a downgraded for indirectness because the studies used SARS data, which does not reflect the presymptomatic infectiousness of COVID-19
b Only 2 cases of SARS out of >95,000 quarantined
Abbreviations: COVID-19: coronavirus disease 2019; NA: not applicable; SARS: severe acute respiratory syndrome

3.2.2 Resource use and comparative effectiveness of quarantine

We did not identify any studies assessing resource use of quarantine for travelers from regions with high transmission rates, nor on comparative effectiveness of diverse types of quarantine.

4 Discussion

To the best of our knowledge, this is the first systematic evidence synthesis on the effectiveness of quarantine for COVID-19 using additional data from other severe coronavirus outbreaks. The evidencebase is very limited because 19 of the 21 included studies are on SARS and MERS while only two are on SARS-CoV-2. Nevertheless, included studies consistently report a benefit of quarantine measures. Quarantine of people exposed to confirmed cases resulted in higher numbers of infections and deaths averted than through quarantine of those who repatriated from regions with an outbreak. In 2011, Jefferson et al. conducted a systematic review on physical interventions to combat respiratory diseases in general. They did not find valid evidence on the effectiveness of quarantine and stressed the need for further research.47
The searches identified only two modeling studies on prevention and control measures for the current COVID-19 outbreak.45,46 Both studies concluded that prevention and control measures can be effective in controlling COVID-19 and emphasized the implementation of early and strong measures. These studies, however, are based on limited data sets and did not specify what combinations of prevention and control measures they mean with “weak” and “strong.” Thus, it is not possible to draw conclusions about the effectiveness of specific quarantine strategies.
The indirect evidence is limited to four cohort studies and 15 modeling studies that used data from SARS and MERS outbreaks. For the effectiveness of quarantine for individuals who were in close contact with a SARS or MERS case, studies consistently confirmed the effectiveness of quarantine but focus mostly on quarantine for individuals who were in contact with confirmed cases. A modeling study from Taiwan showed that even with a very low quarantine rate (4·7% of those who should have been quarantined), a substantial number of SARS cases and deaths could be prevented.36 The evidence on the comparative effectiveness of different types of quarantine was inconclusive.
Only three studies addressed the effectiveness of quarantine for people who traveled from a region with a high rate of SARS transmission.28,36,43 The number of infections and deaths averted with quarantine was substantially lower than for the quarantine of people exposed to SARS cases.
Limitations in the body of evidence
The applicability of the results from SARS and MERS studies is likely to be limited because most models assumed that infectiousness starts with symptom onset. Preliminary estimates of the basic reproduction number of SARS-CoV-2 range from 2·8 to 5·5.4,48,49 Models have shown that the effectiveness of quarantining individuals during outbreaks of diseases with presymptomatic infectiousness and a basic reproduction number of >2·5 is limited. Based on estimates of a basic reproduction number of 3·8 (95% CI, 3·6 and 4·0), Read et al. state, in an unpublished article, that 72% to 75% of the transmission must be stopped by control measures for COVID-19 infections to stop increasing.4
Our report is not applicable to community quarantine, which was outside the scope of this review.
Limitations of the review
Because of time constraints, we conducted a rapid review and abbreviated certain methodological steps of the review process. Specifically, we dually screened only 30% of the titles and abstracts; for the rest, single screening was applied. A recent study showed that single abstract screening misses up to 13% of relevant studies.50 In addition, we applied a single risk of bias assessment, data extraction, and certainty of evidence rating, with a second person checking the plausibility and correctness. These steps are prone to human error and with limited quality assurance with a second, independent person conducting all steps, we introduced some risk of error to this rapid review. Nevertheless, we are confident that none of these methodological limitations would change the overall conclusions of this review.
5 Conclusion
Despite the limited evidence on quarantine to control COVID-19, studies consistently concluded that quarantine is an essential public health measure to control the spread of severe coronaviruses. For both effectiveness and resource use, early implementation of quarantine seems to be key. In addition, studies showed that effectiveness of quarantine of individuals depend on the infectious disease biologic dynamics as well as transmissibility, both areas that are still not clarified for COVID-19.