We have seen that isolating cases soon after diagnosis is not a particularly effective way to reduce transmission. Quarantining all members of a household with a diagnosed case presents a way to remove both the diagnosed case and (more importantly) some newly-infected individuals from circulation before they have become symptomatic, or even infectious.
The most favourable results are achieved when the diagnosis is made as early as possible and the quarantine imposed immediately following diagnosis. We examine situations where the household is diagnosed and quarantined either 3 or 4 days after infection of the primary case. It seems unlikely that quarantine could be imposed less than 3 days after the primary infection, as the primary case will typically not become symptomatic until 48 hours after infection.
For our calculations with household quarantine we assume that household members complying with quarantine have no infectious contacts with other community members after the quarantine starts. There is likely to be a problem with compliance: household quarantine intrinsically involves asking people to stay at home with (presumably) infectious family members, thus placing themselves and their children at relatively higher risk than they would experience in the community. For this reason, we examine the results for different levels of compliance.
The model used
Transmission of infection is divided into intra-household and inter-household components. We estimate the force of infection acting within each household according to the number of infectives inside the household, and the total number of infectives in the wider community.
The population is divided into households of sizes between 1 and 6, in accordance with data from the Australian 2001 Census (Table 2.2).
The force of infection acting in the wider community is computed by counting the total number of infectives in the community. Infectives are divided into two categories depending on whether or not they come from a house that could be quarantined (i.e. whether someone in their household has been infected for long enough to be diagnosed). Infectives who come from a household that is potentially quarantined contribute a diminished force of infection, which adjusts for compliance. The force of infection is reduced to 0.8 × (original value) if households are 20% compliant with quarantine. Infectives from households that could not be quarantined are allocated their full force of infection until someone in their household has been diagnosed.
We present scenarios in which there is 0%, 20%, 50% and 90% compliance with household quarantine, for reproduction numbers of 1.5 and 2.5.
When the reproduction number is R = 1.5, household quarantine implemented three days after the primary case is infected is an effective intervention. Figure 4.11(a) demonstrates how even a low level (20%) of compliance brings about a substantial reduction in epidemic peak height, as well as delaying it in time. A level of 90% compliance is able to suppress the epidemic entirely. However, if quarantine is delayed until the fourth day, see Fig 4.11(b), its benefit is significantly reduced.
If the reproduction number is R = 2.5, quarantine on day 3 after infection conveys some benefit if the compliance is very high (90%), but is almost useless if it is imposed on day 4 after infection or if compliance is low, see Figure 4.11(c) and Figure 4.11(d).
In short, household quarantine by itself is a moderately effective intervention. It is more effective the earlier it is imposed, and therefore depends on early diagnosis of patients.
In Section 4.9 we demonstrate how combining household quarantine and prophylaxis of household members is far more effective. It is natural that these two interventions go together, because a combined approach is socially more acceptable, since it provides prophylactic protection to household members who stay at home with patients.
Figure 4. 11 Epidemic curves for different levels of compliance with household quarantine. ddiag is the day on which the primary case in each household is diagnosed, and subsequently quarantine is implemented starting at the end of that day.
(a) R = 1.5, quarantine implemented on day 3
(b) R = 1.5, quarantine implemented on day 4
(c) R = 2.5, quarantine implemented on day 3
(d) R = 2.5, quarantine implemented on day 4