Review of the ISDS DiSTRIBuTE Project

1
Review of the ISDS DiSTRIBuTE Project 2007/2008
Influenza Season Proof-of-Concept Phase Don
Olson, Research Director International Society
for Disease Surveillance ISDS Annual Conference,
Raleigh NC 4 December 2008 drolson_at_syndromic.org
2
DiSTRIBuTE

• Distributed Surveillance Taskforce
• for Real-time Influenza Burden
• Tracking Evaluation

3

• DiSTRIBuTE an ISDS Project
• Began 2006 (funded through NACCHO/CDC)
• Initiated through network of Local Public Health
  expert practitioners of syndromic surveillance.
• Developed based on local departments using local
  best-practices, maintaining local control.
• Using a minimum data set of aggregated fever,
  respiratory and influenza-like syndrome by broad
  age and zip-3, simply based on the epidemiology
  of the question Does it provide a sufficient
  surrogate measure for reliable population health
  monitoring of influenza related morbidity?
• .

4

• DiSTRIBuTE Recent History
• Developed through 2007 (NACCHO/CDC)
• By growing the network
• Building the foundation
• Proof-of-Concept Phase 2007/2008
• (NACCHO/CDC and Markle Foundation)
• Continue growing it
• Making it work better
• Showing that its working
• Better, Faster Cheaper

5
Federal NeedPandemic and All-Hazards
Preparedness Act
the Secretary, in collaboration with State,
local, and tribal public health officials, shall
establish a near real-time electronic nationwide
public health situational awareness capability
through an interoperable network of systems Such
network shall be built on existing State
situational awareness systems or enhanced systems
that enable such connectivity.
109th Congress of the United States, Amendment to
the Public Service Act. Pandemic and All-Hazards
Preparedness Act. (2006). Pub L No. 109-417, 101
et seq.
6
Federal NeedPresidential Directive (HSPD-21)
The Secretary of Health and Human Services
shall establish an operational national
epidemiologic surveillance system for human
health, with international connectivity where
appropriate, that is predicated on State,
regional, and community-level capabilities and
creates a networked system to allow for two-way
information flow between and among Federal,
State, and local government public health
authorities and clinical health care providers.
Homeland Security Presidential Directive HSPD-21
Public Health and Medical Preparedness
7
Public Health Need and Utility

• Surveillance systems already exist for
  monitoring influenza-like morbidity.

8
ISDS-CSTE State Survey Project

• 52 Survey Respondents 43 Use Syndromic
  Surveillance

Buehler et al Advances in Disease Surveillance
(2008) www.syndromic.org
9
Public Health Need and Utility

• Why DiSTRIBuTE?
• CDC stated Efforts to improve existing Sentinel
  and Syndromic systems have not succeeded
• - CSTE/CDC influenza surveillance
  working-group, 2007
• - CSTE Position Statement Influenza in the
  U.S. 08-ID-10
• asked local public health agencies to make
  electronic Syndromic surveillance systems work,
  and to share their success.

10
Influenza Morbidity Surveillance Can Be Improved
Current US Sentinel System
Syndromic Surveillance

• Manual reporting
• Labor intensive to establish and maintain
• Small proportion of population sampled
• Cant assess burden
• Static ILI definition and numerators by age only
• Reporting delays and frequent non-reporting
• Only covers 33 weeks per year

• Automated reporting
• Uses existing data existing systems
• Can include nearly all ED visits
• Can assess total impact
• Flexible definitions and categories
• Rapid timely reporting (w/in 24 hours)
• Covers 365 days per year

11
Local ViralLocal Sentinel Local
Syndromic
50 w/in 3 wks
gt90 w/in 24 hrs
12

• Public Health Need and Utility
• Morbidity Mortality time series
• A history of shared aggregates
• William Farr and epidemic flu, London 1847
• Reportable pneumonia influenza. U.S. 1918
  (reportable status dropped, 1952).
• Electronic sentinel ILI reporting, The French
  Experiment 25 years ago (in U.S. sentinel
  morbidity de-emphasized over last decade).

13
Markle Foundations Connecting for Health
Paradigm

• Leverage capabilities of the increasingly wide
  variety of distributed information sources and
  users
• Reduce privacy and security vulnerabilities
• Encourage greater participation among data
  sources and users
• Facilitate more timely and responsive
  decision-making
• Decrease burden on data sources who, under
  current paradigm, are required to report same
  information to multiple sources

14
DiSTRIBuTE Methodology

• Share summarized aggregate counts
• Based on existing local surveillance systems
• Febrile, respiratory or influenza-like syndromes
  reported by age group and broad region

15
DiSTRIBuTE Visualizations - Week 2008-21 (ending
Saturday, May 24, 2008) Time-series depict
respiratory, fever and influenza-like syndrome ED
visits by jurisdiction as percent of total.
16
DiSTRIBuTE Visualizations - Week 2008-21 (ending
Saturday, May 24, 2008) Surface plots
depict relative increase in ED syndrome visits as
observed / baseline by jurisdiction and age.
Region and Age (y)
Date
Relative increase (observed/linear-baseline).
17
Something new in Influenza Surveillance?
18
DiSTRIBuTE Visualizations Oct 1, 2006 Nov 8,
2008 Fever, respiratory and influenza-like
illness emergency department (ED) visits in 8
participating U.S. Health Department syndromic
surveillance systems. Graphs show reported
syndrome visits as proportion of all visits
(top), and age-specific temporal epidemic
response surface (TERS) plots (bottom).
Relative increase (observed/lower-quartile
baseline).
19
https//isds-flu.cirg.washington.edu/aegis/basicGr
aphs.php?modenf-ratioyes
20

• ISDS
• Farzad Mostashari
• Marc Paladini
• Jim Buehler
• Ian Painter
• Bill Lober
• John Brownstein

• DiSTRIBuTE Participants
• Mike Coletta (VA)
• Karl Soetebier (GA)
• Bob Rolfs, Greg Williams (UT)
• Amy Ising, Lana Dayneka (NC)
• Kieran Moore (Ontario)
• Julia Gunn, Justin Pendarvis (Boston)
• Marc Paladini, Erin Murray, Trang Nguyen (New
  York City)
• Atar Baer (Seattle)