Mary J. Choi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mary J. Choi has a total of 181 co-authors. Here are the top co-authors by number of shared publications:
- Pierre Etienne Rollin
Viral Special Pathogens Branch, Centers for Disease Control and Prevention, Atlanta, Georgia
4 shared publications Barbara Knust
Viral Special Pathogens Branch, Atlanta, Georgia
3 shared publications- Lawrence J. Purpura
Division of Infectious Diseases, Department of Medicine, Columbia University Irving Medical Center, New York, New York, USA
3 shared publications Ute Ströher
Viral Special Pathogens Branch, Atlanta, Georgia
3 shared publications- John David Klena
Viral Special Pathogens Branch, Centers for Disease Control and Prevention, 1600 Clifton Rd. NE, Atlanta, GA 30329, USA
2 shared publications Miles Stewart
Affiliation Applied Physics Laboratory, Johns Hopkins University, Baltimore, Maryland, United States of America
2 shared publications
Latest Publications
- 4 November 2026
- Published by American Society of Tropical Medicine and Hygiene in The American Journal of Tropical Medicine and Hygiene
- Vol. 115 (5) , 977-984
Investigation of Dinga Dinga: A Cluster of Bilateral Lower-Extremity Tremors—Bundibugyo District, Uganda, 2023–2024
: In September 2023, a cluster of unexplained illness in women affecting their ability to stand/walk was reported at Bundibugyo Hospital, Uganda. Field investigations were conducted to verify reports, conduct active case finding, and identify potential causes using a hypothesis-generating questionnaire, a neurological examination tool, and a standardized laboratory panel. As of January 31, 2025, 136 patients were identified; all were female. Among 108 with age data, the median age was 18 years old, and all but one lived in Bundibugyo District. All reported bilateral lower-extremity tremors interfering with their ability to stand/walk that resolved when sitting or lying down. Of 75 women evaluated, 33 (44%) were acutely symptomatic at assessment. Tremors lasted between 4 days and 3 months. Of 90 women who provided symptom recurrence information, 20 (22%) reported recurrent episodes. No shared environmental exposures of concern were identified among 76 patients with data. Although limited female-specific exposures were noted, no known toxins were identified among patients. Blood specimens were submitted for laboratory testing, and 24 of 93 (26%) tested positive for Plasmodium . One of 34 (3%) tested positive for Shigella spp. Three of 87 (3%) were positive for Ekpoma virus 2 (EKV2), and 41 of 80 (51%) were anti-EKV2 IgG positive. Metagenomic sequencing confirmed two of the EKV2 cases and identified rickettsiae-specific reads in one individual. It is still unknown whether the cause is an infectious agent or an environmental toxin. This outbreak highlights the importance of strong field epidemiology and laboratory capacity in investigating emerging neurological syndromes.
- 4 November 2026
- Published by American Society of Tropical Medicine and Hygiene in The American Journal of Tropical Medicine and Hygiene
- Vol. 115 (5) , 977-984
Investigation of Dinga Dinga: A Cluster of Bilateral Lower-Extremity Tremors—Bundibugyo District, Uganda, 2023–2024
: In September 2023, a cluster of unexplained illness in women affecting their ability to stand/walk was reported at Bundibugyo Hospital, Uganda. Field investigations were conducted to verify reports, conduct active case finding, and identify potential causes using a hypothesis-generating questionnaire, a neurological examination tool, and a standardized laboratory panel. As of January 31, 2025, 136 patients were identified; all were female. Among 108 with age data, the median age was 18 years old, and all but one lived in Bundibugyo District. All reported bilateral lower-extremity tremors interfering with their ability to stand/walk that resolved when sitting or lying down. Of 75 women evaluated, 33 (44%) were acutely symptomatic at assessment. Tremors lasted between 4 days and 3 months. Of 90 women who provided symptom recurrence information, 20 (22%) reported recurrent episodes. No shared environmental exposures of concern were identified among 76 patients with data. Although limited female-specific exposures were noted, no known toxins were identified among patients. Blood specimens were submitted for laboratory testing, and 24 of 93 (26%) tested positive for Plasmodium . One of 34 (3%) tested positive for Shigella spp. Three of 87 (3%) were positive for Ekpoma virus 2 (EKV2), and 41 of 80 (51%) were anti-EKV2 IgG positive. Metagenomic sequencing confirmed two of the EKV2 cases and identified rickettsiae-specific reads in one individual. It is still unknown whether the cause is an infectious agent or an environmental toxin. This outbreak highlights the importance of strong field epidemiology and laboratory capacity in investigating emerging neurological syndromes.
- 24 April 2025
- Published by Oxford University Press (OUP) in Clinical Infectious Diseases
- Vol. 81 (5) , 891-899
Case Series of Patients With Laboratory-Confirmed Marburg Virus Disease in 2023 in Equatorial Guinea
Background: Marburg virus disease (MVD) is a severe viral infection caused by the Marburg marburgvirus species. In February 2023, Equatorial Guinea declared its first outbreak. This case series describes the natural history of MVD in 5 patients with laboratory-confirmed disease. Methods: Patients with confirmed MVD who were admitted to the national treatment center in Bata, Equatorial Guinea were monitored for vital signs and symptoms. Comprehensive clinical data were collected to understand the progression and outcome of the disease. Results: Five patients were confirmed to have MVD. Three male healthcare workers received a diagnosis diagnosed early in their disease and subsequently survived. The other 2 patients, both female, were admitted later in their disease progression and died within 24 hours of admission. Four patients received remdesivir under a protocol for the monitored emergency use of unregistered and experimental interventions. The early symptoms were nonspecific, with rapid progression to more severe conditions in the later stages of the disease. Early treatment with remdesivir showed the drug to be well tolerated. Conclusions: Contrary to some reports and the recommended case definition for MVD, our patients presented with a rash but did not exhibit vomiting or diarrhea. Hemorrhagic signs were solely observed in the terminal stage, preceding demise. Despite the limited sample size, these findings emphasize the importance of tailoring the case definition to the specific outbreak. Further evidence on the efficacy and safety of therapeutics for MVD, including remdesivir, should be gathered through well-designed trials during future epidemic responses.
- 24 April 2025
- Published by Oxford University Press (OUP) in Clinical Infectious Diseases
- Vol. 81 (5) , 891-899
Case Series of Patients With Laboratory-Confirmed Marburg Virus Disease in 2023 in Equatorial Guinea
Background: Marburg virus disease (MVD) is a severe viral infection caused by the Marburg marburgvirus species. In February 2023, Equatorial Guinea declared its first outbreak. This case series describes the natural history of MVD in 5 patients with laboratory-confirmed disease. Methods: Patients with confirmed MVD who were admitted to the national treatment center in Bata, Equatorial Guinea were monitored for vital signs and symptoms. Comprehensive clinical data were collected to understand the progression and outcome of the disease. Results: Five patients were confirmed to have MVD. Three male healthcare workers received a diagnosis diagnosed early in their disease and subsequently survived. The other 2 patients, both female, were admitted later in their disease progression and died within 24 hours of admission. Four patients received remdesivir under a protocol for the monitored emergency use of unregistered and experimental interventions. The early symptoms were nonspecific, with rapid progression to more severe conditions in the later stages of the disease. Early treatment with remdesivir showed the drug to be well tolerated. Conclusions: Contrary to some reports and the recommended case definition for MVD, our patients presented with a rash but did not exhibit vomiting or diarrhea. Hemorrhagic signs were solely observed in the terminal stage, preceding demise. Despite the limited sample size, these findings emphasize the importance of tailoring the case definition to the specific outbreak. Further evidence on the efficacy and safety of therapeutics for MVD, including remdesivir, should be gathered through well-designed trials during future epidemic responses.
- Open Access
- 1 February 2022
- Published by Elsevier in International Journal of Infectious Diseases
- Vol. 115, 126-133
Ebola virus disease nosocomial infections in the Democratic Republic of the Congo: a descriptive study of cases during the 2018–2020 outbreak
- 1 February 2022
- Published by Elsevier in International Journal of Infectious Diseases
- Vol. 115, 126-133
Open AccessEbola virus disease nosocomial infections in the Democratic Republic of the Congo: a descriptive study of cases during the 2018–2020 outbreak
- Open Access
- 5 August 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (8) , e0255631
Development and implementation of the Ebola Exposure Window Calculator: A tool for Ebola virus disease outbreak field investigations
During an Ebola virus disease (EVD) outbreak, calculating the exposure window of a confirmed case can assist field investigators in identifying the source of infection and establishing chains of transmission. However, field investigators often have difficulty calculating this window. We developed a bilingual (English/French), smartphone-based field application to assist field investigators in determining the exposure window of an EVD case. The calculator only requires the reported date of symptoms onset and the type of symptoms present at onset or the date of death. Prior to the release of this application, there was no similar electronic capability to enable consistent calculation of EVD exposure windows for field investigators. The Democratic Republic of the Congo Ministry of Health endorsed the application and incorporated it into trainings for field staff. Available for Apple and Android devices, the calculator continues to be downloaded even as the eastern DRC outbreak resolved. We rapidly developed and implemented a smartphone application to estimate the exposure window for EVD cases in an outbreak setting
- 5 August 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (8) , e0255631
Open AccessDevelopment and implementation of the Ebola Exposure Window Calculator: A tool for Ebola virus disease outbreak field investigations
During an Ebola virus disease (EVD) outbreak, calculating the exposure window of a confirmed case can assist field investigators in identifying the source of infection and establishing chains of transmission. However, field investigators often have difficulty calculating this window. We developed a bilingual (English/French), smartphone-based field application to assist field investigators in determining the exposure window of an EVD case. The calculator only requires the reported date of symptoms onset and the type of symptoms present at onset or the date of death. Prior to the release of this application, there was no similar electronic capability to enable consistent calculation of EVD exposure windows for field investigators. The Democratic Republic of the Congo Ministry of Health endorsed the application and incorporated it into trainings for field staff. Available for Apple and Android devices, the calculator continues to be downloaded even as the eastern DRC outbreak resolved. We rapidly developed and implemented a smartphone application to estimate the exposure window for EVD cases in an outbreak setting
- Open Access
- 19 February 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (2) , e0247002
Chainchecker: An application to visualise and explore transmission chains for Ebola virus disease
2020 saw the continuation of the second largest outbreak of Ebola virus disease (EVD) in history. Determining epidemiological links between cases is a key part of outbreak control. However, due to the large quantity of data and subsequent data entry errors, inconsistencies in potential epidemiological links are difficult to identify. We present chainchecker, an online and offline shiny application which visualises, curates and verifies transmission chain data. The application includes the calculation of exposure windows for individual cases of EVD based on user defined incubation periods and user specified symptom profiles. It has an upload function for viral hemorrhagic fever data and utility for additional entries. This data may then be visualised as a transmission tree with inconsistent links highlighted. Finally, there is utility for cluster analysis and the ability to highlight nosocomial transmission. chainchecker is a R shiny application which has an offline version for use with VHF (viral hemorrhagic fever) databases or linelists. The software is available athttps://shiny.dide.imperial.ac.uk/chaincheckerwhich is a web-based application that links to the desktop application available for download and the github repository,https://github.com/imperialebola2018/chainchecker.
- 19 February 2021
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 16 (2) , e0247002
Open AccessChainchecker: An application to visualise and explore transmission chains for Ebola virus disease
2020 saw the continuation of the second largest outbreak of Ebola virus disease (EVD) in history. Determining epidemiological links between cases is a key part of outbreak control. However, due to the large quantity of data and subsequent data entry errors, inconsistencies in potential epidemiological links are difficult to identify. We present chainchecker, an online and offline shiny application which visualises, curates and verifies transmission chain data. The application includes the calculation of exposure windows for individual cases of EVD based on user defined incubation periods and user specified symptom profiles. It has an upload function for viral hemorrhagic fever data and utility for additional entries. This data may then be visualised as a transmission tree with inconsistent links highlighted. Finally, there is utility for cluster analysis and the ability to highlight nosocomial transmission. chainchecker is a R shiny application which has an offline version for use with VHF (viral hemorrhagic fever) databases or linelists. The software is available athttps://shiny.dide.imperial.ac.uk/chaincheckerwhich is a web-based application that links to the desktop application available for download and the github repository,https://github.com/imperialebola2018/chainchecker.
- 8 January 2021
- Published by Centers for Disease Control MMWR Office in MMWR. Recommendations and Reports
- Vol. 70 (1) , 1-12
Use of Ebola Vaccine: Recommendations of the Advisory Committee on Immunization Practices, United States, 2020
Summary of the Advisory Committee on Immunization Practices’ recommendations for use of Ebola vaccine in the United States.
- 8 January 2021
- Published by Centers for Disease Control MMWR Office in MMWR. Recommendations and Reports
- Vol. 70 (1) , 1-12
Use of Ebola Vaccine: Recommendations of the Advisory Committee on Immunization Practices, United States, 2020
Summary of the Advisory Committee on Immunization Practices’ recommendations for use of Ebola vaccine in the United States.
- Open Access
- 1 July 2018
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 5 (7) , ofy131
A Case of Lassa Fever Diagnosed at a Community Hospital—Minnesota 2014
Background: In April 2014, a 46-year-old returning traveler from Liberia was transported by emergency medical services to a community hospital in Minnesota with fever and altered mental status. Twenty-four hours later, he developed gingival bleeding. Blood samples tested positive for Lassa fever RNA by reverse transcriptase polymerase chain reaction. Methods: Blood and urine samples were obtained from the patient and tested for evidence of Lassa fever virus infection. Hospital infection control personnel and health department personnel reviewed infection control practices with health care personnel. In addition to standard precautions, infection control measures were upgraded to include contact, droplet, and airborne precautions. State and federal public health officials conducted contract tracing activities among family contacts, health care personnel, and fellow airline travelers. Results: The patient was discharged from the hospital after 14 days. However, his recovery was complicated by the development of near complete bilateral sensorineural hearing loss. Lassa virus RNA continued to be detected in his urine for several weeks after hospital discharge. State and federal public health authorities identified and monitored individuals who had contact with the patient while he was ill. No secondary cases of Lassa fever were identified among 75 contacts. Conclusions: Given the nonspecific presentation of viral hemorrhagic fevers, isolation of ill travelers and consistent implementation of basic infection control measures are key to preventing secondary transmission. When consistently applied, these measures can prevent secondary transmission even if travel history information is not obtained, not immediately available, or the diagnosis of a viral hemorrhagic fever is delayed.
- 1 July 2018
- Published by Oxford University Press (OUP) in Open Forum Infectious Diseases
- Vol. 5 (7) , ofy131
Open AccessA Case of Lassa Fever Diagnosed at a Community Hospital—Minnesota 2014
Background: In April 2014, a 46-year-old returning traveler from Liberia was transported by emergency medical services to a community hospital in Minnesota with fever and altered mental status. Twenty-four hours later, he developed gingival bleeding. Blood samples tested positive for Lassa fever RNA by reverse transcriptase polymerase chain reaction. Methods: Blood and urine samples were obtained from the patient and tested for evidence of Lassa fever virus infection. Hospital infection control personnel and health department personnel reviewed infection control practices with health care personnel. In addition to standard precautions, infection control measures were upgraded to include contact, droplet, and airborne precautions. State and federal public health officials conducted contract tracing activities among family contacts, health care personnel, and fellow airline travelers. Results: The patient was discharged from the hospital after 14 days. However, his recovery was complicated by the development of near complete bilateral sensorineural hearing loss. Lassa virus RNA continued to be detected in his urine for several weeks after hospital discharge. State and federal public health authorities identified and monitored individuals who had contact with the patient while he was ill. No secondary cases of Lassa fever were identified among 75 contacts. Conclusions: Given the nonspecific presentation of viral hemorrhagic fevers, isolation of ill travelers and consistent implementation of basic infection control measures are key to preventing secondary transmission. When consistently applied, these measures can prevent secondary transmission even if travel history information is not obtained, not immediately available, or the diagnosis of a viral hemorrhagic fever is delayed.
- Open Access
- 7 September 2017
- Published by American Society of Tropical Medicine and Hygiene in The American Journal of Tropical Medicine and Hygiene
- Vol. 97 (3) , 964-973
The Effects of an Integrated Community Case Management Strategy on the Appropriate Treatment of Children and Child Mortality in Kono District, Sierra Leone: A Program Evaluation
Integrated community case management (iCCM) aims to reduce child mortality in areas with poor access to health care. iCCM was implemented in 2009 in Kono district, Sierra Leone, a postconflict area with high under-five mortality rates (U5MRs). We evaluated iCCM’s impact and effects on child health using cluster surveys in 2010 (midterm) and 2013 (endline) to compare indicators on child mortality, coverage of appropriate treatment, timely access to care, quality of care, and recognition of community health workers (CHWs). The sample size was powered to detect a 28% decline in U5MR. Clusters were selected proportional to population size. All households were sampled to measure mortality and systematic random sampling was used to measure coverage in a subset of households. We used program data to evaluate utilization and access; 5,257 (2010) and 3,649 (2013) households were surveyed. U5MR did not change significantly (4.54 [95% confidence interval [CI]: 3.47–5.60] to 3.95 [95% CI: 3.06–4.83] deaths per 1,000 per month (P = 0.4)) though a relative change smaller than 28% could not be detected. CHWs were the first source of care for 52% (2010) and 50.9% (2013) of children. Coverage of appropriate treatment of fever by CHWs or peripheral health units increased from 45.5% [95% CI: 39.2–52.0] to 58.2% [95% CI: 50.5–65.5] (P = 0.01); changes for diarrhea and pneumonia were not significant. The continued reliance on the CHW as the first source of care and improved coverage for the appropriate treatment of fever support iCCM’s role in Kono district.
- 7 September 2017
- Published by American Society of Tropical Medicine and Hygiene in The American Journal of Tropical Medicine and Hygiene
- Vol. 97 (3) , 964-973
Open AccessThe Effects of an Integrated Community Case Management Strategy on the Appropriate Treatment of Children and Child Mortality in Kono District, Sierra Leone: A Program Evaluation
Integrated community case management (iCCM) aims to reduce child mortality in areas with poor access to health care. iCCM was implemented in 2009 in Kono district, Sierra Leone, a postconflict area with high under-five mortality rates (U5MRs). We evaluated iCCM’s impact and effects on child health using cluster surveys in 2010 (midterm) and 2013 (endline) to compare indicators on child mortality, coverage of appropriate treatment, timely access to care, quality of care, and recognition of community health workers (CHWs). The sample size was powered to detect a 28% decline in U5MR. Clusters were selected proportional to population size. All households were sampled to measure mortality and systematic random sampling was used to measure coverage in a subset of households. We used program data to evaluate utilization and access; 5,257 (2010) and 3,649 (2013) households were surveyed. U5MR did not change significantly (4.54 [95% confidence interval [CI]: 3.47–5.60] to 3.95 [95% CI: 3.06–4.83] deaths per 1,000 per month (P = 0.4)) though a relative change smaller than 28% could not be detected. CHWs were the first source of care for 52% (2010) and 50.9% (2013) of children. Coverage of appropriate treatment of fever by CHWs or peripheral health units increased from 45.5% [95% CI: 39.2–52.0] to 58.2% [95% CI: 50.5–65.5] (P = 0.01); changes for diarrhea and pneumonia were not significant. The continued reliance on the CHW as the first source of care and improved coverage for the appropriate treatment of fever support iCCM’s role in Kono district.
- Open Access
- 1 January 2017
- Published by Elsevier in The Lancet Global Health
- Vol. 5 (1) , e33
Screening of genital fluid for Ebola virus – Authors' reply
- 1 January 2017
- Published by Elsevier in The Lancet Global Health
- Vol. 5 (1) , e33
Open AccessScreening of genital fluid for Ebola virus – Authors' reply
- 1 October 2016
- Published by Elsevier in The Lancet Infectious Diseases
- Vol. 16 (10) , 1107-1108
Zika virus in semen: lessons from Ebola
- 1 October 2016
- Published by Elsevier in The Lancet Infectious Diseases
- Vol. 16 (10) , 1107-1108
Zika virus in semen: lessons from Ebola