Too Many Systems Collecting Data on Disease Outbreaks?

[CC-Conrad Clyburn-MedForeSight]

Colleagues,

As reported in Federal Computer Week ... About 300 systems at federal, state and local agencies monitor disease outbreaks and chemical exposure. Some critics say that multiplicity is a problem.

Serena Vinter, a senior research associate at the Trust for America’s Health, a nonprofit public health organization, said ... “There are too many disease surveillance systems, and they do not necessarily communicate with each other,” Vinter said. “It is hard to get a good picture.”

Others say the organizations in charge of alerting the public are often slow to act. Veratect, a company that tracks disease outbreaks, said it had sounded the alarm about the swine flu outbreak in Mexico March 30, but several weeks passed before the World Health Organization and Centers for Disease Control and Prevention began widely publicizing the illness...


Read on at: http://fcw.com/articles/2009/10/26/cdc-side-murky-crystal-ball.aspx...

CC

[Carl Taylor]

Conrad-This is a more complex issue than just system clash. First under the federalist approach to disaster response each state has played a role in buying or designing some biosurveillence/situational awareness/surge capacity management tool, generally focused on hospital ED reporting. These systems exist alongside a series of federal initiatives which I will not comment on. Regardless, any system (and by way of a disclaimer we designed the one in my state) that is mostly hospital centric will not provide an adequate picture of H1N1's impact since much of the front lines of care for this disease lives at the primary care physician level, off the radar for most systems. This is why in August Steve Prior and I were suggesting the first wave of H1 was over 1.5 million cases, when the public numbers where somewhere around 100,000. Ok Steve and I were wrong, it looks like more recent data suggests a number closer to 5 million, but we were the highest without going over so we must win something. In any event the first question then is how do we improve data collection on the ground and then make it transparent to all parties who need to know from the local responders up to the Federal and WHO level. The second is how do we aggregate this data, when needed to regional or Federal resources.There are at least some efforts HavABed comes to mind as one. The third issue however is the most troubling and is the one perhaps raised inferentially by Veratect. Lets assume we can dramatically improve the data collection (and I believe we can by grabbing data at the claims level which can be done relatively non invasively to a doctors practice and is inexpensive), who cares and what are the public health systems going to do about it? At this point many jurisdictions on a global level have quit collecting numbers. H1 has simply gotten too widespread. I would respectfully suggest this is a mistake. By not pushing biosurv. more aggressively we run the risk of not knowing. By not knowing we will lose direct patient intervention opportunities which will leave our EDs overwhelmed when there are far better surge management strategies readily available, we will also lose lessons learned such as comorbid impact on disease severity, follow on care requirements for some patients such as babies born to mothers who had H1 (1918 produce issues in this space) and challenge our ability to accurately assess future disease waves. I am reasonably convinced we can do better in grabbing and aggregating the data to monitor the outbreaks but concomitant with improving our data collection and aggregation abilities will be the need to rethink how we should intervene to use the data so what we collect isnt simply more statistical noise.

[CC-Conrad Clyburn-MedForeSight]

Carl,

Thank you for the cogent and important analysis. If I recall, the 1918 flu epidemic came in at least two waves, with the second being the most serious. You raise critical points. I will share them with the community for their consideration.

CC