Household perceptions of risk as drivers for FMD vaccination

Transcription

1 Household perceptions of risk as drivers for FMD vaccination Ashley Railey Washington State University Tiziana Lembo Boyd Orr Centre for Population and Ecosystem Health; Institute of Biodiversity, Animal Health and Comparative Medicine; University of Glasgow, Scotland Guy Palmer Paul G Allen School for Global Animal Health, Washington State University, USA Gabriel Shirima Nelson Mandela African Institute of Science and Technolology, Tanzania Tom Marsh Paul G Allen School for Global Animal Health, Washington State University, USA

2 FMD in Tanzania 50% of households own livestock* Open-grazing FMD is endemic 4 of 7 serotypes in Tanzania Up to 3 outbreaks a year (Casey and Reeve et al., 2018) Vaccines NOT matched to serotypes Targeted vaccination (Casey and Reeve et al., 2018) What drives household FMD vaccination decisions? *2015 Tanzanian Livestock Modernization Initiative 2 of 10

3 Problem Uncertainty in decision-making Human health decisions* Immediacy and proximity of risk (Brook & McLachlan 2006) Negative past experiences with FMD vaccines in Tanzania Vaccination for FMD may improve household wellbeing Household income and wealth (Marsh et al., 2016) Production losses (Casey and Reeve et al., 2018) *see seasonal influenza literature (Bloom et al 2017; Galvani, Reluga & Chapman 2007) 3 of 10

4 Survey Serengeti KENYA April 2016-July 2016 Multi-stage random sampling 432 households TANZANIA Ngorongoro National parks Study Districts Study Sites 4 of 10

5 Descriptive Statistics Household characteristics 80% male head of household 85% have some formal education 46% off-farm income 38% crop income 63% sold cattle (past year) Livestock characteristics 42 cows (median 20) 91% produce milk Disease Factors 70% reported FMD in past year 19% vaccinated livestock 5 of 10

6 Methodology 1) What drives FMD vaccination decisions? Stated preference willingness to pay- Derived from individual preferences, not the market Hypothetical Double-Bounded Contingent Valuation (Hanemann, Loomis, & Kanninen, 1991) One cow, 6 mo duration If yes Q1 Initial Price Q2 Higher price If no Q2 Lower Price 6 of 10

7 Methodology 2) How are vaccination decisions influenced by proximity of perceived risk? Routine vaccination Biannual, planned application Population-level protection Lower perceived risk of infection Emergency vaccination Spatial and temporal immediacy Individual-focused protection Higher perceived risk of infection Vaccine efficacy: 50 or 100 percent Outbreak distance: village or neighbor 7 of 10

8 Predicted WTP Distributions Amount in Tanzanian Shillings (2100Tsh~USD 1) 8 of 10

9 Table 3 Vaccination Determinants Variable Routine Marginal Effects (CI 95%) P value Emergency Marginal Effects (CI 95%) P value Education (0=Formal; 1=No Formal) 681 (-7,1356) (-369,1679) Income Off-Farm ( 25,000 Tsh) Base Case Off-Farm (25-100,000) 589 (-34,1213) (835,3090) Off-Farm (>100,000) 1022 (360,1685) (672,2854) Crops ( 100,000 Tsh) Base Case Crops ( ,000) 1635 (806,2465) (1034,3554) Crops (>500,000) -445 (-1067,176) (-1513,3554) Herd Size 26 (-192,243) (-348,432) Expected Milk Loss (in liters per cow) 306 (-94,707) (-205,1051) Cattle sold in past year 36 (-.33,71) (-48,71) FMD experience in past year (0=No; 1=Yes) -241 (751,270) (-1156,590) Vaccinated for any cattle disease in past year (0=No; 1=Yes) -247 (-795,299) (-754,1186) Use of government vet (0=No; 1=Yes) -663 (-1113,-214) (-2626,-1008) Vaccine efficacy (0=100%; 1=50%) 1573 (370,2778) (107,4529) Gender (0=Female; 1=Male) 1031 (321,1740) (-478,2192) Gender*efficacy (0=100%; 1=50%) (-2740,-174) (-5066,406) District (0=Ngorongoro; 1=Serengeti) -270 (-751,212) (-779,967) Outbreak (0=Village; 1=@Neighbor) -476 (-1244,293) Log Likelihood Chi-2 Statistic Log of variable USD 1.00=2100 Tanzanian shillings AF Railey, T Lembo, GH Palmer, G Shirima, and TL Marsh (2018). Spatial and temporal risk as drivers for adoption of foot and mouth disease vaccination. Vaccine 36,

10 How do we increase vaccination uptake? Demonstrate field effectiveness Better quality vaccines and supply chains Improve surveillance and household participation Predict outbreaks in space and time Contributions to community control? Market assessment 10 of 10

11 Acknowledgements UNIVERSITY SUPPORT: Dr. Thomas L. Marsh (Washington State University) Dr. Timothy Baszler (Washington State University) Dr. Thomas Rotolo (Washington State University) Dr. Felix Lankester (Washington State University) Dr. Tiziana Lembo (University of Glasgow) Dr. Gabriel Shirima (Nelson Mandela African Institute of Science and Technology) FUNDING: Bill and Melinda Gates Foundation Special Thanks: Administrative and grant assistance: Rose Mosha Sara Pepper and Allen School staff Dr. Guy Palmer FIELD TEAM: Emmanuel Sindoya, Victor Sianga, Isaya Ole Seki, Jackson Dudui, Loserian Ole Maoi, Timas Yohana, Sunguro Matonge, Ryoba Makorere, Mayaya Masunga, Ahmed, and Zilpah. Linus Otieno Program for Enhancing the Health and Productivity of Livestock (PEHPL) MSD Animal Health Tanzania Research Platform