in Japan Infectious Disease Surveillance Center, Japan

Transcription

1 GIS for Pandemic Simulation in Japan Infectious Disease Surveillance Center, National Institute of Infectious Diseases, Japan

2 Pandemic Simulation Evidence for preparation planning for pandemic flu Required the most realistic and detail simulation to determine policies 2

3 Brief History pf Mathematical modeling for Pandemic Flu from SIR to Ribm No Geographical Space Hypothetical ti Area Real Data of Location Real Data of Location SIR ibm Ribm Urban Engineering Homogeneous Hypothetical Hypothetical Population Behavior Actual Behavior Behavior Ferguson2005 Longini2005 Germann2006 Ferguson2006 This study Eubank2004 Eubank2005 Ohkusa2006 Glass2006 Shichijo

4 Person-Trip Data of Tokyo Metropolitan Area Data of the location and transportation of 0.9 million persons a day in the Tokyo metropolitan area,which has a population of 33 million, is available about 2% of the population is randomly chosen and is surveyed. This data includes family members, transportation mode, and dlocation. Location was recorded at home, or school, workplace, shopping center..., and the area where these people can be identified in terms of more than 1,600 zones Moreover, it contains the name of the station where they get and get off the train, and the timing. 4

5 7:00 8:00 9:00 10:00 C Station 11:00 BSt Stationti A Station An Example of Person Trip Data 12:00 C Station D Station 13:00 14:00 15:00 CSt Stationti CSt Stationti 16:00 17:00 18:00 19:00 20:00 21:00 CSt Stationti B Station A Station 5

6 Estimated Number of Contact Identifies location of all person in the data in every 6 minutes. Number of contact was estimated as number of persons who were less than 1m distances from patients If data indicates n person in the same area or train or bus, and at the same time, the number of contacts was estimated as n 3.14 / (sampling rate area width in m 2 ) 1m 6

7 Natural lhistory Infection Final day φ S Susceptible I0 Incubation period 1-3 days p 0 I1 Symptomatic period p days δ Recover Death 7

8 Scenario Coming back from the affected area (Day 3) Go to work (Day4 ) Afternoon in Day 5 Go to work (Day 5 ) Visiting a doctor

9 GIS for Pandemic Simulation Understanding the speed of spread intuitively Evaluation and planning of policies Area closure School closure and voluntary staying at home Location of medical facility Simulation for Tokyo metropolitan Effectiveness of area closure Effectiveness of school closure and voluntary staying at home Nation wide simulation Simulation for local city 9

10 Simulation for Tokyo Metropolitan

11 Train Line This area is Pandemic Flu Simulation In Tokyo Metropolitan Area By NIID. 11

12 Day 3 Initial Case H5N1 of 12

13 Day 4 13

14 Day 5 14

15 Day 6 15

16 Day 7 16

17 Day 8 17

18 Day 9 18

19 Day 10 19

20 Area Closure

21 21

22 22

23 Simulation at Fukuoka

24 School Closure and Voluntary Staying at Home

25 Epidemic Curve in Tokyo Num mber of Ne ewly Infect ted No Intervention Voluntary Staying at Home One Day Delay Days after initial case was infected 25

26 No Intervention Day 4 Voluntary Restriction 26

27 No Intervention Day 5 Voluntary Restriction 27

28 No Intervention Day 6 Voluntary Restriction 28

29 No Intervention Day 7 Voluntary Restriction 29

30 No Intervention Day 8 Voluntary Restriction 30

31 No Intervention Day 9 Voluntary Restriction 31

32 No Intervention Day 10 Voluntary Restriction 32

33 No Intervention Day 11 Voluntary Restriction 33

34 Nation Wide Simulation

35 Day 3

36 Day 4

37 Day 5

38 Day 6

39 Day 7

40 Day 8

41 Day 9

42 Day 10

43 Day 11

44 Day 12

45 Day 13

46 Day 14

47 Simulation for Local City

48 Containment at the Local City 8day 9day 10day 11day 12day 13day 14day

49 Conclusion Speed in the geographical dispersion is very fast in a big city Nation wide spread just delays for 2 or 3 days after the initial case was confirmed Therefore area closure seems to be meaningless in the big city However, in the local city, it may possible and may containment On the eother hand, d,school closure eand dvoluntary ystaying gat home is a very powerful countermeasure It can reduce new infections by more than 90%, hopefully we can control the outbreak Conversely, geographic dispersion may not be affected very much 49

50 Area which is included and will be included to the model Included :75 million population Under construction 函館秋田栃木新潟群馬島根 鳥京阪神取福岡中京静岡広島佐賀浜松宮崎 旭川札幌 岩手仙台福島茨城北部首都圏 沖縄

51 Challenges We extend the simulation area to all area where the survey was conducted We also extend the model to the area where the survey was not conducted 51

52 Thank you for your attention 52