff... a lot... it's kinda weird. So, if I was some evil terrorist, How whould I do this? Very simple, you do what nature has already done in massive pandemics in the past: You don't do something stupid like infecting one single area, thus making it easy for your enemy to at least try to contain your attack; NO. You release this stuff in 16 cities, simultaneously. You probably say, "that's crazy". Well, duhhh... it is crazy, but let's look at from the perspective of a military strategist:

What resources does our enemy have (assuming, in my opinion, a worst case scenario that we are dealing with an Al-Quaida type of enemy)

• Psycho, religious fanatics who believe that they will go to heaven and have a harem of virgins if they die fighting the United States, Israel, or their allies, thus the suicide bomber or suicide terrorist phenomenon
• Plenty of these zealots
• Plenty of money
• Plenty of time
• Plenty of determination
• Not a whole lot else

So, with just this limited resource, what has Al-Quaida already done?

1. With just 16 people (15 actually), destroyed an entire section of New York City thus taking down two of the largest buildings in the world and killing a few thousand innocent civilians as well as destroying a good portion of the headquarters for the most sophisticated military in all of human history. Oh, and one plane (so a 75% success rate) didn't make it's target to the capital and all of this happened in a just a couple hours.
2. There are many other "coordinated attack" such as bombings in Africa.
3. Multiple, too many to stomach actually the number, of suicide bombings in Israel and Iraq.

So, then, how do you estimate how this enemy might use a new weapon (Smallpox)?

1. Convince your followers that your God will protect you from any attack that you unleash on your enemy - I'm pretty sure this has already been done.
2. You recruit as many zealots as you can who are willing to die for their cause, in essence suicide bombers - I gaurantee that this is already done.
3. Brain wash these guys in the usual way that they do this sort of thing - really, that's about all the training that you would have to do for this plan... and, again, this is already done.
4. Set up cells in 16 major cities. (in the U.S.) I'm going to estimate 3 or 4 of these zealots per cell. Has already been done in the past, and is theorized that there are active cells around the world already.
5. OK, now for the ONLY hard part: Get some smallpox virus, probably just costs some money to pay off an unemployed Russian scientist, or convince Kim Jong Ill to give you some from the Korean stockpile.
6. Now, where the plan gets specific: Get one person with some knowledge of microbiology who is able to smuggle samples of of this to the cells - all he has to do then is inoculate these people. Very easy to do. And smuggling this would be easy as well since a sample of smallpox virus looks like "nothing", just some powder and there is NO TEST to detect this at any security station anywhere in the world (i.e. airports).
7. Give your suicide terrorists a few hundred dollars - easy for Al Quaida to come across.
8. Now, just let your infected suicide terrorists travel around to as many public places as they possibly can; ride as many mass-transit systems as possible. Instruct them to avoid going to a hospital for as long as they can. Instruct them to cough in peoples faces, spit on tables, anything to disseminate the virus as soon as they start getting sick.
9. Now, my friends, nature takes it's course...
10. I'm not sophisticated enough to model out how this sort of scenario would take place, but it scares me to death. Imagine the number of contacts that could be made after good 10 to 14 days of people spreading this intentionally before getting too sick to go any further, and then some good semaritan might be likely to find this person sick and physically take them to a hospital where there is a chance that an admitting physician is not bright enough to recognize smallpox and then let this spread into a hospital of (relatively) immunocomprised patients.
11. How do we trace contact all people who could have been exposed? how well did we just do with ONE guy who chose to do some international travel?
12. There are varying estimates to the number of other people that one contact can infect, but the estimates are ALL greater than one, which means that this naturally spreads into a population.
13. Natural Smallpox kills 33% of all people infected.

Just my 2 cents.

CDC Webcast on XDR TB Case - Failure of our Current Beauracracy to Protect Against Biological Dissemination

A webcast was published today by the CDC on the XDR TB case which I reported yesterday. Thankfully the patient seems to be stable and it seems that he is still "smear negative". This is important becuase it is believed that the infectivity of a person with TB is directly related to the amount of bacteria found in the sputum; thus, it could be surmised that this person very well has not infected any of the other people who were on these trans-atlantic flights. However, I recall that there is evidence that transmission rates are still about 15% for culture positive/smear negative cases.

There does not seem to be much other information at this time as the investigation is still early. The CDC is proceeding with contact tracing, but it seems that there is quite a bit of beauracracy in this process. big surprise...

Here is the timeline as I understand it:

• January 2007??? incidental finding on a chest x-ray is consistent with right upper lobe tuberculosis - the date needs to be clarified per this webcast, furthermore, they did not have any other information on any other preceding events
• May 10th: Patient has laboratory documentation of MDR TB and he was instructed "Not to travel"
• May 12th: This patient chooses to go ahead and travel from the U.S. to Europe
• May 22nd: Laboratory results are made available which shows that this patient's isolate is an XDR strain
• May 23: The patient was finally contacted and was again instructed "Not to do any further travel"
• May 24th: This patient chooses to go ahead and fly back to the Canada and drives via automobile to the United States
• May 25th: The patient is finally quarantined by federal order and placed into inpatient treatment

I would submit that this ought to be taken into consideration in legislation and planning issues when it comes to other areas of "homeland security". I mean, seriously, how damn hard should it be for an Epidemic Intelligence Service Officer (EIS) to be able to gather simple information such as a flight manifest? I understand that there are issues of privacy, but what if this was something a bit more contagious and rapidly progressing? What if this was a case of a terrorist self-inoculated with a rogue sample of Smallpox and we were dealing with a situation where extra hours of delay in contact tracing literally translates into many new infections. And, FYI, the mortality of smallpox is approximately 30%, so let me ask you, would it be acceptable to not have a basic manifest of potential contacts nearly a week after a case was identified? In this regard, thank God this is just uber-resistant tuberculosis...

CDC Investigates U.S. Traveler with Extensively Drug Resistant Tuberculosis

The Centers for Disease Control posted a Health Advisory alerting to the investigation of a travelor found to have Extensively Drug Resistant Tuberculosis (XDR TB) who may have exposed other people while traveling on commercial aircraft:

"CDC learned that a patient with XDR TB traveled to Europe via commercial airline (Air France # 385) departing Atlanta on May 12 and arriving in Paris on May 13, 2007, and returned to the United States after taking a commercial flight on May 24 from Prague, Czech Republic to Montreal, Canada (Czech Air # 0104). The patient re-entered the U.S. on May 24 via automobile."

XDR TB refers to strains of Tuberculosis (caused by Mycobacterium tuberculosis) that are resistant to nearly all available antibiotics. In particular, this requires resistance to two first line agents (INH and Rifampin), to fluoroquinolones (i.e. Moxifloxacin which may soon become a preferred first line agent) as well as at least three of the second line agents. With this level of resitance, treatment usually hinges on a very lengthy duration of treatment and the use of potentially toxic drugs.

Recent reports from Africa have described a rapidly fatal course with a median survival of 16 days. This seems to be much more rapid than 'typical' tuberculosis and it is not clear at this point if there are associated changes in virulence in the XDR strains compared to routine or even MDR strains of TB. The reports from Africa were also associated with HIV infection; the current reports do not specify the HIV status of the contact from this CDC report.

According to the CDC report, there has never been an investigation of commercial airline associated contact with an XDR case. It is being recommended by the CDC to notify all persons who were associated with these flights in order to be evaluated.

Avian Influenza A/(H7N2) Reported in the United Kingdom

On 25 May 2007, the United Kingdom Health Protection Agency (HPA) announced that influenza A/H7N2 virus infection had been laboratory confirmed in four individuals exposed to infected poultry in Corwen Farm, Conwy, Wales.

The poultry outbreak in Wales started on a smallholding, Corwen Farm, Conwy, on 8 May 2007, was laboratory confirmed on 24 May 2007, and notified to the World Organisation for Animal Health on 25 May 2007.

It appears that all cases involved were characterized by mild flu-like symptoms with testing otherwise negative for other etiologies. Of note, H7 strains (such as this H7N2 strain) are generally a milder strain of influenza in humans compared to the H5N1 strains of influenza that are predominantly circulating in Southeast Asia and typified by highly lethal "avian influenza".

Changes to US Immigration Policy Regarding HIV Status?

The Center for Strategic & International Studies (CSIS), which is a American "think tank", issued a report advising to relax restrictions on immigration policies for people infected with the Human Immunodeficiency Virus (HIV) which causes Acquired Immunodeficiency Syndrome (AIDS).

According to this 16 page report, the current US policy (and law) is to deny the admission of immigrants and non-immigrants with HIV infection. This policy is based on early 1980's fears and lack of understanding of the emerging HIV epidemic. Furthermore, this report concludes that current US policy represents one of the most stringent migrant pre-departure screening approaches in the world and suggests that this approach is obsolete and discriminatory.

One basis for these conclusions does ring of truth in that this is inconsistent with the United State's stated goals of fighting the HIV epidemic.

However, I do not see this report addressing the additional cost to American taxpayers for picking up the burden of treating non-citizens with HIV in an environment where US HIV providers already find it difficult to receive adequate compensation for taking care of HIV patients. Many of the patients in the United States, who have HIV, are indigent thus relying entirely on programs such as Medicare, Medicaid, or specialized funding such as the Ryan White act. New immigrants, with HIV, will likely tax these systems.

I would agree that changes to our current policy are in order, as this is fair. However, I hope that the policy makers keep in mind that this can cause additional stress to the care of HIV patients who are already here.

Fashionable Pandemic Personal Protective Equipment (PPE)?

I notice one of the Google AdSense links on my blog directs you to a site called Fashion Flu Masks. I was quite interested in seeing what this was all about and so I gave it a click:

The sites states that "We do not guarantee a Fashion Flu Mask alone will protect you from a pandemic flu, but it is a vital part of a flu defense package".

There is formal guidance regarding the use of masks in times of epidemic influenza as outlined by the CDC.

I requested specifications on the masks used. According to the maker, these are originally 3M N95 masks (which is a fairly standard multi-purpose mask for respiratory isolation). However, the designer specifically does not list this specification due to concerns regarding the ability to continue filtering at an N95 level after the decorating process.

These do look quite fun and I am struck by the creativity, although I would not personally recommend (over an original N95 mask) for any real outbreak of a respiratory infection.

I personally like the pink mask with the image of a bird shown to the right. Another quote from the site, "it's best to be prepared, but why not look stylish too?"

West Nile Virus: First Cases Being Reported to Arbonet

It is nearly June and the first cases of West Nile Virus (WNV) are being reported by the Centers for Disease Control (CDC).

As of yesterday (May 22nd), there are 3 cases reported, according to the Arbonet site. All three of these cases so far are being reported from Mississippi. However, reporting is likely to be under-reported at this point since there is usually a delay between the initial reports from local health departments until this information is updated to the Arbonet website.

The typical WNV season in the United States starts slowly in May with significant reporting that begins in June. Reporting typically dies down by October in most states. Last year there were a total of 4268 cases reported by the CDC.

First Post: Pandemic Influenza

Hmmm... what should I say?

Nothing profound today, but I am composing an article on pandemic influenza for a state medical society journal. This is quite interesting; although I have started out with some knowledge of influenza from prior study, the history of such things as the great 1918 pandemic are not readily known.

I suggest reading the book "The Great Influenza" by John M Barry. This book does an excellent job of setting the 1918 pandemic (AKA "Spanish Influenza") into an historical context.

For start, it becomes readily apparent that American medicine, at the turn of the century, was mostly non-scientific and primarily still relied on all the false presumptions left over from Hippocratic medicine. So, in this, the great pandemic of 1918 is most interesting because this occurred at a time when the medical community was just beginning to come to grips with crucial concepts such as Germ Theory, yet the essential details of molecular genetics or even the recognition of viruses was still lacking.

The historical context is further interesting given the relationship to World War I. This is even more relative when considering the current political climate. I don't think many people realize that during World War I (a war which Woodrow Wilson tried very hard to avoid entering) the United States was entirely controlled by the government. In the spirit of "total war", the government did all but outright censor the press and public. It becomes apparent that this was detrimental in the event of pandemic influenza, but overall it would seem that history as tolerated and accepted that course in our history. It is somewhat interesting to make comparisons to the current day popular mentality and governmental climate.

Nevertheless, imagine what it would be like to have approximately 50% of the entire US population sick with a disease where most of those are too ill to work and then about 10% die... and to have all of that occur over the period of just a few months. And it would appear that the mortality from "bird flu" (mostly attributed to H5N1 virus) is even more virulent than this.

Well, more to say about this some other time.