Interventions for SARS-CoV-2 Prevention among Incarcerated Adults: A Network-Based Modeling Analysis Público
Schneider, Isaac (Spring 2023)
Abstract
Background. COVID-19 presents challenges in settings like jails or prisons with a high density of contacts. The state of Georgia has seen limited COVID vaccine uptake with one of the highest incarceration rates in the United States. Using a network-based SARS-CoV-2 transmission model parameterized with data from the Fulton County Jail, this study investigates the impact of three SARS-CoV-2 prevention strategies: vaccination, contact tracing and quarantining, and jail release.
Methods. Contact networks were simulated at two different overlapping network layers: cell and block. Cell-level contacts represented shared confined sleeping space, whereas block-level contacts represented shared common space. Contact tracing and quarantining were simulated at the cell-level or both the cell- and block-levels. A reference scenario and nine intervention scenarios were simulated 300 times to estimate the median and interquartile range (IQR) of the outcome measures. Each scenario simulated a 185-day period to measure the prolonged effects of the interventions in the midst of a potential COVID outbreak in the jail. The cumulative incidence, number of infections averted (NIA), and percentage of infections averted (PIA) were calculated for all scenarios. For the seven scenarios involving contact tracing and quarantining, total quarantines over the simulation and the number of quarantines per day were calculated to determine the quarantine requirements. Additionally, a sensitivity analysis was conducted to compare the interaction between vaccination rates and contact tracing rates.
Results. We found that cell-level contact tracing was a relatively ineffective intervention by itself (3.2% PIA), but its effectiveness increased when combining it with other interventions (i.e., vaccination or increased jail release rate). The other intervention strategies each produced a PIA of over 10%, with the jail release scenario producing a PIA of nearly 20% despite only resulting in a 13% reduction in the jail population. The all-level contact tracing only scenario was effective at both 50% and 100% of contacts traced, but feasibility is limited without a reduction in the jail population.
Conclusions. Implementing a combination intervention approach could substantially reduce the morbidity and mortality from COVID-19 and future respiratory viruses in this jail setting while providing secondary protection to the community.
Table of Contents
INTRODUCTION. 1
METHODS. 2
RESULTS. 5
DISCUSSION. 8
REFERENCES. 13
TABLES AND FIGURES. 15
Table 1. General model parameters. 15
Table 2. Parameters adjusted for different intervention scenarios. 16
Table 3. Median and IQR cumulative incidence and median and IQR number and percentage of infections averted for nine scenarios against a baseline with 300 simulations per scenario. 17
Table 4. Median and IQR total number of quarantined individuals and median and IQR quarantined per day for seven scenarios with 300 simulations per scenario. 18
Figure 1. Disease transmission process displaying the stages of infection, transition between stages, and entry/exit processes. 19
Figure 2. Percentage of cell and block-level infections averted across 300 simulations for each of the nine scenarios – a: vaccination, b: 50% cell-level contact tracing (CT), c: 100% cell-level CT, d: 50% all-level CT, e: 100% all-level CT, f: jail release, g: all combined (cell-level CT), h: all combined (all-level CT, limited quarantine capacity), i: all combined (all-level CT, full quarantine capacity). 20
Figure 3. Cell and block-level cumulative incidence across 300 simulations for each of the nine scenarios and a baseline scenario – a: vaccination, b: 50% cell-level contact tracing (CT), c: 100% cell-level CT, d: 50% all-level CT, e: 100% all-level CT, f: jail release, g: all combined (cell-level CT), h: all combined (all-level CT, limited quarantine capacity), i: all combined (all-level CT, full quarantine capacity), z: baseline. 21
Figure 4. Number quarantined per day across 300 simulations for each of the three cell-level contact tracing scenarios – b: 50% cell-level contact tracing (CT), c: 100% cell-level CT, g: all combined (cell-level CT). 22
Figure 5. Number quarantined per day across 300 simulations for each of the four all-level contact tracing scenarios – d: 50% all-level contact tracing (CT), e: 100% all-level CT, h: all combined (all-level CT, limited quarantine capacity), i: all combined (all-level CT, full quarantine capacity). 23
Figure 6. Sensitivity analysis of percentage of infections averted (PIA) between (A) cell-level contact tracing or (B) all-level contact tracing and varying daily vaccination rates. 24
About this Master's Thesis
| School | |
|---|---|
| Department | |
| Subfield / Discipline | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| Palabra Clave | |
| Committee Chair / Thesis Advisor |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Interventions for SARS-CoV-2 Prevention among Incarcerated Adults: A Network-Based Modeling Analysis () | 2023-04-19 17:36:16 -0400 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|