Evaluating a multiplier model approach to burden estimation by estimating COVID-19 disease burden in Maryland, USA from April 2020-March 2021 and discussing international feasibility Público
Rohraff, Dallas (Summer 2021)
Published
Abstract
Background: COVID-19 disease burden estimation is valuable to understand the true impact of the disease considering that reported COVID-19 case counts are likely to be an underreport of the true number of infections. There remains a need for a straightforward approach to COVID-19 disease burden estimation which can be utilized in a variety of settings.
Methods: We developed a multiplier model approach to estimate COVID-19 burden, which can be modified for use in local or international settings. Using data from the state of Maryland as an example, we evaluated the use of the COVID-19 disease burden multiplier model to estimate COVID-19 associated symptomatic cases, medically attended illnesses, hospitalizations, and deaths from April 2020-March 2021. As a comparison, we estimated excess deaths due to COVID-19 during this time frame using age-specific time-series regression models.
Results: The multiplier approach estimated a total of 615,495 symptomatic illnesses, 234,853 medically attended illnesses, 33,567 hospitalizations, and 9,662 deaths across all age groups in the state of Maryland from April 2020 to March 2021. Those aged <50 years contributed to a majority of the estimated symptomatic and medically attended illnesses, but most hospitalizations and deaths estimated were among those aged ≥50 years. The regression model estimated 8,173 COVID-19 attributable deaths in the same time frame in Maryland.
Discussion: The multiplier model estimated COVID-19 burden in the state of Maryland reasonably well with estimates that were greater than reported case counts and deaths, but less than CDC seroprevalence estimates. This method may prove valuable in local areas if a straightforward approach is desired and the available data sources are well understood. The multiplier model seems feasible to use in Albania and South Africa, but further studies will be needed to evaluate its efficacy in international settings.
Table of Contents
INTRODUCTION 1
METHODS 3
Data Sources 3
Analytic Methods 6
Multiplier Approach to COVID-19 Burden Estimation 6
Dynamic Susceptible Population 6
Symptomatic Illnesses 7
Medically Attended Illnesses 8
Hospitalizations 9
Deaths 10
COVID-19 Time-series Excess Mortality Regression Model 11
RESULTS 12
COVID-19 Burden Estimates 12
COVID-19 Time-series Excess Mortality Regression Model 13
DISCUSSION 14
International Feasibility 19
Albania 21
South Africa 23
Study Summary and Impact 24
NOTES 26
TABLES 27
FIGURES 29
SUPPLEMENTAL FIGURES 32
REFERENCES 35
About this Master's Thesis
School | |
---|---|
Department | |
Subfield / Discipline | |
Degree | |
Submission | |
Language |
|
Research Field | |
Palabra Clave | |
Committee Chair / Thesis Advisor | |
Committee Members |
Primary PDF
Thumbnail | Title | Date Uploaded | Actions |
---|---|---|---|
|
Evaluating a multiplier model approach to burden estimation by estimating COVID-19 disease burden in Maryland, USA from April 2020-March 2021 and discussing international feasibility () | 2021-08-05 13:50:01 -0400 |
|
Supplemental Files
Thumbnail | Title | Date Uploaded | Actions |
---|