Jennifer Priem
Research Subjects
Jennifer Priem has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Andrew McWilliams
Center for Outcomes Research and Evaluation, Atrium Health, Charlotte, NC, United States
4 shared publications - Shih-Hsiung Chou
Atrium Health
4 shared publications - Yhenneko J. Taylor
Center for Outcomes Research and Evaluation, Atrium Health, Charlotte, NC
4 shared publications - Marc A. Kowalkowski
Center for Health System Sciences, Atrium Health Charlotte North Carolina USA
3 shared publications - Pooja Palmer
Center for Outcomes Research and Evaluation, Atrium Health
3 shared publications - Catherine Passaretti
Wake Forest University School of Medicine
2 shared publications
Latest Publications
- Open Access
- 27 November 2023
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 10 (Supplement)
980. Risk Factors for Central Line Associated Bloodstream Infection Prior to and During the COVID-19 Pandemic
Central line associated bloodstream infections (CLABSIs) increased during the COVID-19 pandemic1-3, especially in patients infected with SARS-CoV-2,4 but there has been limited analysis on other risk factors associated with this increase. We conducted a retrospective cohort study to assess risk factors for developing a CLABSI prior to (1/2019-3/2020) and during the COVID-19 pandemic (3/2020-2021) across 12 acute care hospitals in the Southeastern U.S. A multivariable logistic regression analysis was performed comparing patients with and without CLABSI. We detected a shift in risk factors for CLABSI during the pandemic. While COVID-19 diagnosis increased the risk of CLABSI, as has been shown in other studies4, our data additionally revealed a shift in risk factors in patients without COVID-19 showing a higher risk in those with black race, Medicare for insurance and presence of a dialysis catheter during the pandemic. These results suggest that alterations may have occurred in healthcare delivery, specifically in vulnerable populations, laying the foundation for future investigations into potential care improvements. All Authors: No reported disclosures
- 27 November 2023
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 10 (Supplement)
Open Access980. Risk Factors for Central Line Associated Bloodstream Infection Prior to and During the COVID-19 Pandemic
Central line associated bloodstream infections (CLABSIs) increased during the COVID-19 pandemic1-3, especially in patients infected with SARS-CoV-2,4 but there has been limited analysis on other risk factors associated with this increase. We conducted a retrospective cohort study to assess risk factors for developing a CLABSI prior to (1/2019-3/2020) and during the COVID-19 pandemic (3/2020-2021) across 12 acute care hospitals in the Southeastern U.S. A multivariable logistic regression analysis was performed comparing patients with and without CLABSI. We detected a shift in risk factors for CLABSI during the pandemic. While COVID-19 diagnosis increased the risk of CLABSI, as has been shown in other studies4, our data additionally revealed a shift in risk factors in patients without COVID-19 showing a higher risk in those with black race, Medicare for insurance and presence of a dialysis catheter during the pandemic. These results suggest that alterations may have occurred in healthcare delivery, specifically in vulnerable populations, laying the foundation for future investigations into potential care improvements. All Authors: No reported disclosures
- Open Access
- 12 December 2022
- Published by JMIR Publications Inc. in JMIR Preprints
A Matched, Retrospective Comparative Effectiveness Study of Home-based versus Inpatient Hospital Care for Acutely Ill COVID-19 Patients (Preprint)
BACKGROUND Hospital at Home (HaH) programs have been shown to be an effective care delivery model in diverse settings. Comparative effectiveness evaluations between HaH and traditional inpatient care for COVID-19 patients are limited. OBJECTIVE To test the effectiveness of care provided through Atrium Health-HaH compared to inpatient hospitalization for acutely ill patients with COVID-19. METHODS Retrospective matched sample comparative effectiveness study of 116 patients who presented to an emergency department and tested positive for COVID-19 at Atrium Health, a large integrated health care organization in the southeastern United States. Patients who were subsequently hospitalized between May 8 and June 25, 2020 (n = 58) were included in the control group, patients admitted to AH-HaH between July 15 and September 31, 2020 (n = 58) were included in the treatment condition. The primary outcome was 30-day composite of severe illness and death, indexed by the WHO illness severity scale. Patient demographic characteristics, comorbidities and hospital outcomes were collected from electronic health records. RESULTS 30-days post discharge, the number of patients who experienced severe illness or death was equivalent (n = 11/group; strata adjusted OR 1.0, 95% CI 0.37-2.69). CONCLUSIONS AH-HaH provides a feasible and efficacious alternative to traditional inpatient treatment for acutely ill patients with COVID-19.
- 12 December 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessA Matched, Retrospective Comparative Effectiveness Study of Home-based versus Inpatient Hospital Care for Acutely Ill COVID-19 Patients (Preprint)
BACKGROUND Hospital at Home (HaH) programs have been shown to be an effective care delivery model in diverse settings. Comparative effectiveness evaluations between HaH and traditional inpatient care for COVID-19 patients are limited. OBJECTIVE To test the effectiveness of care provided through Atrium Health-HaH compared to inpatient hospitalization for acutely ill patients with COVID-19. METHODS Retrospective matched sample comparative effectiveness study of 116 patients who presented to an emergency department and tested positive for COVID-19 at Atrium Health, a large integrated health care organization in the southeastern United States. Patients who were subsequently hospitalized between May 8 and June 25, 2020 (n = 58) were included in the control group, patients admitted to AH-HaH between July 15 and September 31, 2020 (n = 58) were included in the treatment condition. The primary outcome was 30-day composite of severe illness and death, indexed by the WHO illness severity scale. Patient demographic characteristics, comorbidities and hospital outcomes were collected from electronic health records. RESULTS 30-days post discharge, the number of patients who experienced severe illness or death was equivalent (n = 11/group; strata adjusted OR 1.0, 95% CI 0.37-2.69). CONCLUSIONS AH-HaH provides a feasible and efficacious alternative to traditional inpatient treatment for acutely ill patients with COVID-19.
- Open Access
- 28 November 2022
- Published by International Association for Statistical Computing in Journal of Data Science, Statistics, and Visualisation
- Vol. 2 (7) , 1-13
Implementation of an Adaptable COVID-19 Utilization and Resource Visualization Engine (CURVE) to Depict In-Hospital Resource Forecasts Over Time
We developed an interactive web-based, decision support application that can adapt to the rapid pace of change in region-specific pandemic related variables and knowledge, thereby providing timely, accurate insights to inform a large healthcare system’s proactive response to COVID-19 hospital resource planning. We designed the COVID-19 Utilization and Resource Visualization Engine (CURVE) app to be adaptable to real-time changes as the pandemic evolved, enabling decisions to be supported by contemporary local data and accurate predictive models. To demonstrate this flexibility, we sequentially implemented a Susceptible-Infected-Removed (SIR) model that incorporates social-distancing and imperfect detection (SIR-D2), an extended-state-space Bayesian SIR model (eSIR), and a time-series model (ARIMA). CURVE improves upon other pandemic forecasting solutions by providing adaptable decision support that generates locally calibrated forecasts aligned to health system specific data to guide COVID-19 pandemic planning. The app additionally enables systematic monitoring of forecast model performance and realignment that keeps pace with the pandemic’s volatile spread and behavior. CURVE provides a flexible pandemic decision support framework that places the most accurate, locally relevant information in front of decision makers to enable health systems to be proactive and prepared.
- 28 November 2022
- Published by International Association for Statistical Computing in Journal of Data Science, Statistics, and Visualisation
- Vol. 2 (7) , 1-13
Open AccessImplementation of an Adaptable COVID-19 Utilization and Resource Visualization Engine (CURVE) to Depict In-Hospital Resource Forecasts Over Time
We developed an interactive web-based, decision support application that can adapt to the rapid pace of change in region-specific pandemic related variables and knowledge, thereby providing timely, accurate insights to inform a large healthcare system’s proactive response to COVID-19 hospital resource planning. We designed the COVID-19 Utilization and Resource Visualization Engine (CURVE) app to be adaptable to real-time changes as the pandemic evolved, enabling decisions to be supported by contemporary local data and accurate predictive models. To demonstrate this flexibility, we sequentially implemented a Susceptible-Infected-Removed (SIR) model that incorporates social-distancing and imperfect detection (SIR-D2), an extended-state-space Bayesian SIR model (eSIR), and a time-series model (ARIMA). CURVE improves upon other pandemic forecasting solutions by providing adaptable decision support that generates locally calibrated forecasts aligned to health system specific data to guide COVID-19 pandemic planning. The app additionally enables systematic monitoring of forecast model performance and realignment that keeps pace with the pandemic’s volatile spread and behavior. CURVE provides a flexible pandemic decision support framework that places the most accurate, locally relevant information in front of decision makers to enable health systems to be proactive and prepared.
- 7 March 2022
- Published by Springer Nature in Journal of Racial and Ethnic Health Disparities
- Vol. 10 (2) , 817-825
Evaluating Racial/Ethnic Differences in Care Escalation Among COVID-19 Patients in a Home-Based Hospital
The novel coronavirus disease 2019 (COVID-19) has infected over 414 million people worldwide with 5.8 million deaths, as of February 2022. Telemedicine-based interventions to expand healthcare systems’ capacity and reduce infection risk have rapidly increased during the pandemic, despite concerns regarding equitable access. Atrium Health Hospital at Home (AH-HaH) is a home-based program that provides advanced, hospital-level medical care and monitoring for patients who would otherwise be hospitalized in a traditional setting. Our retrospective cohort study of positive COVID-19 patients who were admitted to AH-HaH aims to investigate whether the rate of care escalation from AH-HaH to traditional hospitalization differed based on patients’ racial/ethnic backgrounds. Logistic regression was used to examine the association between care escalation within 14 days from index AH-HaH admission and race/ethnicity. We found approximately one in five patients receiving care for COVID-19 in AH-HaH required care escalation within 14 days. Odds of care escalation were not significantly different for Hispanic or non-Hispanic Blacks compared to non-Hispanic Whites. However, secondary analyses showed that both Hispanic and non-Hispanic Black patients were younger and with fewer comorbidities than non-Hispanic Whites. The study highlights the need for new care models to vigilantly monitor for disparities, so that timely and tailored adaptations can be implemented for vulnerable populations.
- 7 March 2022
- Published by Springer Nature in Journal of Racial and Ethnic Health Disparities
- Vol. 10 (2) , 817-825
Evaluating Racial/Ethnic Differences in Care Escalation Among COVID-19 Patients in a Home-Based Hospital
The novel coronavirus disease 2019 (COVID-19) has infected over 414 million people worldwide with 5.8 million deaths, as of February 2022. Telemedicine-based interventions to expand healthcare systems’ capacity and reduce infection risk have rapidly increased during the pandemic, despite concerns regarding equitable access. Atrium Health Hospital at Home (AH-HaH) is a home-based program that provides advanced, hospital-level medical care and monitoring for patients who would otherwise be hospitalized in a traditional setting. Our retrospective cohort study of positive COVID-19 patients who were admitted to AH-HaH aims to investigate whether the rate of care escalation from AH-HaH to traditional hospitalization differed based on patients’ racial/ethnic backgrounds. Logistic regression was used to examine the association between care escalation within 14 days from index AH-HaH admission and race/ethnicity. We found approximately one in five patients receiving care for COVID-19 in AH-HaH required care escalation within 14 days. Odds of care escalation were not significantly different for Hispanic or non-Hispanic Blacks compared to non-Hispanic Whites. However, secondary analyses showed that both Hispanic and non-Hispanic Black patients were younger and with fewer comorbidities than non-Hispanic Whites. The study highlights the need for new care models to vigilantly monitor for disparities, so that timely and tailored adaptations can be implemented for vulnerable populations.
- Open Access
- 29 July 2021
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 1 (S1) , s19-s20
Evaluating the Relationship between Cycle Threshold Values and Reported COVID-19 Symptoms among Healthcare Workers
Background: SARS-CoV-2 detected by reverse transcription polymerase chain reaction (RT-PCR) can persist for weeks to months in some individuals. Cycle threshold (Ct) values represent the number of cycles needed to amplify viral ribonucleic acid (RNA) to reach a detectable level. As such, Ct values are inversely related to the amount of virus in a sample. As knowledge of SARS-CoV-2 viral dynamics continues to evolve, understanding the relationship between Ct values, type of symptoms, and timing of symptom onset can help determine when infected individuals are most likely to be infectious. Methods: We conducted a retrospective cohort study of 1,027 healthcare workers (HCWs) who tested positive for SARS-CoV-2 by RT-PCR from nasopharyngeal specimens between June 27, 2020, and September 21, 2020. All HCWs were interviewed within 72 hours of their diagnosis for symptom history. Due to multiple PCR platforms being in use in our facility, only 360 HCWs (35%) had Ct values available for analysis. Multivariate linear regression models examined the effect of COVID-19–related symptoms and timing of symptom onset to test on Ct values. Results: The most frequently reported symptoms were congestion (55.6%), cough (50.3%), and headache (46.7%). Other symptoms less commonly reported were fatigue (36.7%), loss of taste or smell (36.4%), fever (35.4%), muscle aches (33.3%), sore throat (27.4%), and diarrhea (26.7%). Symptomatic HCWs (88.3% of sample) had lower Ct values (ORF-1 M = 22.66, SD = 5.17; E-Gene M = 24.34, SD = 6.60) than asymptomatic individuals (ORF-1 M = 25.46, SD = 6.06; E-Gene M = 29.34, SD = 7.96). Of all symptoms measured, only presence of fever, congestion, and muscle aches predicted significantly lower Ct values. Mean Ct values decreased 2 days prior to symptom onset, were lowest the day of symptom onset, then increased in a curvilinear fashion. There were no significant 2-way interactions between symptoms and time of symptom onset to testing. Conclusions: The curvilinear pattern of Ct values over time from symptom onset are consistent with disease progression patterns and support current understanding of infectivity being highest 2 days prior to symptom onset through day 8. Presence of fever, congestion, and muscle aches are significantly correlated with lower Ct values, suggesting that these symptoms are associated with higher viral load. Although Ct values are not without limitations, our findings support the current understanding that presymptomatic and symptomatic individuals, particularly those with fever, congestion, and muscle aches, may pose higher risk of transmission to others. Funding: No Disclosures: None Figure 1. Figure 2. Table 1.
- 29 July 2021
- Published by Cambridge University Press (CUP) in Antimicrobial Stewardship & Healthcare Epidemiology
- Vol. 1 (S1) , s19-s20
Open AccessEvaluating the Relationship between Cycle Threshold Values and Reported COVID-19 Symptoms among Healthcare Workers
Background: SARS-CoV-2 detected by reverse transcription polymerase chain reaction (RT-PCR) can persist for weeks to months in some individuals. Cycle threshold (Ct) values represent the number of cycles needed to amplify viral ribonucleic acid (RNA) to reach a detectable level. As such, Ct values are inversely related to the amount of virus in a sample. As knowledge of SARS-CoV-2 viral dynamics continues to evolve, understanding the relationship between Ct values, type of symptoms, and timing of symptom onset can help determine when infected individuals are most likely to be infectious. Methods: We conducted a retrospective cohort study of 1,027 healthcare workers (HCWs) who tested positive for SARS-CoV-2 by RT-PCR from nasopharyngeal specimens between June 27, 2020, and September 21, 2020. All HCWs were interviewed within 72 hours of their diagnosis for symptom history. Due to multiple PCR platforms being in use in our facility, only 360 HCWs (35%) had Ct values available for analysis. Multivariate linear regression models examined the effect of COVID-19–related symptoms and timing of symptom onset to test on Ct values. Results: The most frequently reported symptoms were congestion (55.6%), cough (50.3%), and headache (46.7%). Other symptoms less commonly reported were fatigue (36.7%), loss of taste or smell (36.4%), fever (35.4%), muscle aches (33.3%), sore throat (27.4%), and diarrhea (26.7%). Symptomatic HCWs (88.3% of sample) had lower Ct values (ORF-1 M = 22.66, SD = 5.17; E-Gene M = 24.34, SD = 6.60) than asymptomatic individuals (ORF-1 M = 25.46, SD = 6.06; E-Gene M = 29.34, SD = 7.96). Of all symptoms measured, only presence of fever, congestion, and muscle aches predicted significantly lower Ct values. Mean Ct values decreased 2 days prior to symptom onset, were lowest the day of symptom onset, then increased in a curvilinear fashion. There were no significant 2-way interactions between symptoms and time of symptom onset to testing. Conclusions: The curvilinear pattern of Ct values over time from symptom onset are consistent with disease progression patterns and support current understanding of infectivity being highest 2 days prior to symptom onset through day 8. Presence of fever, congestion, and muscle aches are significantly correlated with lower Ct values, suggesting that these symptoms are associated with higher viral load. Although Ct values are not without limitations, our findings support the current understanding that presymptomatic and symptomatic individuals, particularly those with fever, congestion, and muscle aches, may pose higher risk of transmission to others. Funding: No Disclosures: None Figure 1. Figure 2. Table 1.
- Open Access
- 19 June 2020
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 6 (2) , e19353
Modeling COVID-19 Latent Prevalence to Assess a Public Health Intervention at a State and Regional Scale: Retrospective Cohort Study
Background: Emergence of the coronavirus disease (COVID-19) caught the world off guard and unprepared, initiating a global pandemic. In the absence of evidence, individual communities had to take timely action to reduce the rate of disease spread and avoid overburdening their health care systems. Although a few predictive models have been published to guide these decisions, most have not taken into account spatial differences and have included assumptions that do not match the local realities. Access to reliable information that is adapted to local context is critical for policy makers to make informed decisions during a rapidly evolving pandemic. Objective: The goal of this study was to develop an adapted susceptible-infected-removed (SIR) model to predict the trajectory of the COVID-19 pandemic in North Carolina and the Charlotte Metropolitan Region, and to incorporate the effect of a public health intervention to reduce disease spread while accounting for unique regional features and imperfect detection. Methods: Three SIR models were fit to infection prevalence data from North Carolina and the greater Charlotte Region and then rigorously compared. One of these models (SIR-int) accounted for a stay-at-home intervention and imperfect detection of COVID-19 cases. We computed longitudinal total estimates of the susceptible, infected, and removed compartments of both populations, along with other pandemic characteristics such as the basic reproduction number. Results: Prior to March 26, disease spread was rapid at the pandemic onset with the Charlotte Region doubling time of 2.56 days (95% CI 2.11-3.25) and in North Carolina 2.94 days (95% CI 2.33-4.00). Subsequently, disease spread significantly slowed with doubling times increased in the Charlotte Region to 4.70 days (95% CI 3.77-6.22) and in North Carolina to 4.01 days (95% CI 3.43-4.83). Reflecting spatial differences, this deceleration favored the greater Charlotte Region compared to North Carolina as a whole. A comparison of the efficacy of intervention, defined as 1 – the hazard ratio of infection, gave 0.25 for North Carolina and 0.43 for the Charlotte Region. In addition, early in the pandemic, the initial basic SIR model had good fit to the data; however, as the pandemic and local conditions evolved, the SIR-int model emerged as the model with better fit. Conclusions: Using local data and continuous attention to model adaptation, our findings have enabled policy makers, public health officials, and health systems to proactively plan capacity and evaluate the impact of a public health intervention. Our SIR-int model for estimated latent prevalence was reasonably flexible, highly accurate, and demonstrated efficacy of a stay-at-home order at both the state and regional level. Our results highlight the importance of incorporating local context into pandemic forecast modeling, as well as the need to remain vigilant and informed by the data as we enter into a critical period of the outbreak.
- 19 June 2020
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 6 (2) , e19353
Open AccessModeling COVID-19 Latent Prevalence to Assess a Public Health Intervention at a State and Regional Scale: Retrospective Cohort Study
Background: Emergence of the coronavirus disease (COVID-19) caught the world off guard and unprepared, initiating a global pandemic. In the absence of evidence, individual communities had to take timely action to reduce the rate of disease spread and avoid overburdening their health care systems. Although a few predictive models have been published to guide these decisions, most have not taken into account spatial differences and have included assumptions that do not match the local realities. Access to reliable information that is adapted to local context is critical for policy makers to make informed decisions during a rapidly evolving pandemic. Objective: The goal of this study was to develop an adapted susceptible-infected-removed (SIR) model to predict the trajectory of the COVID-19 pandemic in North Carolina and the Charlotte Metropolitan Region, and to incorporate the effect of a public health intervention to reduce disease spread while accounting for unique regional features and imperfect detection. Methods: Three SIR models were fit to infection prevalence data from North Carolina and the greater Charlotte Region and then rigorously compared. One of these models (SIR-int) accounted for a stay-at-home intervention and imperfect detection of COVID-19 cases. We computed longitudinal total estimates of the susceptible, infected, and removed compartments of both populations, along with other pandemic characteristics such as the basic reproduction number. Results: Prior to March 26, disease spread was rapid at the pandemic onset with the Charlotte Region doubling time of 2.56 days (95% CI 2.11-3.25) and in North Carolina 2.94 days (95% CI 2.33-4.00). Subsequently, disease spread significantly slowed with doubling times increased in the Charlotte Region to 4.70 days (95% CI 3.77-6.22) and in North Carolina to 4.01 days (95% CI 3.43-4.83). Reflecting spatial differences, this deceleration favored the greater Charlotte Region compared to North Carolina as a whole. A comparison of the efficacy of intervention, defined as 1 – the hazard ratio of infection, gave 0.25 for North Carolina and 0.43 for the Charlotte Region. In addition, early in the pandemic, the initial basic SIR model had good fit to the data; however, as the pandemic and local conditions evolved, the SIR-int model emerged as the model with better fit. Conclusions: Using local data and continuous attention to model adaptation, our findings have enabled policy makers, public health officials, and health systems to proactively plan capacity and evaluate the impact of a public health intervention. Our SIR-int model for estimated latent prevalence was reasonably flexible, highly accurate, and demonstrated efficacy of a stay-at-home order at both the state and regional level. Our results highlight the importance of incorporating local context into pandemic forecast modeling, as well as the need to remain vigilant and informed by the data as we enter into a critical period of the outbreak.
- Open Access
- 14 April 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Modeling COVID-19 Latent Prevalence to Assess a Public Health Intervention at a State and Regional Scale: Retrospective Cohort Study (Preprint)
BACKGROUND Emergence of the coronavirus disease (COVID-19) caught the world off guard and unprepared, initiating a global pandemic. In the absence of evidence, individual communities had to take timely action to reduce the rate of disease spread and avoid overburdening their health care systems. Although a few predictive models have been published to guide these decisions, most have not taken into account spatial differences and have included assumptions that do not match the local realities. Access to reliable information that is adapted to local context is critical for policy makers to make informed decisions during a rapidly evolving pandemic. OBJECTIVE The goal of this study was to develop an adapted susceptible-infected-removed (SIR) model to predict the trajectory of the COVID-19 pandemic in North Carolina and the Charlotte Metropolitan Region, and to incorporate the effect of a public health intervention to reduce disease spread while accounting for unique regional features and imperfect detection. METHODS Three SIR models were fit to infection prevalence data from North Carolina and the greater Charlotte Region and then rigorously compared. One of these models (SIR-int) accounted for a stay-at-home intervention and imperfect detection of COVID-19 cases. We computed longitudinal total estimates of the susceptible, infected, and removed compartments of both populations, along with other pandemic characteristics such as the basic reproduction number. RESULTS Prior to March 26, disease spread was rapid at the pandemic onset with the Charlotte Region doubling time of 2.56 days (95% CI 2.11-3.25) and in North Carolina 2.94 days (95% CI 2.33-4.00). Subsequently, disease spread significantly slowed with doubling times increased in the Charlotte Region to 4.70 days (95% CI 3.77-6.22) and in North Carolina to 4.01 days (95% CI 3.43-4.83). Reflecting spatial differences, this deceleration favored the greater Charlotte Region compared to North Carolina as a whole. A comparison of the efficacy of intervention, defined as 1 – the hazard ratio of infection, gave 0.25 for North Carolina and 0.43 for the Charlotte Region. In addition, early in the pandemic, the initial basic SIR model had good fit to the data; however, as the pandemic and local conditions evolved, the SIR-int model emerged as the model with better fit. CONCLUSIONS Using local data and continuous attention to model adaptation, our findings have enabled policy makers, public health officials, and health systems to proactively plan capacity and evaluate the impact of a public health intervention. Our SIR-int model for estimated latent prevalence was reasonably flexible, highly accurate, and demonstrated efficacy of a stay-at-home order at both the state and regional level. Our results highlight the importance of incorporating local context into pandemic forecast modeling, as well as the need to remain vigilant and informed by the data as we enter into a critical period of the outbreak.
- 14 April 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessModeling COVID-19 Latent Prevalence to Assess a Public Health Intervention at a State and Regional Scale: Retrospective Cohort Study (Preprint)
BACKGROUND Emergence of the coronavirus disease (COVID-19) caught the world off guard and unprepared, initiating a global pandemic. In the absence of evidence, individual communities had to take timely action to reduce the rate of disease spread and avoid overburdening their health care systems. Although a few predictive models have been published to guide these decisions, most have not taken into account spatial differences and have included assumptions that do not match the local realities. Access to reliable information that is adapted to local context is critical for policy makers to make informed decisions during a rapidly evolving pandemic. OBJECTIVE The goal of this study was to develop an adapted susceptible-infected-removed (SIR) model to predict the trajectory of the COVID-19 pandemic in North Carolina and the Charlotte Metropolitan Region, and to incorporate the effect of a public health intervention to reduce disease spread while accounting for unique regional features and imperfect detection. METHODS Three SIR models were fit to infection prevalence data from North Carolina and the greater Charlotte Region and then rigorously compared. One of these models (SIR-int) accounted for a stay-at-home intervention and imperfect detection of COVID-19 cases. We computed longitudinal total estimates of the susceptible, infected, and removed compartments of both populations, along with other pandemic characteristics such as the basic reproduction number. RESULTS Prior to March 26, disease spread was rapid at the pandemic onset with the Charlotte Region doubling time of 2.56 days (95% CI 2.11-3.25) and in North Carolina 2.94 days (95% CI 2.33-4.00). Subsequently, disease spread significantly slowed with doubling times increased in the Charlotte Region to 4.70 days (95% CI 3.77-6.22) and in North Carolina to 4.01 days (95% CI 3.43-4.83). Reflecting spatial differences, this deceleration favored the greater Charlotte Region compared to North Carolina as a whole. A comparison of the efficacy of intervention, defined as 1 – the hazard ratio of infection, gave 0.25 for North Carolina and 0.43 for the Charlotte Region. In addition, early in the pandemic, the initial basic SIR model had good fit to the data; however, as the pandemic and local conditions evolved, the SIR-int model emerged as the model with better fit. CONCLUSIONS Using local data and continuous attention to model adaptation, our findings have enabled policy makers, public health officials, and health systems to proactively plan capacity and evaluate the impact of a public health intervention. Our SIR-int model for estimated latent prevalence was reasonably flexible, highly accurate, and demonstrated efficacy of a stay-at-home order at both the state and regional level. Our results highlight the importance of incorporating local context into pandemic forecast modeling, as well as the need to remain vigilant and informed by the data as we enter into a critical period of the outbreak.