Callum Arnold
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
Callum Arnold has a total of 102 co-authors. Here are the top co-authors by number of shared publications:
Michelle Science
Public Health Ontario, Toronto, Ontario, Canada
4 shared publicationsElizabeth McLachlan
National Microbiology Laboratory Branch, Public Health Agency of Canada, 1015 Arlington Street, Winnipeg, MB, Canada, R3E 3M4; Department of Medical Microbiology and Infectious Diseases, University of Manitoba, 745 Bannatyne Avenue, Winnipeg, MB, Canada, R3E 0J9
3 shared publications- Shelly Bolotin
Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada; Centre for Vaccine Preventable Diseases, University of Toronto, Toronto, ON, Canada; Public Health Ontario, Toronto, ON, Canada; Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada
3 shared publications Alberto Severini
National Microbiology Laboratory, Public Health Agency of Canada, Ottawa, Ontario, Canada
2 shared publications- Bryan Patenaude
Affiliation Department of International Health, Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland, United States of America
2 shared publications Matthew J. Ferrari
Affiliations Department of Biology, Pennsylvania State University, University Park, Pennsylvania, United States of America, Center for Infectious Disease Dynamics, Pennsylvania State University, University Park, Pennsylvania, United States of America
2 shared publications
Latest Publications
- Open Access
- 29 April 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (4) , e1013309
Individual and population level uncertainty interact to determine the performance of outbreak surveillance systems
Background: Outbreak detection frequently relies on imperfect individual-level case diagnosis. Both outbreaks and cases are discrete events that can be misclassified and uncertainty at the case level may impact the performance of outbreak alert and detection systems. Here, we describe how the performance of outbreak detection depends on individual-level diagnostic test characteristics and population-level epidemiology, and describe settings where imperfect individual-level tests can achieve consistent performance comparable to “perfect” diagnostic tests. Methodology: We generated a stochastic SEIR model to simulate daily incidence of measles (i.e., true) and non-measles (i.e., noise) febrile rash illness. We modeled non-measles sources as either independent static (Poisson) noise, or dynamical noise consistent with an independent SEIR process (e.g., rubella). Defining outbreak alerts as the exceedance of a threshold by the 7-day rolling average of observed test positives, we optimized the threshold that maximized outbreak detection accuracy across sets of noise structures and magnitudes, diagnostic test accuracy (consistent with either a perfect test, or proposed rapid diagnostic tests), and testing rates. Conclusions: The optimal threshold for each diagnostic test typically increased monotonically with testing rate. With static noise, outbreak detection with RDT-like and perfect tests achieved accuracies of 90%, with comparable delays to outbreak detection. With dynamical noise, the accuracy of perfect test scenarios was superior to those achieved with RDTs (≈ 90% vs. ≤ 80%). Outbreak detection accuracy declined as dynamical noise increased and leads to permanent alert status with RDT-like tests at very high noise. The performance of an outbreak detection system is highly sensitive to the structure and the magnitude of the background noise. Depending on the epidemiological context, outbreak detection using RDTs can perform as well as perfect tests.
- 29 April 2026
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 22 (4) , e1013309
Open AccessIndividual and population level uncertainty interact to determine the performance of outbreak surveillance systems
Background: Outbreak detection frequently relies on imperfect individual-level case diagnosis. Both outbreaks and cases are discrete events that can be misclassified and uncertainty at the case level may impact the performance of outbreak alert and detection systems. Here, we describe how the performance of outbreak detection depends on individual-level diagnostic test characteristics and population-level epidemiology, and describe settings where imperfect individual-level tests can achieve consistent performance comparable to “perfect” diagnostic tests. Methodology: We generated a stochastic SEIR model to simulate daily incidence of measles (i.e., true) and non-measles (i.e., noise) febrile rash illness. We modeled non-measles sources as either independent static (Poisson) noise, or dynamical noise consistent with an independent SEIR process (e.g., rubella). Defining outbreak alerts as the exceedance of a threshold by the 7-day rolling average of observed test positives, we optimized the threshold that maximized outbreak detection accuracy across sets of noise structures and magnitudes, diagnostic test accuracy (consistent with either a perfect test, or proposed rapid diagnostic tests), and testing rates. Conclusions: The optimal threshold for each diagnostic test typically increased monotonically with testing rate. With static noise, outbreak detection with RDT-like and perfect tests achieved accuracies of 90%, with comparable delays to outbreak detection. With dynamical noise, the accuracy of perfect test scenarios was superior to those achieved with RDTs (≈ 90% vs. ≤ 80%). Outbreak detection accuracy declined as dynamical noise increased and leads to permanent alert status with RDT-like tests at very high noise. The performance of an outbreak detection system is highly sensitive to the structure and the magnitude of the background noise. Depending on the epidemiological context, outbreak detection using RDTs can perform as well as perfect tests.
- Open Access
- 13 February 2026
- Published by Elsevier in Journal of Infection
- Vol. 92 (3) , 106701
Assessing measles immunity in individuals of childbearing age in Toronto, Ontario, Canada
- 13 February 2026
- Published by Elsevier in Journal of Infection
- Vol. 92 (3) , 106701
Open AccessAssessing measles immunity in individuals of childbearing age in Toronto, Ontario, Canada
- Open Access
Individual and Population Level Uncertainty Interact to Determine the Performance of Outbreak Surveillance Systems
Background: Outbreak detection frequently relies on imperfect individual-level case diagnosis. Both outbreaks and cases are discrete events that can be misclassified and uncertainty at the case level may impact the performance of outbreak alert and detection systems. Here, we describe how the performance of outbreak detection depends on individual-level diagnostic test characteristics and population-level epidemiology and describe settings where imperfect individual-level tests can achieve consistent performance comparable to “perfect” diagnostic tests. Methodology: We generated a stochastic SEIR model to simulate daily incidence of measles (i.e., true) and non-measles (i.e., noise) febrile rash illness. We modeled non-measles sources as either independent static (Poisson) noise, or dynamical noise consistent with an independent SEIR process (e.g., rubella). Defining outbreak alerts as the exceedance of a threshold by the 7-day rolling average of observed test positives, we optimized the threshold that maximized outbreak detection accuracy across set of noise structures and magnitudes, diagnostic test accuracy (consistent with either a perfect test, or proposed rapid diagnostic tests), and testing rates. Conclusions: The optimal threshold for each diagnostic test typically increased monotonically with testing rate. With static noise, outbreak detection with RDT-like and perfect tests achieved accuracies of 90%, with comparable delays to outbreak detection. With dynamical noise, the accuracy of perfect test scenarios was superior to those achieved with RDTs (≈ 90% vs.≈ 80%). Outbreak detection accuracy declined as dynamical noise increased and leads to permanent alert status with RDT-like tests at very high noise. The performance of an outbreak detection system is highly sensitive to the structure and the magnitude of the background noise. Depending on the epidemiological context, outbreak detection using RDTs can perform as well as perfect tests. Author Summary: To respond to outbreaks of infectious diseases, we first need to detect them. This detection is inherently flawed, in part, due to imperfect diagnostic tests used to indicate whether individuals are positive or negative for a disease. We evaluated the impact of imperfect diagnostic tests for infectious disease on the accuracy and timeliness of outbreak detection in the context of a set of background infections that could be mistaken for the disease of interest, and consequently cause false positive test results. We find that when outbreak detection performance is highly dependent on the structure and magnitude of the background “noise” infections. When the rate of background infections far exceeds that of the target infection, and are dynamical, such that there are large peaks and troughs of “noise infections”, imperfect diagnostic tests are not able to accurately distinguish the “signal” (target infections) from the background “noise”. If the background “noise” infections are either less cyclical in their dynamics, or do not outnumber true infections by a great deal, imperfect diagnostic tests can perform well.
Open AccessIndividual and Population Level Uncertainty Interact to Determine the Performance of Outbreak Surveillance Systems
Background: Outbreak detection frequently relies on imperfect individual-level case diagnosis. Both outbreaks and cases are discrete events that can be misclassified and uncertainty at the case level may impact the performance of outbreak alert and detection systems. Here, we describe how the performance of outbreak detection depends on individual-level diagnostic test characteristics and population-level epidemiology and describe settings where imperfect individual-level tests can achieve consistent performance comparable to “perfect” diagnostic tests. Methodology: We generated a stochastic SEIR model to simulate daily incidence of measles (i.e., true) and non-measles (i.e., noise) febrile rash illness. We modeled non-measles sources as either independent static (Poisson) noise, or dynamical noise consistent with an independent SEIR process (e.g., rubella). Defining outbreak alerts as the exceedance of a threshold by the 7-day rolling average of observed test positives, we optimized the threshold that maximized outbreak detection accuracy across set of noise structures and magnitudes, diagnostic test accuracy (consistent with either a perfect test, or proposed rapid diagnostic tests), and testing rates. Conclusions: The optimal threshold for each diagnostic test typically increased monotonically with testing rate. With static noise, outbreak detection with RDT-like and perfect tests achieved accuracies of 90%, with comparable delays to outbreak detection. With dynamical noise, the accuracy of perfect test scenarios was superior to those achieved with RDTs (≈ 90% vs.≈ 80%). Outbreak detection accuracy declined as dynamical noise increased and leads to permanent alert status with RDT-like tests at very high noise. The performance of an outbreak detection system is highly sensitive to the structure and the magnitude of the background noise. Depending on the epidemiological context, outbreak detection using RDTs can perform as well as perfect tests. Author Summary: To respond to outbreaks of infectious diseases, we first need to detect them. This detection is inherently flawed, in part, due to imperfect diagnostic tests used to indicate whether individuals are positive or negative for a disease. We evaluated the impact of imperfect diagnostic tests for infectious disease on the accuracy and timeliness of outbreak detection in the context of a set of background infections that could be mistaken for the disease of interest, and consequently cause false positive test results. We find that when outbreak detection performance is highly dependent on the structure and magnitude of the background “noise” infections. When the rate of background infections far exceeds that of the target infection, and are dynamical, such that there are large peaks and troughs of “noise infections”, imperfect diagnostic tests are not able to accurately distinguish the “signal” (target infections) from the background “noise”. If the background “noise” infections are either less cyclical in their dynamics, or do not outnumber true infections by a great deal, imperfect diagnostic tests can perform well.
- Open Access
The Maximal Expected Benefit of SARS-CoV-2 Interventions Among University Students: A Simulation Study Using Latent Class Analysis
Non-pharmaceutical public health measures (PHMs) were central to pre-vaccination efforts to reduce Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) exposure risk; heterogeneity in adherence placed bounds on their potential effectiveness, and correlation in their adoption makes assessing the impact attributable to an individual PHM difficult. During the Fall 2020 semester, we used a longitudinal cohort design in a university student population to conduct a behavioral survey of intention to adhere to PHMs, paired with an IgG serosurvey to quantify SARS-CoV-2 exposure at the end of the semester. Using Latent Class Analysis on behavioral survey responses, we identified three distinct groups among the 673 students with IgG samples: 256 (38.04%) students were in the most adherent group, intending to follow all guidelines, 306 (46.21%) in the moderately-adherent group, and 111 (15.75%) in the least-adherent group, rarely intending to follow any measure, with adherence negatively correlated with seropositivity of 25.4%, 32.2% and 37.7%, respectively. Moving all individuals in an SIR model into the most adherent group resulted in a 76-93% reduction in seroprevalence, dependent on assumed assortativity. The potential impact of increasing PHM adherence was limited by the substantial exposure risk in the large proportion of students already following all PHMs.
Open AccessThe Maximal Expected Benefit of SARS-CoV-2 Interventions Among University Students: A Simulation Study Using Latent Class Analysis
Non-pharmaceutical public health measures (PHMs) were central to pre-vaccination efforts to reduce Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) exposure risk; heterogeneity in adherence placed bounds on their potential effectiveness, and correlation in their adoption makes assessing the impact attributable to an individual PHM difficult. During the Fall 2020 semester, we used a longitudinal cohort design in a university student population to conduct a behavioral survey of intention to adhere to PHMs, paired with an IgG serosurvey to quantify SARS-CoV-2 exposure at the end of the semester. Using Latent Class Analysis on behavioral survey responses, we identified three distinct groups among the 673 students with IgG samples: 256 (38.04%) students were in the most adherent group, intending to follow all guidelines, 306 (46.21%) in the moderately-adherent group, and 111 (15.75%) in the least-adherent group, rarely intending to follow any measure, with adherence negatively correlated with seropositivity of 25.4%, 32.2% and 37.7%, respectively. Moving all individuals in an SIR model into the most adherent group resulted in a 76-93% reduction in seroprevalence, dependent on assumed assortativity. The potential impact of increasing PHM adherence was limited by the substantial exposure risk in the large proportion of students already following all PHMs.
- Open Access
- 10 November 2023
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (11) , e0287765
Maternal varicella antibodies in children aged less than one year: Assessment of antibody decay
Objectives: To investigate maternal antibody levels to varicella in infants <12 months of age in Ontario, Canada. Study design: In this study, we included specimens from infants <12 months of age, born at ≥37 weeks gestational age, who had sera collected at The Hospital for Sick Children (Toronto, Canada) between 2014–2016. We tested sera using a glycoprotein-based enzyme-linked immunosorbent assay (gpELISA). We measured varicella susceptibility (antibody concentration <150mIU/mL) and mean varicella antibody concentration, and assessed the probability of susceptibility and concentration between one and 11 months of age using multivariable logistic regression and Poisson regression. Results: We found that 32% of 196 included specimens represented infants susceptible to varicella at one month of age, increasing to nearly 80% at three months of age. At six months of age, all infants were susceptible to varicella and the predicted mean varicella antibody concentration declined to 62 mIU/mL (95% confidence interval 40, 84), well below the threshold of protection. Conclusions: We found that varicella maternal antibody levels wane rapidly in infants, leaving most infants susceptible by four months of age. Our findings have implications for the timing of first dose of varicella-containing vaccine, infection control measures, and infant post-exposure prophylaxis recommendations.
- 10 November 2023
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 18 (11) , e0287765
Open AccessMaternal varicella antibodies in children aged less than one year: Assessment of antibody decay
Objectives: To investigate maternal antibody levels to varicella in infants <12 months of age in Ontario, Canada. Study design: In this study, we included specimens from infants <12 months of age, born at ≥37 weeks gestational age, who had sera collected at The Hospital for Sick Children (Toronto, Canada) between 2014–2016. We tested sera using a glycoprotein-based enzyme-linked immunosorbent assay (gpELISA). We measured varicella susceptibility (antibody concentration <150mIU/mL) and mean varicella antibody concentration, and assessed the probability of susceptibility and concentration between one and 11 months of age using multivariable logistic regression and Poisson regression. Results: We found that 32% of 196 included specimens represented infants susceptible to varicella at one month of age, increasing to nearly 80% at three months of age. At six months of age, all infants were susceptible to varicella and the predicted mean varicella antibody concentration declined to 62 mIU/mL (95% confidence interval 40, 84), well below the threshold of protection. Conclusions: We found that varicella maternal antibody levels wane rapidly in infants, leaving most infants susceptible by four months of age. Our findings have implications for the timing of first dose of varicella-containing vaccine, infection control measures, and infant post-exposure prophylaxis recommendations.
- Open Access
SARS-CoV-2 Seroprevalence in a University Community: A Longitudinal Study of the Impact of Student Return to Campus on Infection Risk Among Community Members
Background Returning university students represent large-scale, transient demographic shifts and a potential source of transmission to adjacent communities during the COVID-19 pandemic. Methods In this prospective longitudinal cohort study, we tested for IgG antibodies against SARS-CoV-2 in a non-random cohort of residents living in Centre County prior to the Fall 2020 term at the Pennsylvania State University and following the conclusion of the Fall 2020 term. We also report the seroprevalence in a non-random cohort of students collected at the end of the Fall 2020 term. Findings Of 345 community participants, 19 (5·5%) were positive for SARS-CoV-2 IgG antibodies at their first visit between 7 August and 2 October. Of 625 student participants who returned to campus for fall instruction, 195 (31·2%) were positive for SARS-CoV-2 antibodies between 26 October and 23 November. Twenty-eight (8·1%) community participants returned a positive IgG antibody result by 9 December. Only contact with known SARS-CoV-2-positive individuals and attendance at small gatherings (20-50 individuals) were significant predictors of detecting IgG antibodies among returning students (aOR, 95% CI: 3·24, 2·14-4·91; 1·62, 1·08-2·44; respectively). Interpretation Despite high seroprevalence observed within the student population, seroprevalence in a longitudinal cohort of community residents was low and stable from before student arrival for the Fall 2020 term to after student departure. The study implies that heterogeneity in SARS-CoV-2 transmission can occur in geographically coincident populations. Funding The Pennsylvania State University Office of the Provost, Social Science Research Institute, Huck Institute for the Life Sciences, and Clinical and Translational Science Institute; National Institutes of Health. Evidence before this study This study was initiated in Spring of 2020, during the pandemic but prior to the return of students to US universities. There was no prior research on the consequences of the return to in-person university instruction on SARS-CoV-2 transmission. While the explicit case of transmission of SARS-CoV-2 between student and resident communities was poorly understood at the start of this study, there is a well-established literature that has documented the role that large population movements, consistent with the return of students to campuses, play in the introduction and maintenance of pathogen transmission or diseases. Several studies documented the role of travelers in the introduction of SARS-CoV-2 to populations. The return of students to colleges and universities creates the potential for SARS-CoV-2 introduction and transmission with a sustained increase in population size, density, and mixing. No prior study has considered the long-term transmission consequences of such demographic change on the pre-influx population. We searched PubMed, medRxiv, bioRxiv, and Google Scholar between May 2020 and September 2020. We used search terms “seroprevalence”, “anti-SARS-CoV-2”, “IgG”, “COVID antibodies” and “university”, “college”. Several modeling studies had been conducted to project likely prevalence of anti-SARS-CoV-2 antibodies in student and resident populations resulting from the return of students to in-person instruction in the Fall 2020 term in the US and UK. Several studies conducted after September 2020 described the county-level incidence of SARS-CoV-2 in counties that contain colleges and universities. None of these studies specifically disaggregated the incidence of COVID in resident and student populations. Added value of this study There is clear evidence that in-person university instruction is correlated with high incidence in students and higher incidence in counties with colleges and universities in counties those without such institutions. However, this is the first study to specifically quantify the change in seroprevalence in non-student residents associated with the return of students to in-person instruction in a large university setting. Despite high COVID-19 incidence in the student population, seroprevalence in the community resident cohort was low and stable from before student arrival for the Fall 2020 term until after student departure. The results of this study highlight the potential for significant heterogeneity in incidence in geographically coincident sub-populations. Implications of all the available evidence The influx of students into college and university towns reflects a significant increase in population size and density that could result in increases in COVID-19 risk in surrounding communities. Despite the potential risk, it is possible that SARS-CoV-2 transmission may lag behind in surrounding communities This implies the potential to minimize the transmission of SARS-CoV-2 from high prevalence sub-populations with targeted interventions. Future work should investigate the relative efficacy of specific measures to reduce risk of transmission between subpopulations and guide communities with high levels of episodic migration and similar high-density subpopulations.
Open AccessSARS-CoV-2 Seroprevalence in a University Community: A Longitudinal Study of the Impact of Student Return to Campus on Infection Risk Among Community Members
Background Returning university students represent large-scale, transient demographic shifts and a potential source of transmission to adjacent communities during the COVID-19 pandemic. Methods In this prospective longitudinal cohort study, we tested for IgG antibodies against SARS-CoV-2 in a non-random cohort of residents living in Centre County prior to the Fall 2020 term at the Pennsylvania State University and following the conclusion of the Fall 2020 term. We also report the seroprevalence in a non-random cohort of students collected at the end of the Fall 2020 term. Findings Of 345 community participants, 19 (5·5%) were positive for SARS-CoV-2 IgG antibodies at their first visit between 7 August and 2 October. Of 625 student participants who returned to campus for fall instruction, 195 (31·2%) were positive for SARS-CoV-2 antibodies between 26 October and 23 November. Twenty-eight (8·1%) community participants returned a positive IgG antibody result by 9 December. Only contact with known SARS-CoV-2-positive individuals and attendance at small gatherings (20-50 individuals) were significant predictors of detecting IgG antibodies among returning students (aOR, 95% CI: 3·24, 2·14-4·91; 1·62, 1·08-2·44; respectively). Interpretation Despite high seroprevalence observed within the student population, seroprevalence in a longitudinal cohort of community residents was low and stable from before student arrival for the Fall 2020 term to after student departure. The study implies that heterogeneity in SARS-CoV-2 transmission can occur in geographically coincident populations. Funding The Pennsylvania State University Office of the Provost, Social Science Research Institute, Huck Institute for the Life Sciences, and Clinical and Translational Science Institute; National Institutes of Health. Evidence before this study This study was initiated in Spring of 2020, during the pandemic but prior to the return of students to US universities. There was no prior research on the consequences of the return to in-person university instruction on SARS-CoV-2 transmission. While the explicit case of transmission of SARS-CoV-2 between student and resident communities was poorly understood at the start of this study, there is a well-established literature that has documented the role that large population movements, consistent with the return of students to campuses, play in the introduction and maintenance of pathogen transmission or diseases. Several studies documented the role of travelers in the introduction of SARS-CoV-2 to populations. The return of students to colleges and universities creates the potential for SARS-CoV-2 introduction and transmission with a sustained increase in population size, density, and mixing. No prior study has considered the long-term transmission consequences of such demographic change on the pre-influx population. We searched PubMed, medRxiv, bioRxiv, and Google Scholar between May 2020 and September 2020. We used search terms “seroprevalence”, “anti-SARS-CoV-2”, “IgG”, “COVID antibodies” and “university”, “college”. Several modeling studies had been conducted to project likely prevalence of anti-SARS-CoV-2 antibodies in student and resident populations resulting from the return of students to in-person instruction in the Fall 2020 term in the US and UK. Several studies conducted after September 2020 described the county-level incidence of SARS-CoV-2 in counties that contain colleges and universities. None of these studies specifically disaggregated the incidence of COVID in resident and student populations. Added value of this study There is clear evidence that in-person university instruction is correlated with high incidence in students and higher incidence in counties with colleges and universities in counties those without such institutions. However, this is the first study to specifically quantify the change in seroprevalence in non-student residents associated with the return of students to in-person instruction in a large university setting. Despite high COVID-19 incidence in the student population, seroprevalence in the community resident cohort was low and stable from before student arrival for the Fall 2020 term until after student departure. The results of this study highlight the potential for significant heterogeneity in incidence in geographically coincident sub-populations. Implications of all the available evidence The influx of students into college and university towns reflects a significant increase in population size and density that could result in increases in COVID-19 risk in surrounding communities. Despite the potential risk, it is possible that SARS-CoV-2 transmission may lag behind in surrounding communities This implies the potential to minimize the transmission of SARS-CoV-2 from high prevalence sub-populations with targeted interventions. Future work should investigate the relative efficacy of specific measures to reduce risk of transmission between subpopulations and guide communities with high levels of episodic migration and similar high-density subpopulations.
- Open Access
- 12 August 2020
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 42 (3) , 261-267
Healthcare worker perception of a global outbreak of novel coronavirus (COVID-19) and personal protective equipment: Survey of a pediatric tertiary-care hospital
Objective: In this study, we aimed to capture perspectives of healthcare workers (HCWs) on coronavirus disease 2019 (COVID-19) and infection prevention and control (IPAC) measures implemented during the early phase of the COVID-19 pandemic.Design: A cross-sectional survey of HCWs.Participants: HCWs from the Hospital for Sick Children, Toronto, Canada.Intervention: A self-administered survey was distributed to HCWs. We analyzed factors influencing HCW knowledge and self-reported use of personal protective equipment (PPE), concerns about contracting COVID-19 and acceptance of the recommended IPAC precautions for COVID-19.Results: In total, 175 HCWs completed the survey between March 6 and March 10: 35 staff physicians (20%), 24 residents or fellows (14%), 72 nurses (41%), 14 respiratory therapists (8%), 14 administration staff (8%), and 14 other employees (8%). Most of the respondents were from the emergency department (n = 58, 33%) and the intensive care unit (n = 58, 33%). Only 86 respondents (50%) identified the correct donning order; only 60 (35%) identified the correct doffing order; but the majority (n = 113, 70%) indicated the need to wash their hands immediately prior to removal of their mask and eye protection. Also, 91 (54%) respondents felt comfortable with recommendations for droplet and/or contact precautions for routine care of patients with COVID-19. HCW occupation and concerns about contracting COVID-19 outside work were associated with nonacceptance of the recommendations (P= .016 andP= .036 respectively).Conclusion: As part of their pandemic response plans, healthcare institutions should have ongoing training for HCWs that focus on appropriate PPE doffing and discussions around modes of transmission of COVID-19.
- 12 August 2020
- Published by Cambridge University Press (CUP) in Infection Control & Hospital Epidemiology
- Vol. 42 (3) , 261-267
Open AccessHealthcare worker perception of a global outbreak of novel coronavirus (COVID-19) and personal protective equipment: Survey of a pediatric tertiary-care hospital
Objective: In this study, we aimed to capture perspectives of healthcare workers (HCWs) on coronavirus disease 2019 (COVID-19) and infection prevention and control (IPAC) measures implemented during the early phase of the COVID-19 pandemic.Design: A cross-sectional survey of HCWs.Participants: HCWs from the Hospital for Sick Children, Toronto, Canada.Intervention: A self-administered survey was distributed to HCWs. We analyzed factors influencing HCW knowledge and self-reported use of personal protective equipment (PPE), concerns about contracting COVID-19 and acceptance of the recommended IPAC precautions for COVID-19.Results: In total, 175 HCWs completed the survey between March 6 and March 10: 35 staff physicians (20%), 24 residents or fellows (14%), 72 nurses (41%), 14 respiratory therapists (8%), 14 administration staff (8%), and 14 other employees (8%). Most of the respondents were from the emergency department (n = 58, 33%) and the intensive care unit (n = 58, 33%). Only 86 respondents (50%) identified the correct donning order; only 60 (35%) identified the correct doffing order; but the majority (n = 113, 70%) indicated the need to wash their hands immediately prior to removal of their mask and eye protection. Also, 91 (54%) respondents felt comfortable with recommendations for droplet and/or contact precautions for routine care of patients with COVID-19. HCW occupation and concerns about contracting COVID-19 outside work were associated with nonacceptance of the recommendations (P= .016 andP= .036 respectively).Conclusion: As part of their pandemic response plans, healthcare institutions should have ongoing training for HCWs that focus on appropriate PPE doffing and discussions around modes of transmission of COVID-19.
- Open Access
- 12 June 2020
- Published by SAGE Publications in CARTILAGE
- Vol. 13 (1_suppl) , 1761S-1771S
The Response of Hip Joint Cartilage to Exercise in Children: An MRI Study Using T2-Mapping
The aim of this study was to determine the effects of activity and cam morphology on cartilage composition during adolescence and investigate the development of cartilage composition with age. Cross-sectional observational cohort study of individuals from football club academies and an age-matched control population, aged 9 to 18 years. Assessments included questionnaires and T2-mapping of hips. Primary imaging outcome measures were T2 relaxation time of acetabular and femoral cartilage, average alpha angle, and lateral epiphyseal extension. The cohort consisted of 109 elite male footballers, 49 male controls, and 51 female controls. Elite male footballers had an acetabular cartilage T2 value 4.85 ms greater than male controls (P < 0.001). A significant positive correlation existed between Physical Activity Questionnaire Score and acetabular cartilage T2 value (coefficient 1.07, P < 0.001) and femoral cartilage T2 value (coefficient 0.66, P = 0.032). Individuals with a closed physis had an acetabular cartilage T2 value 7.86 ms less than individuals with an open physis. Acetabular cartilage T2 values decreased with age in elite footballers. No correlation existed between alpha angle and anterosuperior acetabular cartilage T2 value and no difference in T2 value existed between individuals with and without cam morphology. This study demonstrates that high activity levels may significantly affect acetabular cartilage composition during adolescence, but cam morphology may not detrimentally affect cartilage composition until after adolescence.
- 12 June 2020
- Published by SAGE Publications in CARTILAGE
- Vol. 13 (1_suppl) , 1761S-1771S
Open AccessThe Response of Hip Joint Cartilage to Exercise in Children: An MRI Study Using T2-Mapping
The aim of this study was to determine the effects of activity and cam morphology on cartilage composition during adolescence and investigate the development of cartilage composition with age. Cross-sectional observational cohort study of individuals from football club academies and an age-matched control population, aged 9 to 18 years. Assessments included questionnaires and T2-mapping of hips. Primary imaging outcome measures were T2 relaxation time of acetabular and femoral cartilage, average alpha angle, and lateral epiphyseal extension. The cohort consisted of 109 elite male footballers, 49 male controls, and 51 female controls. Elite male footballers had an acetabular cartilage T2 value 4.85 ms greater than male controls (P < 0.001). A significant positive correlation existed between Physical Activity Questionnaire Score and acetabular cartilage T2 value (coefficient 1.07, P < 0.001) and femoral cartilage T2 value (coefficient 0.66, P = 0.032). Individuals with a closed physis had an acetabular cartilage T2 value 7.86 ms less than individuals with an open physis. Acetabular cartilage T2 values decreased with age in elite footballers. No correlation existed between alpha angle and anterosuperior acetabular cartilage T2 value and no difference in T2 value existed between individuals with and without cam morphology. This study demonstrates that high activity levels may significantly affect acetabular cartilage composition during adolescence, but cam morphology may not detrimentally affect cartilage composition until after adolescence.
- Open Access
- 30 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0230721
Impact of maternal reproductive factors on cancer risks of offspring: A systematic review and meta-analysis of cohort studies
A number of studies have reported on associations between reproductive factors, such as delivery methods, number of birth and breastfeeding, and incidence of cancer in children, but systematic reviews addressing this issue to date have important limitations, and no reviews have addressed the impact of reproductive factors on cancer over the full life course of offspring. We performed a comprehensive search in MEDLINE, and Embase up to January 2020 and Web of Science up to 2018 July, including cohort studies reporting the association between maternal reproductive factors of age at birth, birth order, number of births, delivery methods, and breastfeeding duration and cancer in children. Teams of two reviewers independently extracted data and assessed risk of bias. We conducted random effects meta-analyses to estimate summary relative estimates, calculated absolute differences between those with and without risk factors, and used the GRADE approach to evaluate the certainty of evidence. For most exposures and most cancers, we found no suggestion of a causal relation. We found low to very low certainty evidence of the following very small possible impact: higher maternal age at birth with adult multiple myeloma and lifetime uterine cervix cancer incidence; lower maternal age at birth with childhood overall cancer mortality (RR = 1.15, 95% CI = 1.01–1.30; AR/10,000 = 1, 95% CI = 0 to 2), adult leukemia and lifetime uterine cervix cancer incidence; higher birth order with adult melanoma, cervix uteri, corpus uteri, thyroid cancer incidence, lifetime lung, corpus uteri, prostate, testis, sarcoma, thyroid cancer incidence; larger number of birth with childhood brain (RR = 1.27, 95% CI = 1.06–1.52; AR/10,000 = 1, 95% CI = 0 to 2), leukemia (RR = 2.11, 95% CI = 1.62–2.75; AR/10,000 = 9, 95% CI = 5 to 14), lymphoma (RR = 4.66, 95% CI = 1.40–15.57; AR/10,000 = 11, 95% CI = 1 to 44) incidence, adult stomach, corpus uteri cancer incidence and lung cancer mortality, lifetime stomach, lung, uterine cervix, uterine corpus, multiple myeloma, testis cancer incidence; Caesarean delivery with childhood kidney cancer incidence (RR = 1.25, 95% CI = 1.01–1.55; AR/10,000 = 0, 95% CI = 0 to 1); and breastfeeding with adult colorectal cancer incidence. Very small impacts existed between a number of reproductive factors and cancer incidence and mortality in children and the certainty of evidence was low to very low primarily due to observational design.
- 30 March 2020
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 15 (3) , e0230721
Open AccessImpact of maternal reproductive factors on cancer risks of offspring: A systematic review and meta-analysis of cohort studies
A number of studies have reported on associations between reproductive factors, such as delivery methods, number of birth and breastfeeding, and incidence of cancer in children, but systematic reviews addressing this issue to date have important limitations, and no reviews have addressed the impact of reproductive factors on cancer over the full life course of offspring. We performed a comprehensive search in MEDLINE, and Embase up to January 2020 and Web of Science up to 2018 July, including cohort studies reporting the association between maternal reproductive factors of age at birth, birth order, number of births, delivery methods, and breastfeeding duration and cancer in children. Teams of two reviewers independently extracted data and assessed risk of bias. We conducted random effects meta-analyses to estimate summary relative estimates, calculated absolute differences between those with and without risk factors, and used the GRADE approach to evaluate the certainty of evidence. For most exposures and most cancers, we found no suggestion of a causal relation. We found low to very low certainty evidence of the following very small possible impact: higher maternal age at birth with adult multiple myeloma and lifetime uterine cervix cancer incidence; lower maternal age at birth with childhood overall cancer mortality (RR = 1.15, 95% CI = 1.01–1.30; AR/10,000 = 1, 95% CI = 0 to 2), adult leukemia and lifetime uterine cervix cancer incidence; higher birth order with adult melanoma, cervix uteri, corpus uteri, thyroid cancer incidence, lifetime lung, corpus uteri, prostate, testis, sarcoma, thyroid cancer incidence; larger number of birth with childhood brain (RR = 1.27, 95% CI = 1.06–1.52; AR/10,000 = 1, 95% CI = 0 to 2), leukemia (RR = 2.11, 95% CI = 1.62–2.75; AR/10,000 = 9, 95% CI = 5 to 14), lymphoma (RR = 4.66, 95% CI = 1.40–15.57; AR/10,000 = 11, 95% CI = 1 to 44) incidence, adult stomach, corpus uteri cancer incidence and lung cancer mortality, lifetime stomach, lung, uterine cervix, uterine corpus, multiple myeloma, testis cancer incidence; Caesarean delivery with childhood kidney cancer incidence (RR = 1.25, 95% CI = 1.01–1.55; AR/10,000 = 0, 95% CI = 0 to 1); and breastfeeding with adult colorectal cancer incidence. Very small impacts existed between a number of reproductive factors and cancer incidence and mortality in children and the certainty of evidence was low to very low primarily due to observational design.
- 1 December 2019
- Published by American Academy of Pediatrics (AAP) in Pediatrics
- Vol. 144 (6)
Measles Antibody Levels in Young Infants
BACKGROUND: Infants are often assumed to be immune to measles through maternal antibodies transferred during pregnancy and, in many countries, receive their first measles-containing vaccine at 12 to 15 months. Immunity may wane before this time in measles-eliminated settings, placing infants at risk for measles and complications. We investigated humoral immunity to measles in infants <12 months of age in Ontario, Canada.METHODS: We selected sera collected at a tertiary pediatric hospital from infants <12 months who were born at ≥37 weeks’ gestational age. We excluded infants with conditions that affect antibody levels. We selected ≤25 sera from 8 predetermined age bands and tested them for measles-neutralizing antibody using the plaque-reduction neutralization test. We calculated the proportion immune at each age band, and predictors of infant susceptibility were assessed by using multivariable logistic regression and Poisson regression.RESULTS: Of 196 infant sera, 56% (110 of 196) were from boys, and 35% (69 of 196) were from infants with underlying medical conditions. In the first month, 20% (5 of 25) of infants had antibodies below the protective threshold, which increased to 92% (22 of 24) by 3 months. By 6 months, all infants had titers below the protective threshold. In a multivariable analysis, infant age was the strongest predictor of susceptibility (odds ratio = 2.13 for each additional month increase; 95% confidence interval: 1.52–2.97).CONCLUSIONS: Most infants were susceptible to measles by 3 months of age in this elimination setting. Our findings inform important policy discussions relating to the timing of the first dose of measles-containing vaccine and infant postexposure prophylaxis recommendations.
- 1 December 2019
- Published by American Academy of Pediatrics (AAP) in Pediatrics
- Vol. 144 (6)
Measles Antibody Levels in Young Infants
BACKGROUND: Infants are often assumed to be immune to measles through maternal antibodies transferred during pregnancy and, in many countries, receive their first measles-containing vaccine at 12 to 15 months. Immunity may wane before this time in measles-eliminated settings, placing infants at risk for measles and complications. We investigated humoral immunity to measles in infants <12 months of age in Ontario, Canada.METHODS: We selected sera collected at a tertiary pediatric hospital from infants <12 months who were born at ≥37 weeks’ gestational age. We excluded infants with conditions that affect antibody levels. We selected ≤25 sera from 8 predetermined age bands and tested them for measles-neutralizing antibody using the plaque-reduction neutralization test. We calculated the proportion immune at each age band, and predictors of infant susceptibility were assessed by using multivariable logistic regression and Poisson regression.RESULTS: Of 196 infant sera, 56% (110 of 196) were from boys, and 35% (69 of 196) were from infants with underlying medical conditions. In the first month, 20% (5 of 25) of infants had antibodies below the protective threshold, which increased to 92% (22 of 24) by 3 months. By 6 months, all infants had titers below the protective threshold. In a multivariable analysis, infant age was the strongest predictor of susceptibility (odds ratio = 2.13 for each additional month increase; 95% confidence interval: 1.52–2.97).CONCLUSIONS: Most infants were susceptible to measles by 3 months of age in this elimination setting. Our findings inform important policy discussions relating to the timing of the first dose of measles-containing vaccine and infant postexposure prophylaxis recommendations.