Limits...
Detecting differential transmissibilities that affect the size of self-limited outbreaks.

Blumberg S, Funk S, Pulliam JR - PLoS Pathog. (2014)

Bottom Line: Our analysis is based on a branching process model that permits statistical comparison of both the strength and heterogeneity of transmission for two distinct types of cases.When applied to chain size data for MERS-CoV transmission before 2014, our method indicates that despite an apparent trend towards improved control, there is not enough statistical evidence to indicate that R eff has declined with time.Human monkeypox, on the other hand, does not show evidence of differential transmission between animals in contact with humans, primary cases, or secondary cases, which assuages the concern that social mixing can amplify transmission by secondary cases.

View Article: PubMed Central - PubMed

Affiliation: Francis I. Proctor Foundation, University of California San Francisco, San Francisco, California, United States of America; Fogarty International Center, National Institutes of Health, Bethesda, Maryland, United States of America.

ABSTRACT
Our ability to respond appropriately to infectious diseases is enhanced by identifying differences in the potential for transmitting infection between individuals. Here, we identify epidemiological traits of self-limited infections (i.e. infections with an effective reproduction number satisfying [0 < R eff < 1) that correlate with transmissibility. Our analysis is based on a branching process model that permits statistical comparison of both the strength and heterogeneity of transmission for two distinct types of cases. Our approach provides insight into a variety of scenarios, including the transmission of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) in the Arabian peninsula, measles in North America, pre-eradication smallpox in Europe, and human monkeypox in the Democratic Republic of the Congo. When applied to chain size data for MERS-CoV transmission before 2014, our method indicates that despite an apparent trend towards improved control, there is not enough statistical evidence to indicate that R eff has declined with time. Meanwhile, chain size data for measles in the United States and Canada reveal statistically significant geographic variation in R eff, suggesting that the timing and coverage of national vaccination programs, as well as contact tracing procedures, may shape the size distribution of observed infection clusters. Infection source data for smallpox suggests that primary cases transmitted more than secondary cases, and provides a quantitative assessment of the effectiveness of control interventions. Human monkeypox, on the other hand, does not show evidence of differential transmission between animals in contact with humans, primary cases, or secondary cases, which assuages the concern that social mixing can amplify transmission by secondary cases. Lastly, we evaluate surveillance requirements for detecting a change in the human-to-human transmission of monkeypox since the cessation of cross-protective smallpox vaccination. Our studies lay the foundation for future investigations regarding how infection source, vaccination status or other putative transmissibility traits may affect self-limited transmission.

No MeSH data available.


Related in: MedlinePlus

Assessing temporal variation of MERS-CoV transmission in the Arabian Peninsula before Aug 8, 2013.A) The results of estimating the effective reproduction number, , for six inter-related models of transmission are shown. The contours show the 95% confidence regions for three models that allow different values of  for cases occurring before versus after June 1, 2013. The distinction is that each model makes different assumptions about the degree of transmission heterogeneity (as explained in the text). The black dot shows the maximum likelihood estimation (MLE) estimate of the  values for these three models. The dashed grey line indicates when  does not change with time. The dashed colored lines show the MLE estimate and 95% confidence interval of  for the three models that assume transmissibility of cases is independent of time. The slight displacement of the colored lines from the dashed grey line is only for visual clarity. B) The fit of our preferred model to the early MERS-CoV chain size data is shown (Table 1). The error bars of the data correspond to 95% confidence intervals as determined by non-parametric bootstrapping of chain sizes. C) The fit of our preferred model to the late MERS-CoV chain size data is shown.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4214794&req=5

ppat-1004452-g002: Assessing temporal variation of MERS-CoV transmission in the Arabian Peninsula before Aug 8, 2013.A) The results of estimating the effective reproduction number, , for six inter-related models of transmission are shown. The contours show the 95% confidence regions for three models that allow different values of for cases occurring before versus after June 1, 2013. The distinction is that each model makes different assumptions about the degree of transmission heterogeneity (as explained in the text). The black dot shows the maximum likelihood estimation (MLE) estimate of the values for these three models. The dashed grey line indicates when does not change with time. The dashed colored lines show the MLE estimate and 95% confidence interval of for the three models that assume transmissibility of cases is independent of time. The slight displacement of the colored lines from the dashed grey line is only for visual clarity. B) The fit of our preferred model to the early MERS-CoV chain size data is shown (Table 1). The error bars of the data correspond to 95% confidence intervals as determined by non-parametric bootstrapping of chain sizes. C) The fit of our preferred model to the late MERS-CoV chain size data is shown.

Mentions: To avoid artifacts of assembling multiple data sources, we restricted our analysis to the previously reported chain size distribution for all MERS-CoV cases in the Arabian Peninsula occurring before August 8, 2013 [34]. Previous analysis of these data shows that is 0.74 (95% CI 0.53–1.03) before June 1, 2013 and 0.32 (95% CI 0.14–0.65) after June 1, 2013. Our results replicate the finding that independent evaluation of cases before and after June 1, 2013 results in an estimate of 0.7 and 0.3 for respectively (Figure 2 and Table 1). When our six models are compared, we do not find statistical support for models with different values of before and after June 1, 2013. This is again consistent with the results of prior studies that determined a p-value of 0.07 for change in , but our analysis allows the possibility of a high degree of transmission heterogeneity.


Detecting differential transmissibilities that affect the size of self-limited outbreaks.

Blumberg S, Funk S, Pulliam JR - PLoS Pathog. (2014)

Assessing temporal variation of MERS-CoV transmission in the Arabian Peninsula before Aug 8, 2013.A) The results of estimating the effective reproduction number, , for six inter-related models of transmission are shown. The contours show the 95% confidence regions for three models that allow different values of  for cases occurring before versus after June 1, 2013. The distinction is that each model makes different assumptions about the degree of transmission heterogeneity (as explained in the text). The black dot shows the maximum likelihood estimation (MLE) estimate of the  values for these three models. The dashed grey line indicates when  does not change with time. The dashed colored lines show the MLE estimate and 95% confidence interval of  for the three models that assume transmissibility of cases is independent of time. The slight displacement of the colored lines from the dashed grey line is only for visual clarity. B) The fit of our preferred model to the early MERS-CoV chain size data is shown (Table 1). The error bars of the data correspond to 95% confidence intervals as determined by non-parametric bootstrapping of chain sizes. C) The fit of our preferred model to the late MERS-CoV chain size data is shown.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4214794&req=5

ppat-1004452-g002: Assessing temporal variation of MERS-CoV transmission in the Arabian Peninsula before Aug 8, 2013.A) The results of estimating the effective reproduction number, , for six inter-related models of transmission are shown. The contours show the 95% confidence regions for three models that allow different values of for cases occurring before versus after June 1, 2013. The distinction is that each model makes different assumptions about the degree of transmission heterogeneity (as explained in the text). The black dot shows the maximum likelihood estimation (MLE) estimate of the values for these three models. The dashed grey line indicates when does not change with time. The dashed colored lines show the MLE estimate and 95% confidence interval of for the three models that assume transmissibility of cases is independent of time. The slight displacement of the colored lines from the dashed grey line is only for visual clarity. B) The fit of our preferred model to the early MERS-CoV chain size data is shown (Table 1). The error bars of the data correspond to 95% confidence intervals as determined by non-parametric bootstrapping of chain sizes. C) The fit of our preferred model to the late MERS-CoV chain size data is shown.
Mentions: To avoid artifacts of assembling multiple data sources, we restricted our analysis to the previously reported chain size distribution for all MERS-CoV cases in the Arabian Peninsula occurring before August 8, 2013 [34]. Previous analysis of these data shows that is 0.74 (95% CI 0.53–1.03) before June 1, 2013 and 0.32 (95% CI 0.14–0.65) after June 1, 2013. Our results replicate the finding that independent evaluation of cases before and after June 1, 2013 results in an estimate of 0.7 and 0.3 for respectively (Figure 2 and Table 1). When our six models are compared, we do not find statistical support for models with different values of before and after June 1, 2013. This is again consistent with the results of prior studies that determined a p-value of 0.07 for change in , but our analysis allows the possibility of a high degree of transmission heterogeneity.

Bottom Line: Our analysis is based on a branching process model that permits statistical comparison of both the strength and heterogeneity of transmission for two distinct types of cases.When applied to chain size data for MERS-CoV transmission before 2014, our method indicates that despite an apparent trend towards improved control, there is not enough statistical evidence to indicate that R eff has declined with time.Human monkeypox, on the other hand, does not show evidence of differential transmission between animals in contact with humans, primary cases, or secondary cases, which assuages the concern that social mixing can amplify transmission by secondary cases.

View Article: PubMed Central - PubMed

Affiliation: Francis I. Proctor Foundation, University of California San Francisco, San Francisco, California, United States of America; Fogarty International Center, National Institutes of Health, Bethesda, Maryland, United States of America.

ABSTRACT
Our ability to respond appropriately to infectious diseases is enhanced by identifying differences in the potential for transmitting infection between individuals. Here, we identify epidemiological traits of self-limited infections (i.e. infections with an effective reproduction number satisfying [0 < R eff < 1) that correlate with transmissibility. Our analysis is based on a branching process model that permits statistical comparison of both the strength and heterogeneity of transmission for two distinct types of cases. Our approach provides insight into a variety of scenarios, including the transmission of Middle East Respiratory Syndrome Coronavirus (MERS-CoV) in the Arabian peninsula, measles in North America, pre-eradication smallpox in Europe, and human monkeypox in the Democratic Republic of the Congo. When applied to chain size data for MERS-CoV transmission before 2014, our method indicates that despite an apparent trend towards improved control, there is not enough statistical evidence to indicate that R eff has declined with time. Meanwhile, chain size data for measles in the United States and Canada reveal statistically significant geographic variation in R eff, suggesting that the timing and coverage of national vaccination programs, as well as contact tracing procedures, may shape the size distribution of observed infection clusters. Infection source data for smallpox suggests that primary cases transmitted more than secondary cases, and provides a quantitative assessment of the effectiveness of control interventions. Human monkeypox, on the other hand, does not show evidence of differential transmission between animals in contact with humans, primary cases, or secondary cases, which assuages the concern that social mixing can amplify transmission by secondary cases. Lastly, we evaluate surveillance requirements for detecting a change in the human-to-human transmission of monkeypox since the cessation of cross-protective smallpox vaccination. Our studies lay the foundation for future investigations regarding how infection source, vaccination status or other putative transmissibility traits may affect self-limited transmission.

No MeSH data available.


Related in: MedlinePlus