1 Mycoplasm a net Mycoplasm a net Building the m odel: Bayesian - - PDF document

1

1 Tina Birk Jensen

Bayesian netw ork A case exam ple

Advanced Herd Management October 3rd 2006 Tina Birk Jensen

2 Tina Birk Jensen

Overview

Mycoplasm a net

• Background
• The model
• Consequences of Mycoplasma
• Effect of control strategies

GRO net

3 Tina Birk Jensen

Mycoplasm a net

Part of the CEPROS project on Animal Health Economics Developed by Lars Otto, FØI

4 Tina Birk Jensen

Mycoplasm a net

Purpose To develop a tool for evaluating the economic consequences of different control strategies against Mycoplasma The decisions are often based on intuition from farmers, advisors and vets

5 Tina Birk Jensen

Mycoplasm a hyopneum onia

Causes enzootic pneumonia in finishers Important in the intensive pig production Can exist latent in the herd Outbreak: 50-70 percent of the finishers have lung lesions A well documented disease

6 Tina Birk Jensen

Mycoplasm a net

The model supports decisions but do not take them Bayesian network The parameters in the model are based on:

• Literature studies
• Expert opinions

2

7 Tina Birk Jensen

Mycoplasm a net

Bayesian net Static model for a single pig herd Stochastic and probabilistic model Make use of Bayes formula for probalistic interference

A

A

A

B

( ) ( ) ( ) ( )

A P B P B A P A B P

i i i

| | =

( ) ( ) ( ) ( )

B P A P A B P B A P

i i i

| | =

8 Tina Birk Jensen

Mycoplasm a net

Building the m odel: Biological m odel:

• Based on biological knowledge

Econom ic m odel:

• Economic risk due to the biological variation

9 Tina Birk Jensen

Risk factors Disease level Productivity Economics Constibution margin

10 Tina Birk Jensen

Risk factors Disease level Productivity Economics Constibution margin Control stragegy Diagnostics

11 Tina Birk Jensen

Mycoplasm a net

I nput Risk factors:

• Herd size
• Production type
• Purchase policy
• Season
• Region

12 Tina Birk Jensen

Mycoplasm a net

I nput Diagnostics:

• Clinical examination
• Serology examination
• Postmortem examination

Determines the disease level with more precision

3

13 Tina Birk Jensen

Mycoplasm a net

I nput Control strategy: (Any action with the aim to change the level of disease) Short term strategy: Medication Vaccination Long term strategy: Change in mangement Buildings

14 Tina Birk Jensen

Mycoplasm a net

Output: Probability distribution:

• Disease level
• Production outcome
• Economical outcome

15 Tina Birk Jensen 16 Tina Birk Jensen

The states of the input variables

States: 1-1000 / 1001-3000 / 3001-5000 / 5001- States: Yes / no States: Sept. / Dec. / Mar. / Jun. States: Zero / low / middle / high States: Zero / low / middle / high States: 0-1% / 1-10% / 10-20% / 20%- States: 0g / -30g / -60g / -90g States: 0 / 0-0.15 / 0.15-0.30 / 0.30-0.45 States: 0% / ½% / 1% / 1½% Risk factors Herd size All in all out Seson Diagnostics Clinical examination Serologic examination Postmortem examination Productivity outcome Change in daily weight gain Change in feed conversion rate Change in mortality

Text Label

17 Tina Birk Jensen

Different scenarios

High-risk herd: Large herd size Continous production Buying piglets from many other herds Low-risk herd: Small herd size Managing all in all out production Not buying piglets from other herds

18 Tina Birk Jensen

4

19 Tina Birk Jensen

Change in productivity Calculation: P0= Ppot + ∆P0 P1= Ppot + ∆P1 P1 - P0 = (Ppot + ∆P1) – (Ppot + ∆P0) P1 - P0 = ∆P1 - ∆P0

20 Tina Birk Jensen

The econom ic part

Contribution margin = TR – VC TR: The income from the slaughterhouse VC: Weaner price Cost of feed Dead pigs Labor cost Control strategy Diagnosis

21 Tina Birk Jensen

Economic loss due to Mycoplasma: High-risk herd

22 Tina Birk Jensen

Economic loss due to Mycoplasma: Low-risk herd

23 Tina Birk Jensen

Effect of m edication

24 Tina Birk Jensen

Effect of vaccination

5

25 Tina Birk Jensen 26 Tina Birk Jensen

Perspectives: Mycoplasm a net

Not all needed information can be found in the literature Describes the steady state between risk factors and the severity of Mycoplasma Often it does not pays off to control just one disease It is important to consider several diseases and their interactions

27 Tina Birk Jensen

GRO net

Grise, risiko og økonom i Further development of the Mycoplasma-net Considers tw o disease com plexes:

• Gastro-intestinal disorders
• Air-way disease

28 Tina Birk Jensen 29 Tina Birk Jensen 30 Tina Birk Jensen

Perspectives: GRO net

Not all information to the model can be found in the literature A challenge to estimate the effect of more than one disease The model will be able to estimate long term control strategies Need to test the model in ”real” herds