Regarding the National Academies of Science The Domestic and International Impacts of the 2009-H1N1 Influenza A Pandemic: Global Challenges, Global Solutions: Workshop Summary, there were missed opportunities to take advantage of a deep exploration of lessons learned in the warning challenges of the pandemic.
Before proceeding, it is important to note the definition of warning (from Merriam Webster) as "something that warns or serves to warn", where warn is defined as:
- to give notice to beforehand especially of danger or evil; or
- to call to one's attention
This implies an intelligent process that involves humans facilitated by automated systems to sensitize other humans to orient themselves to an event.
The NAS report contained an appendix, RUMORS OF PANDEMIC: MONITORING EMERGING DISEASE OUTBREAKS ON THE INTERNET, penned by Madoff and Brownstein from their perspectives as the leads for ProMED and the Google.org-funded HealthMap, respectively. The timeline of available information published for the pandemic was offered as follows:
BOX A9-2
Swine Flu Day by Day
11 March: First documented symptoms (as of 5 May) in a Mexico City resident who later would be found to have confirmed infection with A(H1N1) swine flu.
30 March: A 10-year-old boy with fever, cold, and vomiting goes to the Naval Medical Center San Diego in California. As part of a clinical study, a nasopharyngeal swab is sent across town to the Naval Health Research Center (NHRC).
1 April: NHRC researchers determine that the boy is likely infected with influenza A, but they cannot subtype the strain. As per protocol, the sample is sent to Marshfield Labs in Wisconsin. HealthMap, a global disease alert system run by academics, flags a news story from Mexico about a strange respiratory outbreak in the state of Veracruz that has claimed two lives.
11 April: As per the International Health Regulations (IHR), the World Health Organization (WHO) has a pandemic alert and response network, which relies on designated people or institutions in each member country to report unusual disease patterns. PAHO, a regional office of WHO, asks the Mexican IHR “focal point” to verify the outbreak reported in the news.
12 April: Mexico’s director general of epidemiology confirms to PAHO the existence of acute respiratory infections Studies continue. Mexico’s focal point considers outbreak to be a “potential public health event of international importance” because it meets IHR criteria: severe public health impact and an unusual event.
…
21 April: Samples from Mexico arrive at PHAC.
22 April: CDC publishes first dispatch in the Morbidity and Mortality Weekly Report (MMWR) about two cases in California. Mexico reports atypical influenza behavior associated with severe pneumonia in various cities. InDRE ships samples to PHAC’s National Microbiology Laboratory in Winnipeg and CDC.
ProMED’s first report on human cases citing CDC report.
23 April: Samples from Mexico arrive at CDC. PHAC and CDC confirm Mexico cases are the
same A(H1N1) of swine origin.
6 April: Veratect, a Kirkland, Washington-based company that scours news reports for emerging threats, reports in its subscription-only database that local Mexican health officials have declared an alert because of respiratory disease outbreak in La Gloria, Veracruz state, Mexico.
SOURCE: Adapted with permission from Cohen (2009).
Note this time line was reconstructed in hindsight. There is an important difference between historical reconstruction of an epidemic curve versus the time line of social awareness of a biological threat. The key difference is an orienting signal sent to the international public health community that triggers a swarm effect: a sudden and pervasive demand for information as observed during a crisis.
Several clarifications to Madoff and Brownstein's appendix:
Our approach was not an automated online media harvesting approach. A team of 30 analysts served as a human interface with multi-source information, working behind an elaborate set of reporting requirements modeled after a military watch cell. Entry of a report into our system, with subsequent distribution to clients, was a human moderated process by individuals trained in the recognition of indicators of an unusual disease phenomenon or sociological crisis. Further, our information was made freely available to WHO Headquarters Geneva (the GAR), PAHO, and CDC (both the GDDOC and emergency communications groups, to include those managing Epi-X).
Information we reported about the pandemic emergence in Mexico was sent in a human-driven, deliberate escalation of notification that began with a highlighted point on a map / web portal interface, to two emails sent to CDC, to a phone call to CDC's Emergency Operations Center (with a subsequent hour long conversation). It was the first time in my 12 year career doing this work I had the DEOC page the senior health intelligence officer after hours with an insistent message the individual call me back. During the conversation, concern about "atypical pneumonia" with attempted SARS rule out was discussed as well as the string of other "unusual / atypical" respiratory disease reported in various locations of Mexico. Reported involvement of the Canadians for diagnostic support was emphasized. My point here is to highlight the component of human recognition.
Our reporting of the La Gloria situation on April 6th reflected the use of multiple sources of information, not a single media report in a routine procedure by the analyst to clarify information claiming something unusual was going on in that community with pig farm involvement. Of course in hindsight an email should have been triggered at that point. But that is hindsight. Operationally, we see claims of "unusual" or "mystery" disease worldwide as routine- including Mexico. Thus, it was a judgment call on the analyst's part. What was not routine was use of the words "atypical pneumonia" in the Oaxaca reports on 16 April, which was a well-known catch phrase for SARS in 2002 and 2003.
We (and I personally) have always struggled with the notion of reporting credit. To be frank, the first person to report the situation in La Gloria was not HealthMap. It wasn't Veratect. It was the diligent Mexican journalist who took the time to investigate the situation. To-date, I have yet to see anyone identify this individual and recognize the tremendous contribution he made to the warning process. Most certainly the same sentiment goes to the Mexican journalists in Oaxaca State reporting on the situation in Reforma. Lastly, there are bloggers worldwide who have not been credited for helping to "amp the signal" via Twitter and other social media outlets to push orientation of international public health organizations to the crisis. Reporting credit is a necessary concern for organizations continually seeking funding, but a full chain of reporting evidence should be maintained to acknowledge those on the front line.
Here is the NAS timeline, now with our operational perspective added, highlighted in red. Again, the emphasis here is social recognition as events unfolded:
BOX A9-2
Swine Flu Day by Day
11 March: First documented symptoms (as of 5 May) in a Mexico City resident who later would be found to have confirmed infection with A(H1N1) swine flu. (Hindsight- we believe there was evidence in Baja California dating back to Feb. We reported 14 deaths due to unusual / undiagnosed disease with associated social unrest indicators in Mexicali on Feb 23rd that was flagged maximum alert and emailed to CDC- we did not receive word on the test results. Unknown if this was pH1N1. Again, hindsight.)
30 March: A 10-year-old boy with fever, cold, and vomiting goes to the Naval Medical Center San Diego in California. As part of a clinical study, a nasopharyngeal swab is sent across town to the Naval Health Research Center (NHRC). (No alert raised by CDC or the state of California. At this time we reported on a returning Canadian traveler- an attorney- who required ventilatory care following return from Mexico- still unclear if this individual had pH1N1 infection. We did not move further with the notification process beyond posting in our web portal. Again, this kind of thing happens from time to time.)
1 April: NHRC researchers determine that the boy is likely infected with influenza A, but they cannot subtype the strain. As per (routine) protocol, the sample is sent to Marshfield Labs in Wisconsin. (No alert raised by CDC or the state of California.)
HealthMap, a global disease alert system run by academics, flags a news story from Mexico about a strange respiratory outbreak in the state of Veracruz that has claimed two lives. (Of note, this was a point on a map among hundreds of points of a variety of flag colors.)
6 April: We report unusual respiratory disease in La Gloria with claims of pediatric fatalities and an association with a pig farm. Information posted on our web portal.
10 April: A PAHO health intelligence officer accesses the La Gloria, Mexico report in our system.
11 April: As per the International Health Regulations (IHR), the World Health Organization (WHO) has a pandemic alert and response network, which relies on designated people or institutions in each member country to report unusual disease patterns. PAHO, a regional office of WHO, asks the Mexican IHR “focal point” to verify the outbreak reported in the news. (PAHO's health intelligence officer accessed the La Gloria, Mexico report in our system again on April 11th.)
12 April: Mexico’s director general of epidemiology confirms to PAHO the existence of acute respiratory infections Studies continue. Mexico’s focal point considers outbreak to be a “potential public health event of international importance” because it meets IHR criteria: severe public health impact and an unusual event. (No alert raised by WHO, PAHO, or CDC)
17 April: Repeat email of the 16 April report sent to CDC.
20 April: Anonymous source reveals request by Mexican officials for diagnostic assistance from the Canadians and possible deployment of Canadian personnel to Mexico. Veratect immediately phones the CDC Director's Emergency Operations Center (DEOC). CDC indicates they have been focused on the situation in California and Texas. More reports of unusual respiratory disease in other areas of Mexico.
…
21 April: Samples from Mexico arrive at PHAC (in Winnipeg Canada). We
report deaths due to "atypical pneumonia" in Oaxaca with quarantine of
the involved hospital facility tests for SARS and A/H5N1 negative.
Veratect sensitizes the International Federation of Red Cross who in turn requested broader access be provided to the Pan-American Disaster Response Unit (PADRU). Veratect moved to notify several US state and local public health authorities, providing the caveat the situation in Mexico remained unclear due to pending laboratory results. Veratect reached out to World Health Organization (WHO) operations, informing them the Veratect team was on full alert posture and available for situational awareness support. They indicated they and their subordinate, the Pan American Health Organization (PAHO) were now aware of the situation but had no further information. Veratect also extended contact to the British Columbia Center for Disease Control and offered assistance in tracking the events in Mexico. All contacts indicated laboratory results were pending.
22 April: CDC publishes first dispatch in the Morbidity and Mortality Weekly Report (MMWR) about two cases in California. Mexico reports atypical influenza behavior associated with severe pneumonia in various cities. (No international alert for the Mexico cases raised by CDC or mentioned in the MMWR) InDRE ships samples to PHAC’s National Microbiology Laboratory in Winnipeg and CDC.
ProMED’s first report on human cases (in the United States) citing CDC report.
We report Canada announced a national alert for travelers returning
from Mexico with respiratory disease, beginning a campaign of public media
announcements. Potentially ill contacts were identified returning from Mexico
and isolated in Canada. Internet
blogs begin to spin up. CDC
indicates concern about the events unfolding in Mexico. Veratect sensitizes the US community
physician social network managed by Ozmosis.
23 April: Samples from Mexico arrive at CDC. PHAC and CDC confirm Mexico cases are the same A(H1N1) of swine origin.
Veratect issues notification
to additional public health authorities in two states. Veratect reaches
out to the Pan American Health Organization emergency operations team but is
unable to establish contact. Veratect notes no publicly available English
language reporting from ProMED, HealthMap, FluNET, CDC, ECDC, or WHO about the
unfolding events in Mexico. Many of Veratect’s clients, including
Canadian, ask why an alert has not been issued by the US to sensitize their
healthcare community.
24 April: Veratect notifies the United Kingdom National Health Service NaTHNaC (National Travel Health Network and Centre). The NaTHNaC website as of April 24th contained no public announcement of the situation in Mexico. Response from their senior analyst did not indicate awareness. Veratect notifies the private US clinical laboratory community and activates a Twitter feed (twitter.com/veratect) to enable more rapid updating of information until WHO and CDC were able to come up to date with their public reporting.
At no time during this process did we observe that Google's Flu Trend provide early warning information for the California, Texas, or Mexico events.
As with HealthMap andProMED, we consider Flu Trends to be a helpful source of adjunct information that requires corroboration by other sources. As rightly pointed out by Madoff and Brownstein, redundancy and variety of information should be encouraged.
The purpose of highlighting the above is not to criticize individuals or organizations. Detection and warning of emerging influenza pandemics is arguably one of the most difficult challenges in biosurveillance operations, and the international community has struggled with the problem of time-sensitive warning processes since the 1957 pandemic, the first influenza pandemic facilitated by the global air traffic grid. But this represents a critical requirement as part of an effective global biosecurity strategy: effective warning processes.
Here the NAS should consider an opportunity to analyze the challenges in warning during the 2009 H1N1 pandemic and address the emergence of an analytic discipline that underpins an effective process for time-sensitive warning of rapidly evolving biological threats.
Recent Comments