Using Mathematical Models to Assess Responses to an Outbreak of an Emerged Viral Respiratory Disease

6.1 Extent to which protecting health care workers affects transmission

Page last updated: April 2006

One way to examine the effect that protecting HCWs has on transmission is to look at its effect on R. Of particular interest is the question of whether R<1 can be achieved. Towards this end we present in Figure 6.1 contours where R=1, for scenarios in which HCWs are given different levels of protection. These contours are some of those shown in Figure 5.2. We see that, for ƒ < 0.5, the intervention consisting of isolating diagnosed cases, as well as PPE and prophylactic use of antiviral drugs for HCWs provides the opportunity for preventing a major epidemic for a substantial range of R0 values.

contours where R=1, for  scenarios in which HCWs are given different levels of protection.

Figure 6.1    Contours of R=1 corresponding to different levels of protection for HCWs, over a range of values for R0 and ƒ. For points above a curves R>1, so there is a possibility of a major outbreak. For points below a curve R<1, so the infection will fade out.

To explain why protecting HCWs provides a greater opportunity for elimination, we present in Figure 6.2 the graphs of IH, the proportion of all infectives over the first 4 generations that can be attributed to direct infection from a health care worker, for a range of R0 values and some alternative interventions. The interventions considered are isolating cases following diagnosis, with ƒ =0.8 achieved, and

(i) no additional intervention,

(ii) PPE for HCWs, and

(iii) default AV and PPE use for HCWs.

We see that when no protection is given to HCWs a substantial number of the infections are attributable to HCWs (top two curves in Figure 6.2). Specifically, after 4 generations without antiviral distribution or PPE to HCWs, and with R0 in the range 1 to 4, about 30-40% of all infections arise from contacts with health care workers when ƒ =0.6. This is a very large proportion of the infections, bearing in mind that HCWs are only 1% of the population. The infection rate by HCWs is substantially reduced by the use of PPE for HCWs, and dissipates almost completely when HCWs are also given AVs for prophylaxis.

the proportion of all infectives over the first 4 generations that can  be attributed to direct infection from a health care worker

Figure 6.2  Graphs of IH, the proportion of all infectives over the first 4 generations that can be attributed to direct infection from a health care worker, for a range of R0 values.
Curves for ƒ =0.8 are shown by heavy dashed lines, and curves for ƒ =0.6 are shown by dotted lines
The outbreak begins with one initial case from the general population.
(Parameters: p =0.005, ei =0.275, es =0.25, ep =0.23.)

While the model used to create Figure 6.2 does not contain a household structure, such a high percentage of infections due to unprotected HCWs suggests that members of their households are at high risk of infection. Whether this remains true when HCWs are protected by PPE and AVs needs further investigation.

Document download

This publication is available as a downloadable document.

Using Mathematical Models to Assess Responses to an Outbreak of an Emerged Viral Respiratory Disease(PDF 873 KB)