DRAFT – Not for a.ribu2on or distribu2on Modeling the Ebola Outbreak in West Africa, 2014 December 16 th Update Bryan Lewis PhD, MPH ([email protected]) presen2ng on behalf of the Ebola Response Team of Network Dynamics and Simula2on Science Lab from the Virginia Bioinforma2cs Ins2tute at Virginia Tech Technical Report #14130
44
Embed
Modeling the Ebola Outbreak in West Africa, December 16th 2014 update
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
DRAFT – Not for a.ribu2on or distribu2on
Modeling the Ebola Outbreak in West Africa, 2014
December 16th Update
Bryan Lewis PhD, MPH ([email protected]) presen2ng on behalf of the Ebola Response Team of
Network Dynamics and Simula2on Science Lab from the Virginia Bioinforma2cs Ins2tute at Virginia Tech
Technical Report #14-‐130
DRAFT – Not for a.ribu2on or distribu2on
NDSSL Ebola Response Team Staff: Abhijin Adiga, Kathy Alexander, Chris Barre., Richard Beckman, Keith Bisset, Jiangzhuo Chen, Youngyoun Chungbaek, Stephen Eubank, Sandeep Gupta, Maleq Khan, Chris Kuhlman, Eric Lofgren, Bryan Lewis, Achla Marathe, Madhav Marathe, Henning Mortveit, Eric Nordberg, Paula Stretz, Samarth Swarup, Meredith Wilson,Mandy Wilson, and Dawen Xie, with support from Ginger Stewart, Maureen Lawrence-‐Kuether, Kayla Tyler, Kathy Laskowski, Bill Marmagas Students: S.M. Arifuzzaman, Aditya Agashe, Vivek Akupatni, Caitlin Rivers, Pyrros Telionis, Jessie Gunter, Elisabeth Musser, James Schli., Youssef Jemia, Margaret Carolan, Bryan Kaperick, Warner Rose, Kara Harrison
2
DRAFT – Not for a.ribu2on or distribu2on
Currently Used Data
● Data from WHO, MoH Liberia, and MoH Sierra Leone, available at h.ps://github.com/cmrivers/ebola
● MoH and WHO have reasonable agreement ● Sierra Leone case counts censored up
to 4/30/14. ● Time series was filled in with missing
dates, and case counts were interpolated.
3
Cases Deaths Guinea 2,292 1,428 Liberia 7,797 3,177 Sierra Leone 8,273 1,768 Total 17,608 6,055
• Re-‐run full study with updated simula2on engine
• Analyze transmission tree impact
• Calibrate Sierra Leone – A.empt geographic spread – Run similar prelim study
24
DRAFT – Not for a.ribu2on or distribu2on
POPULATION CONSTRUCTION DETAILS
25
DRAFT – Not for a.ribu2on or distribu2on
Four versions of the Liberia contact network: § Base version: LBR-‐base: ini2al version constructed using the base pipeline
§ Long distance travel: LBR-‐ldt: base version augmented with links corresponding to contacts arising through travel using FlowMinder data.
§ LBR-‐2-‐group and LBR-‐9-‐group: LBR-‐2gp and LBR-‐9gp: Versions based on the Liberia Labor Force Survey.
Synthe2c popula2ons: Liberia
DRAFT – Not for a.ribu2on or distribu2on
Why 4 versions? § LBR-‐base: first version constructed with data that we had found in ini2al data search
§ LBR-‐ldt: the base version construc2on methodology is improved through addi2on of contacts corresponding to long distance travel. FlowMinder data was used to es2mate such contacts and the social contact network of the base version was augmented accordingly.
Synthe2c popula2ons: Liberia
DRAFT – Not for a.ribu2on or distribu2on
Why 4 versions (con2nued)? § LBR-‐2gp and LBR-‐9gp: Aner the construc2ons of the LIB-‐base and LIB-‐ldt social contact networks, we obtained the Liberia Labor Force Survey from 2010. Demographic informa2on from this survey was used to construct these two addi2onal versions that are more closely calibrated against the reported, aggregated 2me-‐use data of this survey. The two versions (2gp/9gp) differ in the number of demographic sub-‐groups (2 and 9) that were used in the calibra2on.
Overall significance: data-‐responsiveness of pipeline; calibra2on, verifica2on and valida2on.
The Problem Road condi2ons in southern Africa are variable and severe.
In order to win the fight against Ebola, it will be necessary to transport medical supplies and pa2ents as efficiently as possible.
DRAFT – Not for a.ribu2on or distribu2on
The Solu2on Eyes on the Ground: a web-‐based tool for tracking road condi2ons. Witnesses report road condi2ons as they encounter them Travelers can then use recent and historical informa2on to plan the best route to help.
DRAFT – Not for a.ribu2on or distribu2on
Future Enhancements
DRAFT – Not for a.ribu2on or distribu2on
Future Reports
DesMnaMon Last Reported
Travel Time
Traffic CondiMons
Road CondiMons
Comments
Bopolu (Gbarpolu)
2014-‐12-‐11
160 Light Passable
Yangaryah (Gbarpolu)
2014-‐12-‐14
190 Heavy Passable
Mecca (Bomi) 2014-‐12-‐10
210 Medium Passable
Tubmanburg (Bomi)
2014-‐12-‐08
240 Medium Passable 4-‐wheel drive
Gbah Jakeh (Bomi)
2014-‐11-‐03
280 Medium Passable
Parker Cornor (Montserrado)
2014-‐12-‐01
300 Heavy Passable
Sinje (Grand Cape Mount)
2014-‐11-‐30
Impassable
DRAFT – Not for a.ribu2on or distribu2on
APPENDIX Suppor2ng material describing model structure, and addi2onal results
37
DRAFT – Not for a.ribu2on or distribu2on
Legrand et al. Model Descrip2on
Exposednot infectious
InfectiousSymptomatic
RemovedRecovered and immune
or dead and buried
Susceptible
HospitalizedInfectious
FuneralInfectious
Legrand, J, R F Grais, P Y Boelle, A J Valleron, and A Flahault. “Understanding the Dynamics of Ebola Epidemics” Epidemiology and Infec1on 135 (4). 2007. Cambridge University Press: 610–21. doi:10.1017/S0950268806007217.
38
DRAFT – Not for a.ribu2on or distribu2on
Compartmental Model
• Extension of model proposed by Legrand et al. Legrand, J, R F Grais, P Y Boelle, A J Valleron, and A Flahault. “Understanding the Dynamics of Ebola Epidemics” Epidemiology and Infec1on 135 (4). 2007. Cambridge University Press: 610–21. doi:10.1017/S0950268806007217.
39
DRAFT – Not for a.ribu2on or distribu2on
Legrand et al. Approach
• Behavioral changes to reduce transmissibili2es at specified days
• Stochas2c implementa2on fit to two historical outbreaks – Kikwit, DRC, 1995 – Gulu, Uganda, 2000
• Finds two different “types” of outbreaks – Community vs. Funeral driven outbreaks
40
DRAFT – Not for a.ribu2on or distribu2on
Parameters of two historical outbreaks
41
DRAFT – Not for a.ribu2on or distribu2on
NDSSL Extensions to Legrand Model
• Mul2ple stages of behavioral change possible during this prolonged outbreak
• Op2miza2on of fit through automated method
• Experiment: – Explore “degree” of fit using the two different outbreak types for each country in current outbreak
42
DRAFT – Not for a.ribu2on or distribu2on
Op2mized Fit Process • Parameters to explored selected – Diag_rate, beta_I, beta_H, beta_F, gamma_I, gamma_D, gamma_F, gamma_H
– Ini2al values based on two historical outbreak • Op2miza2on rou2ne
– Runs model with various permuta2ons of parameters
– Output compared to observed case count
– Algorithm chooses combina2ons that minimize the difference between observed case counts and model outputs, selects “best” one
43
DRAFT – Not for a.ribu2on or distribu2on
Fi.ed Model Caveats
• Assump2ons: – Behavioral changes effect each transmission route similarly
– Mixing occurs differently for each of the three compartments but uniformly within
• These models are likely “overfi.ed” – Many combos of parameters will fit the same curve – Guided by knowledge of the outbreak and addi2onal data sources to keep parameters plausible