Comparison of infection control strategies to reduce COVID-19 outbreaks in homeless shelters in the United States: a simulation study

Table 1 Microsimulation input parameters based on observed outbreak data from homeless shelters in Seattle, Boston, and San Francisco

Parameter*	Base case value	Range in sensitivity analysis^†	References
Natural history
Mean duration of latent infection period, days	3 days	–	[17]
Mean duration of early infectious stage (subclinical/clinical), days	2.3 days	–	[17]
Mean duration of late infectious stage (subclinical/clinical), days	8 days	–	[17, 20, 67, 68]
Relative infectiousness of subclinical infection to clinical infection	1	0.5–1	[19, 69, 70]
Relative infectiousness of early infectious stage to late infectious stage	2	1–3	[17, 18]
Probability of developing clinical symptoms	Age-dependent (see Additional file 1: Table S2)	–	[13]
Background infection rate in community outside shelter	Shelter-specific (see Additional file 1)	0–439 infections/1,000,000 person-days	[49,50,51]
Basic reproduction number, R₀	Variable	1.5–6.2	Estimated
Intervention
Symptom screening
Sensitivity	0.4	0.3–0.5	Assumed based on [72]
Specificity	0.9	0.8–0.9	Assumed
Compliance of symptomatic individuals with PCR testing	80%	50–100%	Assumed
PCR testing
Sensitivity	0.75	0.6–0.9	[28,29,30,31]
Specificity	1	0.95–1	[28, 31]
Frequency	Twice weekly	Daily–monthly	[73,74,75]
Compliance	80%	50–100%	Assumed
Masks
Effectiveness at reducing infectious material exhaled	30%	10–50%	[36, 37, 42]^‡
Effectiveness at reducing infectious material inhaled	40%	20–60%	[36, 37, 42]^‡
Compliance	60%	30–100%	[37, 48]

*See Additional file 1: Table S5 for a complete list of all parameters used in the model calibration and intervention simulations
^†In the sensitivity analysis, each intervention strategy was simulated with all combinations of the minimum and maximum values of the ranges for the indicated parameters to generate the uncertainty ranges around the probability of averting an outbreak in Table 2
^‡See Additional file 1 for a review of current literature on mask effectiveness and a full list of references

ISSN: 1741-7015