CDC's recent claims to the international media hailing
the success of the Haiti IDP Syndromic Surveillance system are
inaccurate. From an operational perspective, the IDPSS has not been
successful in providing the following timely warning of infectious
disease events such as respiratory disease, hepatitis A, dengue,
diphtheria (including a fatal case in a child), increased malaria
transmission activity, and diarrheal disease. The system has, since the
earthquake, been compromised due to time-delayed reporting by
participants, changing participants and "reporter burn-out", lack of
appropriate laboratory and epidemiological investigation capacity to
investigate reports of disease outbreaks, and lack of epidemiological
baselines with which to compare the data. Further, CDC and the MSPP has
largely failed to promote an atmosphere of transparency in reporting
what is and is not known. We have documented several instances of
interference by both CDC and MSPP in our operations to share information
about disease events shared among non-governmental organizations. This is an unacceptable practice in a post-disaster setting of the magnitude seen in Haiti. Indeed, it may be considered unethical.
The definition of an "outbreak" or
"epidemic" cannot be expressed in Haiti due to lack of data. We
strongly challenge any assertion made by any organization claiming "no
outbreaks or epidemics" without scientifically credible data proving
this statistically. Without a denominator, to calculate against, one
cannot assert statistically-significant deviation from baseline. There
is little question that infectious disease is in Haiti, produces serious
morbidity, and has killed people since the earthquake- what is not
known is to what degree this differs from baseline / pre-earthquake
conditions. Since the IDPSS was created after the earthquake, it is not
possible to assert there has been any change from baseline, either increased or decreased for multiple diseases due to lack of proper investigative capacity.
One point we may all agree on is the fact that no infectious disease disaster has been observed in Haiti, as indicated, for example, by abrupt "panic" evacuation of IDP camps or mass fatalities. We may consider the efforts of organizations such as Oxfam and Catholic Relief Services, among many, for the provision of chlorinated drinking water as an example of preventive measures that likely inhibited the spread of disease. Another factor likely at play is herd immunity of the indigenous population, which is largely unquantified both prior and following the quake.
Syndromic surveillance has long held sway
over public health, much to the exclusion of other processes that have
proven operational validity in sensitizing human networks in a tactical
setting to a 'problem' such as social networking. This persistent bias
prevents effective warning from being issued in a truly timely manner,
as opposed to weeks after the pattern of an event is recognized to be an
epidemic curve. We have observed active suppression of social
networking outside of public health channels, which is reflective of a
sociological process that also impaired the vision of the DHS National
Biosurveillance Integration Center from being realized for the United
States and its allies. It is also the reason why BioSense has failed to
meet expectations. While syndromic surveillance has a place in the
orchestra of biosurveillance instruments, however it is but one
instrument with definite weaknesses. Without the integrative discussion
of experienced and astute clinicians and analysts in an operational
setting, such systems have limited value.
The CDC's implementation of the
IDPSS in Haiti is an example of failed integrative biosurveillance, a
microcosm of the same sociological issues that prevent our country from
realizing the full potential of operationally-relevant biosurveillance.
It is our hope this exercise prompts thoughtful and unbiased discussion
about more appropriate strategies for operationally-appropriate
infectious disease crisis warning systems.
Comments