Transcript Texas ILINet Structure and Operation 2002-2008

Texas ILINet Structure and Operation
2002-2008
86th Annual Texas Public Health Association Conference
April 21-23, 2010
South Padre Island, Texas
Gary Heseltine MD MPH
Infectious Disease Control Unit
Chronic Illnesses Demand Chronic Attention
Overview
• What is ILINet
• What is the structure of ILINet
– Where does the data come from
– Who is represented
• How does ILINet operate
– Data handling
– Stability of operation
• Possible Improvements to ILINet
Influenza Like Illness (ILI) Net
• CDC hosted syndromic reporting system
– Fever > 100o F and cough or sore throat
• Clinicians report weekly
– Total number patient visits
– Total number of patients diagnosed with ILI
• Cases divided into 4 age strata
• Statewide weekly ILI index
– Combine data from all clinicians
ILIindex= total ILI patients/total patient visits
Texas ILI Index: What Does It Mean?
ILINet Evolution
Year
Reporters
2002
13
1,237
79,852
9
1.44
2003
44
8,232
184,848
31
1.42
2004
69
18,244
517,845
33
2.09
2005
101
24,020
696,806
42
2.40
2006
111
39,607
1,018,809
52
2.13
2007
123
41,963
1,018,790
61
2.02
2008
124
34,335
1,041,898
58
2.14
167,638
4,558,848
Total
ILI Patients Total Visits Counties Report/County
Cumulative Proportion of Total Patients Visits
by Reporter 2002-08
1
0.9
0.8
Reporters = 224
0.7
Total Visits = 4,558,848
0.6
A single reporter contributed 29% total visits
0.5
Six reporters contributed 51% total visits
0.4
0.3
0.2
0.1
0
1
11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181 191 201 211 221
Reporters
Total Number of Weekly Submissions by Reporter
2002-2008
400
Reporters = 224
350
Total Submissions = 15,494
300
250
200
Mean Submissions/Reporter 69.2
Range 365 to 1
39 reporters provided 50.2% total submissions
150
100
50
0
1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181 191 201 211 221
Weighting
8
Tuning Surveillance Levels
Weightcounty
Data Viewcounty
Weightblock
Data Viewblock
ILINet Data: Two Issues
1. ILINet data reported to the state does not reflect
county populations
–
–
2.
Weight county contributions to reflect underlying
population proportion
Use sampling index to guide resource allocation and
recruiting for ILINet to obtain proportionality
Sampling method itself: non-probability
–
–
–
Clinicians (reporters) are volunteers- bias
Convenience sample - bias
Use a probability sample to correct for ILINet sampling
bias before submission to the state