Disease Prevention vs Data Privacy: using landcover maps to inform - PowerPoint PPT Presentation

Disease Prevention vs Data Privacy: using landcover maps to inform spatial epidemic models Mike Tildesley & Sadie Ryan Centre for Complexity Science University of Warwick/ SUNY ESF

Foot-and-Mouth Disease

Foot-and-Mouth Disease - How to deal with diminishing data. For the UK we have demographic information on: • the location and size of all livestock farms • the movement of all animals For the UK we have epidemic information on: • the location of all infected & culled farms •epidemiological tracing of possible infection routes

20km Ring vaccination – optimal ring size to 15km control outbreak? Cumbria – lots of 10km cattle, high vaccination radius. West Lothian – 5km moderate vaccination. The South – no vaccination.

How to deal with diminishing data. For USA we have limited demographic information: • the number of farms in each county • the number of animals in each county • the movement of some animals (very few!) For USA there is no epidemic information.

US Data Number of farms and Number of farms in Data cattle in Alabama Colorado with between 10 and 19 cattle Region United States Alabama Alaska Arizona Arkansas California Colorado INVENTORY (2002) Cattle and calves farms 1018359 27094 113 2838 29925 17379 13311 number 95497994 1437795 12609 841277 1842273 5234177 2656351 Farms by inventory: 1 t o 9 farms 211030 5031 43 915 4278 5348 3252 number 1104588 28642 172 4100 24631 25038 15353 10 t o 19 farms 189193 6244 15 455 5845 3014 1926 number 2607827 87623 217 5905 81634 40285 26249 20 t o 49 farms 271202 8541 23 535 9702 3209 2997 number 8480012 265152 748 16920 304473 96961 94112 50 t o 99 farms 153618 4036 10 278 5412 1426 1741 number 10648669 274879 703 18556 370887 97596 121413 100 t o 199 farms 103513 2040 12 253 3023 1092 1479 number 14078261 270680 1679 34796 400159 149773 203188 200 t o 499 farms 62774 915 5 190 1367 1215 1120 number 18563597 262901 1177 57884 389844 386061 338206 500 or more farms 27029 287 5 212 298 2075 796 number 40015040 247918 7913 703116 270645 4438463 1857830 No XY locations or animals numbers at the individual farm level are given.

Can we use geographic data to infer farm locations? Would epidemic models utilising such data be useful as predictive tools?

THE MODEL Model developed during 2001 to investigate optimal control strategies for FMD in the UK. Probability of infection for every susceptible farm for each day of the epidemic is calculated using the formula: Σ [ qc + T s N s,j ] Prob = 1 - exp - [ S c N c,i pc + S s N s,i ps ] [ T c N c,j qs ] K ( dij ) i Infected j

Σ [ ] Prob = 1 - exp - [ S c N c,i qc + T s N s,j pc + S s N s,i ps ] qs ] K ( dij ) [ T c N c,j i Infected j Wales S c - susceptibility of cows. T c , T s and S c (and associated powers) are N c,i - number of cows on farm i. Mixture of long and short estimated using the 2001 range transmission UK epidemic data. T c - transmissibility of cows (or how much a cow “transmits” infection). K ( d ij ) - the distance kernel. A parameter which weights the probability of infection based on the distance between farm i and farm j . Transmission Risk Kernel Distance from source

In many countries, precise locations of farms are not available. It may be possible to capture farm demography using other data in the public domain. We use land cover in the UK to determine whether we can accurately predict farm locations.

Farm Locations We use Land Cover Map 2000 to obtain surrogate farm locations for UK livestock farms (to compare with the 2000 UK Agricultural Census). Land Cover Map 2000 defines land use in one of 10 aggregate classes: Our aim is to determine land cover classes that will 1-2. Woodland. potentially correlate with livestock farm locations. 3. Arable and Horticulture. YES! 4-5. Grassland. 6. Mountain/Heath/Bog. SHEEP FARMS? 7. Urban. Land use data is available in parcels 8-10. Water/Coastal. of 25 square metres.

Farm Locations This gives us two sets of surrogate farm databases: LC1 – farms are only located within land cover classes 4 and 5. LC2 – farms are only located within land cover classes 4, 5 and 6. Aggregate Class Sub class 4. Improved Grassland. 14. Improved Grassland 5. Semi-Natural Grass 15. Neutral Grass 16. Setaside Grass 17. Bracken 18. Calcareous Grass 19. Acid grassland 20. Fen, Marsh, Swamp

Farm Locations This gives us two sets of surrogate farm databases: LC1 – farms are only located within land cover classes 4 and 5. LC2 – farms are only located within land cover classes 4, 5 and 6. Aggregate Class Sub class 4. Improved Grassland. 14. Improved Grassland 15. Neutral Grass 5. Semi-Natural Grass 6. Mountain/Heath/Bog. 8. Bog 9. Dense heath 10. Open heath 11. Mountain

Farm Locations This gives us four sets of surrogate farm databases: LC1 – farms are only located within land cover classes 4 and 5. LC2 – farms are only located within land cover classes 4, 5 and 6. Aggregate Class Sub class 4. Improved Grassland. 14. Improved Grassland 15. Neutral Grass 5. Semi-Natural Grass 6. Mountain/Heath/Bog. 8. Bog 9. Dense heath 10. Open heath 11. Mountain LC3 – farms are only located within sub classes 14 and 15. LC4 – farms are only located within sub classes 9, 10, 11, 14 and 15.

Disease Simulations We generate surrogate farm databases assuming the same number of farms in a county as in the recorded data. Farms are only located in accepted land cover types. Cumbria Clwyd We generate surrogate data for three counties in the UK. Devon 100 surrogate farm databases are generated for each land cover type (LC1-LC4) and epidemics are simulated. Epidemic sizes and optimal control policies are determined and compared with that predicted by the “truth data”.

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181)

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181) Random 196 (11-841) 168 (15-649) 63 (10-209)

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181) Random 196 (11-841) 168 (15-649) 63 (10-209) LC1 608 (15-1544) 566 (19-1889) 106 (11-383)

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181) Random 196 (11-841) 168 (15-649) 63 (10-209) LC1 608 (15-1544) 566 (19-1889) 106 (11-383) LC2 481 (16-1489) 479 (17-1584) 90 (10-323)

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181) Random 196 (11-841) 168 (15-649) 63 (10-209) LC1 608 (15-1544) 566 (19-1889) 106 (11-383) LC2 481 (16-1489) 479 (17-1584) 90 (10-323) LC3 1880 (17-2908) 509 (18-1882) 481 (13-1164)

Model Results Mean Epidemic Size (95% CIs) Cumbria Devon Clwyd True Data 2320 (100-2908) 480 (15-1891) 700 (12-1181) Random 196 (11-841) 168 (15-649) 63 (10-209) LC1 608 (15-1544) 566 (19-1889) 106 (11-383) LC2 481 (16-1489) 479 (17-1584) 90 (10-323) LC3 1880 (17-2908) 509 (18-1882) 481 (13-1164) LC4 1563 (15-2669) 419 (17-1565) 246 (11-799)

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd True Data 3.6 2.8 3.6

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd True Data 3.6 2.8 3.6 Random 1.6 [1358] 1.2 [595] 0.8 [291]

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd True Data 3.6 2.8 3.6 Random 1.6 [1358] 1.2 [595] 0.8 [291] LC1 2.8 [199] 2.0 [181] 1.3 [232]

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd True Data 3.6 2.8 3.6 Random 1.6 [1358] 1.2 [595] 0.8 [291] LC1 2.8 [199] 2.0 [181] 1.3 [232] LC2 2.8 [199] 1.8 [227] 1.4 [221]

Optimal Control – Ring Culling Δ Emax is the maximum increase in epidemic size if we applied the optimal ring cull as predicted by the surrogate database. Optimal Ring Cull Radius (km) [Δ E max ] Cumbria Devon Clwyd True Data 3.6 2.8 3.6 Random 1.6 [1358] 1.2 [595] 0.8 [291] LC1 2.8 [199] 2.0 [181] 1.3 [232] LC2 2.8 [199] 1.8 [227] 1.4 [221] LC3 3.5 [17] 2.8 [0] 3.5 [12]