Listeria monocytogenes (Stillbirths): Dose Response Models

From QMRAwiki
Revision as of 03:48, 12 October 2012 by KaitlynDelancey (talk | contribs) (Summary Data)
Jump to: navigation, search

Listeria monocytogenes (Stillbirths in animals)

Author: Sushil Tamrakar


General overview of Listeria monocytogenes

Listeria monocytogenes is a gram-positive rod-shaped bacterium. It is the causative agent of listeriosis, a serious infection caused by eating food contaminated with the bacteria. The disease generally affects older adults, pregnant women, newborns, and adults with weakened immune systems. However, rarely, persons without these risk factors can also be affected [1] .

The overt form of the disease has mortality greater than 25 percent. The two main clinical manifestations are sepsis and meningitis. Meningitis is often complicated by encephalitis, a pathology that is unusual for bacterial infections [1].




Summary Data

Smith et al.(2008)[2] studied dose-response model for Listeria monocytogenes- induced stillbirths in nonhuman primates and Williams et al ( 2007 and 2009)[3] explored fetal mortality in guinea pigs after oral exposure.


Experiment serial number Reference Host type Agent strain Route # of doses Dose units Response Best fit model Optimized parameter(s) LD50/ID50
289,290* [2][3] pooled oral 13 CFU stillbirths beta-Poisson α = 4.22E-02 , N50 = 1.78E+09 1.78E+09
289 [2] rhesus monkeys oral 8 CFU stillbirths beta-Poisson α = 4E-02 , N50 = 8.26E+08 8.26E+08
290 [3] guinea pig oral 5 CFU fetal mortality beta-Poisson α = 9.36E-02 , N50 = 4.67E+07 4.67E+07
*This model is preferred in most circumstances. However, consider all available models to decide which one is most appropriate for your analysis.


Exponential and betapoisson model.jpg

Optimization Output for pooled data (experiment 289 and 290)

Rhesus monkey/Listeria monocytogenes
Dose Stillbirths Not stillbirths Total
316 0 1 1
1580 2 6 8
1E+04 0 4 4
2E+04 0 3 3
1E+05 2 9 11
126000 1 4 5
1E+06 2 7 9
1580000 2 4 6
1E+07 3 6 9
1.26E+07 2 3 5
1E+08 3 1 4
1.26E+08 2 2 4
3.98E+10 1 0 1


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 71.8 63.9 12 3.84
1.33e-15
21
1.49e-10
Beta Poisson 7.88 11 19.7
0.724
Beta-Poisson fits better than exponential; cannot reject good fit for beta-Poisson.


Optimized parameters for the beta-Poisson model, from 500 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
α 4.22E-02 1.14E-02 1.54E-02 1.78E-02 1.26E-01 1.51E-01 2.71E-01
N50 1.78E+09 1.74E+06 4.42E+06 7.04E+06 2.28E+15 8.63E+16 8.36E+20


Parameter scatter plot for beta Poisson model ellipses signify the 0.9, 0.95 and 0.99 confidence of the parameters.
beta Poisson model plot, with confidence bounds around optimized model




Optimization Output for experiment 289

Rhesus monkey/Listeria monocytogenes [2]
Dose Stillbirths Not stillbirths Total
316 0 1 1
1580 2 6 8
2E+04 0 3 3
126000 1 4 5
1580000 2 4 6
1.26E+07 2 3 5
1.26E+08 2 2 4
3.98E+10 1 0 1


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 51.4 47.7 7 3.84
4.95e-12
14.1
7.71e-09
Beta Poisson 3.68 6 12.6
0.72
Beta-Poisson fits better than exponential; cannot reject good fit for beta-Poisson.
Optimized parameters for the beta-Poisson model, from 10000 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
α 4E-02 9.86E-04 9.94E-04 1.02E-02 1.22E-01 1.96E-01 3.05E-01
N50 8.26E+08 1.94E+03 4.54E+04 5.30E+05 8.90E+18 6.18E+22 4.86E+65



Parameter scatter plot for beta Poisson model ellipses signify the 0.9, 0.95 and 0.99 confidence of the parameters.
beta Poisson model plot, with confidence bounds around optimized model



Optimization Output for experiment 290

Guinea pig/Listeria monocytogenes [3]
Dose Fetal mortality Not fetal mortality Total
1E+04 0 4 4
1E+05 2 9 11
1E+06 2 7 9
1E+07 3 6 9
1E+08 3 1 4


Goodness of fit and model selection
Model Deviance Δ Degrees
of freedom
χ20.95,1
p-value
χ20.95,m-k
p-value
Exponential 19.3 17.6 4 3.84
2.71e-05
9.49
0.000677
Beta Poisson 1.72 3 7.81
0.632
Beta-Poisson fits better than exponential; cannot reject good fit for beta-Poisson.


Optimized parameters for the beta-Poisson model, from 10000 bootstrap iterations
Parameter MLE estimate Percentiles
0.5% 2.5% 5% 95% 97.5% 99.5%
α 9.36E-02 1.19E-02 2.03E-02 2.95E-02 1.10E+00 1.02E+02 8.41E+02
N50 4.67E+07 9.56E+05 2.12E+06 3.23E+06 8.13E+11 4.28E+14 2.83E+23



Parameter scatter plot for beta Poisson model ellipses signify the 0.9, 0.95 and 0.99 confidence of the parameters.
beta Poisson model plot, with confidence bounds around optimized model



Summary

The risk of fetal mortality in nonhuman primate ( rhesus monkeys) and guinea pigs statistically same and hence the pooled model with larger data points is recommended model.

References

  1. 1.0 1.1 Todar K (2012) Todar's Online Textbook of Bacteriology.Todar
  2. 2.0 2.1 2.2 2.3 Smith, M. A., K. Takeuchi, et al. (2008). "Dose-Response Model for Listeria monocytogenes-Induced Stillbirths in Nonhuman Primates." Infection and Immunity 76(2): 726-731
  3. 3.0 3.1 3.2 3.3 Williams, D., E. A. Irvin, et al. (2007)."Dose-Response of Listeria monocytogenes after Oral Exposure in Pregnant Guinea Pigs." Journal of Food Protection 70(5): 1122-1128


Williams D, Castleman J, et al. (2009) Risk of Fetal Mortality After Exposure to Listeria monocytogenes Based on Dose-Response Data from Pregnant Guinea Pigs and Primates. Risk Analysis 29(11): 1495-1505.