Upload
alice-coghill
View
213
Download
1
Embed Size (px)
Citation preview
Estimating the best way forward: Expert and farmer evaluations of environmental interventions to reduce human exposure to E. coli O157
Paul Cross, Dan Rigby & Gareth Edwards-Jones
RELU funded ‘Reducing Escherichia coli O157 risk in rural communities’
Uncertainty & O157 Management
Uncertainty regarding:
How people become infected (sporadic infection)
Effectiveness of measures
Likelihood of measures being adopted
Is there a problem?
Many potential measures+
Absence of hard (e.g. RCT) evidence on measures to reduce risk+
A (perceived) need to act
= a problem
Managing Uncertainty
1. Identify all possible interventions2. Elicit ‘expert’ opinion on interventions
Aim: Identify best candidate interventionsIdeally =
highly effective+highly practical
Managing Uncertainty
Which experts? Experts (effectiveness)
Inter alia; Public Health, Veterinary Microbiology (Food), Microbiology (Agricultural/Environmental/Clinical), Risk Assessment, Business, Land Management
Farmers (practicality) Members of farmers unions in Wales and Scotland. Farmers’ markets
How to elicit their views? Novel method: Best Worst Scaling
Best-Worst Scaling
Market research tool
Possible to carry out over distance; no face to face; anonymous
Multiple choice based Scaled, fine resolution results Allows respondents to rank long lists without the associated
cognitive gymnastics (bite-sized chunks)
Best-Worst Scaling
Mosteffective
Leasteffective
A Encourage Farmers and farm visitors to wash hands
following contact with farm animals.
B Vaccinate cattle to control pathogen colonisation and
faecal excretion of E. coli O157.
C Remove farm animals from proximity of private water
supplies (e.g. at least 50m from well, borehole or other private water supply by fencing-off).
D Prevent children under the age of 11, and other
vulnerable groups, coming into contact with animals at petting, or public visitor farms.
E No application of manure to ready-to-eat crops within 12
months of harvest and 6 months of drilling/planting.
Best-Worst Scaling analysis
Take all the “most effective” & “least effective” choices
5 item set we gain information on 7of the 10 paired combinations
Respondent chooses the two measures with the maximum difference in performance (best and worst)
Maximises the ability to predict peoples choices
Intervention generation
Literature review Non-systematic Published and grey literature included Project members Opportunity for experts to add interventions to list
Respondent sample
Results of the expert elicitationExperts (Effectiveness)Round 1 Contacted 53 experts 31(75%) completed survey Reduced initial list of 99 to 30Round 2 Contacted 70 experts 41 (60%) completed survey of 30 interventions
Famers (Practicality)Round 3 50 in Wales 50 in Scotland
CV scores: a measure of agreement?
Experts had higher CV scores than farmers for ‘effectiveness’ (p<0.001) and ‘practicality’ (p<0.01)
Farmer CV mean scores were very similar between Scotland and Wales.
The mean confidence interval for the practicality assessment was 0.86 for farmers and 1.46 for experts (p<0.001) and for effectiveness was 0.94 and 1.35 respectively (p=0.002).
Double fencing
Vaccination
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 300
1
2
3
4
5
6
7
8
9
10
Daily cleaning of water troughs
Hand washing
Effectiveness scores
Is there consensus?
-6 -4 -2 0 2 4 6 8 10
-10
-8
-6
-4
-2
0
2
4
Practicality
Effectiveness
Vaccination (Intervention 19)
Is there consensus?
-1 0 1 2 3 4 5
-4
-2
0
2
4
6
8
Practicality
Effectiveness
Reducing stocking densities by 50%
Is there consensus?
-6 -4 -2 0 2 4 6
-4
-2
0
2
4
6
8
Practicality
EffectivenessP &E Scores are widely distributed indicating poor agreement amongst respondents
16. HACCP for manure handling
Practicality scores
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 300
1
2
3
4
5
6
7
8
9
Vaccination
Hand washing
Reduce cattle stocking densities by 50%.
Scottish and Welsh farmers (practicality)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 300
1
2
3
4
5
6
7
8
9
Wales
Scotland
Intervention
Mea
n sc
ore
Locate manure >50m
-3 -2 -1 0 1 2 3 4 5
-4
-3
-2
-1
0
1
2
3
4
5
6
1
2 3
4
5
6
7
89
10
11
12
13
14 15
16
17 18
19
20
21
2223
24
25
2627
282930
Practicality
Effectiveness
19: Vaccination of cattle
1: Hand washing
27: Reduce cattle by 50%
12: Septic tank leakage High practicality/
low effectivenessLow practicality/ low effectiveness
Low practicality/ High effectiveness
High practicality/ High effectiveness
Best-Worst Scaling 2 x 2 plots
Method suitability
Best suited to the evaluation of large sets of standalone measures
Best suited to the evaluation of multidimensional measures (effectiveness and practicality)
Smaller multi-level sets of measures, identify the ‘best’ bundles (ACA, CBC)
Future
Applications in other areas E.g. Cryptosporidium, Salmonella, anti-microbial resistant genes
Modelling of interventions % reduction in disease prevalence
Development of MACCs % reduction/cost
Participating institutions
Health Protection Agency; Health Protection Scotland; University of Glasgow: School of Veterinary Medicine; Scottish Agricultural Ccollege; Veterinary Laboratory Agency; Scottish Infection Research Network; HPS Colindale; FSA: Microbiological Safety Division; Bioss Scotland; Wageningen University and Research Centre; Liverpool John Moores University: Centre for Public Health; NFUScotland; NFUCymru; Farmers Union Wales; Office of the Chief Veterinary Officer Wales; Meat Promotion Wales; Quality Meat Scotland; Countryside Council Wales
Top interventions by effectiveness and practicality
No. Intervention1 Encourage Farmers and farm visitors to wash hands following contact with farm animals.
4 Prohibit recreational activities (such as walking and camping) to land where manure, slurry or abattoir waste have been applied, or animals and faeces present, in the previous four weeks.
5 Monitoring of private water supplies to identify those with either high indicator counts, or those in areas of high risk. These supplies would need to be treated (e.g. by ozonation, chlorination or ultra-violet treatment).
7 Keep livestock and pets out of ready-to-eat crop areas, using fencing for example.
8 No application of manure to land at high risk of direct flow to watercourses (e.g. adjacent to a watercourse, borehole or road culvert, or areas with a dense network of open drains.
9 Locate solid manure heaps and slurry pits at least 50m away from watercourses, field drains and ready-to-eat crops.
13 No slurry or livestock manure to be applied to high risk fields (ie high risk of transport into adjacent areas watercourses (e.g. when soils saturated or frozen, or heavy rain expected).
19 Vaccinate cattle to control pathogen colonisation and faecal excretion of E. coli O157.