Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Abstracts / International Journal of Infectious Diseases 53S (2016) 4–163 31
16.004
Ecology and environmental drivers of
antimicrobial resistance
U. Theuretzbacher
Center for Anti-Infective Agents, Vienna/AT
Given the multifaceted nature of the resistance problem, the
focus of attention has expanded from human and animal antibiotic
use to the human influence on resistance in the environment. The
link between the animal and human sector are well studied and
led to policy changes in some parts of the world. Such regulatory
initiatives are still missing in the environmental field which is usu-
ally not included in the One Health approach to tackle the global
resistance problem.
The direct release of multidrug resistant bacteria from health-
care settings and animal farms into the environment as well as the
pollution of the environment with high concentrations of antibi-
otics create a dangerous resistance reservoir. Recent metagenomics
studies highlighted the role of mobile genetic elements (mobilome)
as environmental pollutants and their role in co-assembling of
resistance determinants and horizontal transfer from environ-
mental bacteria to pathogens and vice-versa. Phages are widely
distributed in nature and may act as vehicles for such co-localized
resistance cluster genes resistance genes with significant implica-
tions for the horizontal spread of antibiotic resistance. They are
enriched in microbial genomes or independent of their bacte-
rial host in hospital wastewater systems, animal husbandry and
its wastes, aquaculture, and in wastewater treatment plants. This
resistance gene pool (environmental resistome) has been described
in natural waters, sediments near effluent pipe openings, and in
soil especially agricultural farm land due to irrigation with con-
taminated water. High levels of antibiotic residues in wastewater
plants, natural waters but also reclaimed water supply systems
due to unregulated pollution from antibiotic manufacturing plants
exert unprecedented selection pressure in nature. Addressing this
problem requires concerted policy actions and needs to be included
in the One Health approach of current global initiatives.
http://dx.doi.org/10.1016/j.ijid.2016.11.084
17.001
Estimating FluNearYou correlation to CDC’s
ILINet
R. Arafata,∗, E. Bakotab, E. Santosb
aHouston Health Department, Office of Surveillance
and Public Health Preparedness, Houston, TEXAS/US
bHHD, HHD, Houston/US
Purpose: To provide evidence for the data quality of Flu Near
You (FNY) by evaluating the national and Houston datasets against
CDC influenza-like illness (ILI) data.
Methods & Materials: Eachweek, FNY users submitsurveys that
describe the symptoms experienced for the previous week. The sur-
vey tracks if and when a user has received a flu shot and experienced
ILI. This study used those survey responses. The data were deiden-
tified and provided by the Skoll Global Threats Fund to the Houston
Health Department (HHD). The FNY data were compared to ILINet’s
national summary of ILI and influenza positive tests by estimating
the correlation coefficient for the 2014-2015 influenza season. FNY
total ILI counts were correlated to total positive influenza tests, and
FNY percent ILI was compared to ILINet’s unweighted percent ILI.
Mean correlation coefficients for 1,000 bootstraps were estimated
for a sequence of weekly user counts of 10 to 10,435 in increments
of 10. Bootstrapped samples were stratified by ZIP code to account
for fluctuations in weekly participation for both FNY and ILINet, as
both datasets see an increase in user participation during influenza
season. R version 3.2 was used for all analyses; HHD received the
line-list dataset from FNY that contained nearly 400,000 entries.
Each entry corresponds to a single person.
Results: •Correlation of the full FNY dataset against ILI & flu
tests are very high (r2= .94 and .92 respectively).
•Weekly reports from < 200 weekly users have high variance
in their correlation to ILINet and a moderate correlation coefficient
(r2between 0.3 and 0.7).
•At low participation counts, (< 400 per week) FNY correlates
better with positive influenza tests than percentage with ILI.
•Overall, FNY data correlates well with national ILINet data,
even at limited participation levels.
Conclusion: Approximately two-thirds of the counties within
the United States have a population of < 50,000. As such, FNY
provides a simple, low-cost opportunity for public health officials
within those jurisdictions to obtain data that reasonably mirrors
ILINet. For larger jurisdictions, FNY is another tool available to track
and identify seasonal influenza and engage the public on preven-
tion.
http://dx.doi.org/10.1016/j.ijid.2016.11.085
17.002
The first phase of PREDICT: Surveillance for
emerging infectious zoonotic diseases of
wildlife origin (2009-2014)
D. Jolya,∗, C. Kreuder Johnsonb, T. Goldsteinc, S.J.
Anthonyd, W. Kareshe, P. Daszake, N. Wolfef,S.
Murrayg, J. Mazeth
aMetabiota, Nanaimo, BRITISH COLUMBIA/CA
bUniversity of California - Davis, Wildlife Health
Center, Davis, CA/US
cUniversity of California - Davis, Davis, CA/US
dColumbia University, Center for Infection and
Immunity, MSPH, New York, NY/US
eEcoHealth Alliance, New York, NY/US
fGlobal Viral Forecasting Initiative, San Francisco,
CA/US
gSmithsonian, Washington, DC/US
hUC Davis, Davis, CA/US
Purpose: Based on the premise that the majority of emerg-
ing infectious zoonotic diseases originate in wildlife species, the
United States Agency for International Development created the
Emerging Pandemic Threats Program to increase capacity in the
developing world to detect and respond to emerging threats. A
coalition of organizations led by the University of California at
Davis, and including EcoHealth Alliance, Wildlife Conservation
Society, Metabiota Inc., and the Smithsonian Institution, imple-
mented the first phase of PREDICT (2009-2014), the component of
the program tasked with developing the capacity for early detection
of these emerging threats.
Methods & Materials: Based on an iterative process of field and
digital data collection and statistical computer modeling, PREDICT
identified geographic, taxonomic, and behavioural interfaces likely
to lead to disease emergence.
Results: Between 2009 and 2014, over 250,000 samples from
over 56,000 animals were collected from wildlife in close proxim-
ity to humans (46.6%), free-ranging wildlife and hunted wildlife
(31.8%), traded and market wildlife (14.6%), and other sampling
sources (7%). Family-level viral screening was conducted using
32 Abstracts / International Journal of Infectious Diseases 53S (2016) 4–163
consensus PCR, in 32 laboratories in 20 developing countries
around the world. Over 800 hundred novel viruses were found,
based on molecular characterisation and on the percentage
sequence identity between established species, in addition to over
a hundred known viruses.
Conclusion: Implementation of the PREDICT surveillance strat-
egy and prioritization process has improved the capacity in hotspot
countries to detect and respond to emerging disease threats.
http://dx.doi.org/10.1016/j.ijid.2016.11.086
17.003
Emerging and re-emerging infectious diseases
in displaced populations 1998 to 2016: An
analysis of ProMED-mail reports
J.W. Ramatowskia,∗, L. Madoffa, B. Lassmanna,N.
Maranob
aInternational Society for Infectious Diseases,
Brookline, MA/US
bCDC, Atlanta, GA/US
Purpose: Understanding the occurrence of emerging and re-
emerging infectious disease outbreaks in displaced populations is
important to ensure adequate control measures.
Methods & Materials: The 1994–2015 ProMED-mail record
database was queried for records containing the term “refugee,”
“asylum seeker,” and “displaced.” For the purpose of this analysis,
together these groups are termed displaced populations (DPs). Of
the 52,247 records, 600 were returned. Records containing one of
the listed terms were then assessed for the following information:
reported disease outbreak location, reported disease, origin of DPs,
and number of people affected by the outbreak. Unique outbreak
events were then identified. One outbreak event possibly contained
multiple records. Rates of outbreak events, per total number of
ProMED-mail reports each year, were calculated to ensure that
any changes, over time, were not simply secondary to changes
in the total number of reports posted on ProMED and were com-
pared using a two-sided t-test; P <0.05 was considered statistically
significant.
Results: Of 600 records, 118 disease outbreaks spanning years
1998–2015 were identified for use by this review. The mean inci-
dence of reported outbreak events increased across three, 5-year
interval periods (Figure 1). Kenya, Uganda, and Sudan had seven or
more outbreak events between years 1998-2015. The number of
outbreak events in DPs per total ProMED-mail posts between the
first and third 5-year interval increased by 277% (P <0.01). In total,
>559,000 cases of emerging and re-emerging infectious diseases
were reported from the 118 events. Of these, >520,000 cases were
related to the cholera outbreak in internally displaced people after
the 2010 Haitian earthquake. Additionally, >14,000 vaccine pre-
ventable disease cases (measles, chickenpox, polio, and tetanus)
and >10,000 Hepatitis E cases were reported. Less common out-
breaks included malaria, dengue, hemorrhagic fevers, meningitis,
leishmaniasis, louse-borne relapsing fever, anthrax and typhoid
fever.
Time Period Sum of Reports Average Reports-Per-Year
2010-2015 65 11
2004-2009 33 6
1998-2003 17 3
Conclusion: As the number of displaced people grows, there
has been an associated rise in reports related to emerging and re-
emerging diseases in DPs. The results of this analysis underscore
the importance of adequate infrastructure, human resources, clean
water access, and ongoing support needed to prevent, diagnose,
and treat infectious diseases in DPs, with particular emphasis in
under-resourced countries.
http://dx.doi.org/10.1016/j.ijid.2016.11.087
17.004
Digital functions in a participatory One Health
surveillance initiative aiming for pandemic
averting
P. Susampaoa,∗, K. Chanachaib, P. Petraa,T.
Yanoc, S. Pattamakaewd, E. Laiyad,L.
Srikitjakarne, A. Crawleyf, J. Olsenf,M.
Smolinskig
aOpendream Co., Ltd., Bangkok/TH
bDepartment Of Livestock Development, Bangkok/TH
cFaculty of Veterinary Medicine, Chiang Mai
University, Department of Food Animal Clinics,
Chiang Mai/TH
dFaculty of Veterinary Medicine, Chiang Mai
University, Chiang Mai/TH
eFaculty of Veterinary Medicine, Chiang Mai
University, Veterinary Biosciences and Veterinary
Public Health, Chiang Mai/TH
fSkoll Global Threats Fund, San Francisco/US
gSkoll Global Threats Fund, San Francisco, CA/US
Purpose: A community-based Participatory One Health Dis-
ease Detection system (PODD) using smart phone technology was
piloted in Chiang Mai, Thailand. Volunteers from 300 villages and
74 community governmental agencies were selected purposively
to submit daily surveillance reports of poultry health and disease
in their communities The primary objective of PODD in pilot phase
was to detect abnormal deaths in backyard animal in order to
elicit rapid investigation and response. Abnormal numbers or types
of death can be a signal of zoonotic diseases which transmits to
human and causes pandemic as a subsequence, such as abnormal
death in poultry could be an early clinical sign of highly pathogenic
avian influenza (HPAI), a potential precursor of an AI pandemic in
humans. Use of smart phones and digital technology is one of the
key factors making the PODD system workable.
Methods & Materials: The daily reports of poultry health and
abnormal poultry death are automatically captured, filtered with
predefined case and outbreak definitions, and projected onto a GIS
mapping system. The real time analysis of incoming reports allows
rapid detection of outbreaks and the generation of automatic SMS
warning messages to activate community contingency plans. A dis-
ease investigation team is dispatched to confirm the outbreak by
clinical examination and, as necessary, laboratory confirmation.
The system follows up automatically until 3 weeks after the last
report of sick animals or death in the affected area. All stakeholders
are notified after complete recovery to normal.
Results: During the first 16 months of PODD system piloting 25
abnormal death outbreaks were detected. Eight of the outbreaks
were laboratory confirmed with devastating epizootic pathogens,
while 17 of the outbreaks unable to confirmed the causes. Within
eight laboratory confirmed outbreaks, two of which resulted in
almost all chickens in the villages dying. The other six outbreaks
could be timely and effectively controlled by the communities.
Conclusion: Those early outbreak detection and rapid response
demonstrated the potential to integrate this PODD surveillance sys-
tem under their one health operation centres to prevent pandemic
in their community.
http://dx.doi.org/10.1016/j.ijid.2016.11.088
