A second major impact of an emerged pandemic on the health care system will be the number of patients requiring hospitalisation. In particular, if the epidemic cannot be controlled, then there will be a period of time near the peak of the epidemic in which the health care system will have to cope with a large number of extra patients in need of hospitalisation.

The model used above can also calculate the number of cases requiring hospitalisation for the various antiviral strategies described above. Estimates of hospitalisation rates for an emerged pandemic vary considerably in the literature – as an example here, we consider the moderate assumption that 2% of cases will require hospitalisation, and that on average patients are hospitalised for 1 week. The hospitalisation rate is higher than that of Meltzer et al. (1999) to allow for the possibility that a future pandemic resembles the 1918 pandemic rather than the 1957 influenza pandemic. Those who become infected while on prophylaxis are assumed to have the same rate of hospitalisation and the same length of stay, although treatment with AVs is assumed to reduce the average stay in hospital from 7 to 5 days. Treatment is not assumed to reduce the rate of hospitalisation in this example.

In Figure 6.4, we graph the number of cases requiring hospitalisation over time for an epidemic that (without intervention) would produce (a) a 50% attack rate (R0 = 1.4) and (b) a 70% attack rate (R0 = 1.7), and with use of the antiviral stockpile as outlined in Section 6.2.

Cases  in need of hospitalisation over time for a baseline attack rate of (a) 50%, and (b) 70%, for the four community AV-use strategies described in the  text.

Figure 6.4    Cases in need of hospitalisation over time for a baseline attack rate of (a) 50%, and (b) 70%, for the four community AV-use strategies described in the text.

In this example we considered 10,000 currently hospitalised cases to be a threshold indicating a high demand on hospital beds, that might have to be met through alternative hospitalisation arrangements. In the case of a 50% attack rate, this threshold is crossed if AVs are not used for community care, but that with 50% of cases treated, the number of hospitalised cases peaks below 10,000 (see Table 6.1), while prophylaxis of 40% of case contacts halts the epidemic entirely, so that hospitalisations are comparatively negligible.

In the case of a 70% attack rate, none of these intervention strategies can prevent the number of hospitalisations from rising above 10,000. However, the peak number is reduced in the treatment only strategy, while prophylaxis can delay the time at which this threshold is crossed by several months. Details on the period of time over which hospitalisations exceed 10,000 cases, the peak number of cases requiring hospitalisation and the time of this peak are presented in Table 6.2, along with an estimate of the number of patients requiring an alternative to hospitalisation over the course of the epidemic. The latter figure is calculated by assuming that 10,000 is the maximum number of currently hospitalised influenza patients that the health care system can accommodate, and then counting any excess patients over this number.

Hosp. Length Attack Rate Intervention Period of >104 cases (weeks) Peak cases Peak time
(weeks)
Extra patients that need to be accommodated
7 days 50% None 24.8-33.5 19,370 28.9 52,000
50% Treated Never 8,123 39.6 0
40% given Prophylaxis Never Negligible 8.3 0
70% None 13.2-21.5 46,570 17 162,000
50% Treated 17.1-26.2 29,520 57.2 126,000
40% given Prophylaxis 31.8-40 43,930 35.5 162,000
50% Treated & 40% given Prophylaxis 53.4-61.8 42,570 57.2 152,000

Table 6.2  Period in which more than 10,000 cases are in need of hospitalisation, timing of the peak number of such cases, the peak number of cases and an estimation of the number of additional patients over this period of time that will be in need of hospital treatment.


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)