Validating International Classification of Disease 10th revision algorithms for identifying influenza and respiratory syncytial virus hospitalizations

Objective: Routinely collected health administrative data can be used to efficiently assess disease burden in large populations, but it is important to evaluate the validity of these data. The objective of this study was to develop and validate International Classification of Disease 10PthP revision (ICD -10) algorithms that identify laboratory-confirmed influenza or laboratory-confirmed respiratory syncytial virus (RSV) hospitalizations using population-based health administrative data from Ontario, Canada. Study Design and Setting: Influenza and RSV laboratory data from the 2014-15 through to 2017-18 respiratory virus seasons were obtained from the Ontario Laboratories Information System (OLIS) and were linked to hospital discharge abstract data to generate influenza and RSV reference cohorts. These reference cohorts were used to assess the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the ICD-10 algorithms. To minimize misclassification in future studies, we prioritized specificity and PPV in selecting top-performing algorithms. Results: 83,638 and 61,117 hospitalized patients were included in the influenza and RSV reference cohorts, respectively. The best influenza algorithm had a sensitivity of 73% (95% CI 72% to 74%), specificity of 99% (95% CI 99% to 99%), PPV of 94% (95% CI 94% to 95%), and NPV of 94% (95% CI 94% to 95%). The best RSV algorithm had a sensitivity of 69% (95% CI 68% to 70%), specificity of 99% (95% CI 99% to 99%), PPV of 91% (95% CI 90% to 91%) and NPV of 97% (95% CI 97% to 97%). Conclusion: We identified two highly specific algorithms that best ascertain patients hospitalized with influenza or RSV. These algorithms may be applied to hospitalized patients if data on laboratory tests are not available, and will thereby improve the power of future epidemiologic studies of influenza, RSV, and potentially other severe acute respiratory infections.

Predicting Methicillin-Resistant Staphylococcus aureus (MRSA) Bloodstream Infection Incidence Rates using Canadian Nosocomial Infection Surveillance Program (CNISP)

Methicillin-Resistant Staphylococcus aureus (MRSA) is among the most prevalent nosocomial pathogens globally, causing significant morbidity, mortality, and healthcare costs. MRSA bloodstream infection (BSI) incidence rates in Canadian hospitals have significantly risen by almost 60% and have a mortality of over 20% upon Intensive Care Unit admission. MRSA is believed to be spread through healthcare workers; thus, high hand hygiene compliancy in addition to environmental cleaning are the cornerstone countermeasures to disrupting its transmission. The Public Health Agency of Canada (PHAC), in collaboration with the Canadian Nosocomial Infection Surveillance Program (CNISP), conducts national, sentinel surveillance on healthcare-associated infections like MRSA. As a Student Epidemiologist, I developed a research proposal detailing two study objectives: 1) develop a regression model to predict all incident MRSA BSI rates among acute-care hospitals in Canada using CNISP MRSA BSI incident cases from 2000 to 2019, and 2) create a compartmental (Susceptible-Infected-Recovered-Deceased) model to determine the impact of various Infection Prevention and Control (IPC) measures on the risk of healthcare-associated MRSA BSI transmission specifically. This study hopes to demonstrate that proper IPC compliance is associated with lower incident MRSA BSI rates with the goal being to produce a manuscript draft by 2021. MRSA poses a serious threat to patient safety globally and is becoming a growing national public health concern in Canada; determining which IPC strategy is most effective at disrupting MRSA transmission is essential to reducing incidence and mortality rates.

Epigenetics of skin intrinsic florescence

Introduction & Objective: Cumulated advanced glycation end products (AGEs) in the bloodstream and tissues contribute to the pathogenesis of diabetes complications. The skin intrinsic fluorescence (SIF) is a non-invasive measurement of dermal AGEs level using spectrometer, and it can be used as a biomarker in AGEs-related diseases. Previously, specific epigenomic factor has been found to be associated with haemoglobin A1c (HbA1c). HbA1c is a type of glycated haemoglobin – the HbA1c test measures the average glycemic control over the period of 3 months. However, the effect of epigenetic factors on the level of AGEs in the skin remains unknown. We hypothesize that some cytosine-guanine dinucleotides (CpGs) are associated with SIF. An epigenome-wide associations study (EWAS) was performed to identify CpG sites associated with SIF in people with type 1 diabetes. Methods: 499 people with type 1 diabetes that have both methylation and SIF from the Diabetes Control Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) study were included. We fit linear regression models for SIF with each CpG site one at a time. The epigenome-wide significance level (p=5e-8) was applied. Then the result is compared with the null hypothesis where CpGs are not associated with SIF to check the inflation. In order to check the assumptions of the multiple linear models at a single CpG, we use diagnostic plots. Results: We did not identify a specific CpG that is significantly associated with neither skin intrinsic fluorescence 1 (SIF1) nor skin intrinsic fluorescence 12 (SIF 12).The CpG site with strongest effect is cg06538183 ([SE] -2.73 [0.61], p = 8.72e-6) on SIF1 and cg12871967 ([SE] 2.52, 0.53, p = 2.71e-6) on SIF12. Conclusion: We did not find any specific CpG that was significantly associated with either SIF 1 or SIF12. In general, the result suggests that DNA methylation does not impact the accumulation of AGEs in skin cells. DNA methylation data has a unique pattern of distribution that drives the non-uniform distribution of the p-values. The group of 275,301 CpGs that have means above the median and standard deviations below the median has the expected uniform p-value distribution.

Patient and Family Engagement

British Columbia Mental Health and Substance Use Services (BCMHSUS) provides mental health services, education, and health promotion initiatives to people with mental health and substance use issues across the province of BC. As a Project Coordinator in the Patient and Community Engagement portfolio, I performed a variety of work to support patient and family engagement under the newly created Patient Engagement Framework. Engaging patients and families as active participants and co-designers of their own care is an important component of patient-centred care that improves healthcare quality, health outcomes, and overall experiences of care at a system level. To work towards this goal, I developed a trauma-informed policy and procedure for BCMHSUS on patient and family engagement to serve as a guideline for giving patients and families a voice in the design and delivery of their mental health care. I also drafted two patient engagement playbooks called Managing Conflict and Respecting Emotions and Engaging Mandated and Incarcerated Patients, which focus on barriers and solutions to engaging patients in vulnerable circumstances. Moreover, I worked with provincial stakeholders to write the annual report for the BC Partners, which is a collaborative mental health promotion partnership between BCMHSUS and 7 provincial organizations with different mental health and substance use specialties (e.g. BC Schizophrenia Society, The Mood Disorders Association of BC, Canadian Institute for Substance Use Research, etc.). I also performed a literature review of the evidence supporting family engagement in patient- and family-centred care, and I made infographics and other visual designs to translate research and knowledge in visually appealing ways. Overall, my practicum helped me contribute towards advancing public mental health by valuing patients' knowledge, skills, and lived experience in the health system and working on a variety of initiatives to promote mental health in the province.

The effects of non-pharmaceutical interventions on SARS-CoV-2 transmission in different socioeconomic populations in Kuwait: A modelling study

Background: Aggressive non-pharmaceutical interventions (NPIs) may reduce transmission of SARS-CoV2. The extent to which these interventions are successful in stopping the spread have not been characterized in countries with distinct socioeconomic groups. We compared the effects of a partial lockdown on disease transmission among Kuwaitis (P1) and non-Kuwaitis (P2) living in Kuwait. Methods: We fit a metapopulation Susceptible-Exposed-Infectious-Recovered (SEIR) model to reported cases stratified by two groups to estimate the impact of a lockdown on the effective reproduction number (Re). We estimated the basic reproduction number (R0) for the transmission in each group and simulated the potential trajectories of an outbreak from the first recorded case of community transmission until 12 days after the lockdown. We estimated R­e values of both groups before and after the lockdown, simulated the effect of these values on epidemic curves and explored a range of cross-transmission scenarios. Results: We estimate R0 at 1·06 (95% CI: 1·05-1·28) for P1 and 1·83 (1·58-2·33) for P2. On March 22nd, Re for P1 and P2 are estimated at 1·13 (1·07-1·17) and 1·38 (1·25-1·63) respectively. After the curfew had taken effect, Re for P1 dropped modestly to 1·04 (1·02-1·06) but almost doubled for P2 to 2·47 (1·98-3·45). Our simulated epidemic trajectories show that the partial curfew measure modestly reduced and delayed the height of the peak in P1, yet significantly elevated and hastened the peak in P2. Modest cross-transmission from P2 to P1 elevated the height of the peak in P1 and brought it forward in time closer to the peak of P2. Conclusion: Our results demonstrate that a lockdown can reduce SARS-CoV2 transmission in one subpopulation but accelerate it in another. At the population level, the consequences of lockdowns may vary across the socioeconomic spectrum. Any public health intervention needs to be sensitive to disparities within populations.

Evaluating the Impact of Alcohol Control Interventions on Suicide Mortality Rates

Introduction & Objective: Given that the impact of regulatory and public policy initiatives cannot usually be tested through traditional randomized controlled trial designs, well-selected, -designed, and -analyzed natural experiments are the method of choice when examining the effects of such enactments on a variety of outcomes. The classic methodology for such evaluations is interrupted time-series (ITS) analysis, which is considered as one of the quasi-experimental designs that use both pre- and post-policy data without randomization. This study tests the impact of alcohol control interventions implemented in different period of times on suicide mortality rates among people 25-74 years of age using ITS. Methods: We mainly use the generalized additive mixed model (GAMM) to capture trend and seasonality in suicide mortality rates while controlling for unemployment rates, financial crisis during 2007-2008, and legal alcohol consumption records. Given the notable differences in alcohol consumption and suicide mortality between males and females, all analyses are conducted gender-specifically. Results: The ITS shows that the intervention introduced in 2017 has a significant effect on reducing suicide mortality rates for males between 25 and 74. Following the implementation of the intervention, suicide mortality rates decreased by 23.8% (95% CI: 10.2% - 35.4%) on average. Conclusion: The alcohol control intervention that strictly increased the excise tax on alcohol products has been shown to have a strong impact on reducing suicide mortality rates among male adults 25-74 years of age. ITS analyses are one of the strongest evaluative designs and allow a more detailed assessment of the longitudinal impact of an intervention than may be possible with a randomized control trial.

Developing Resources for Staff and Adapting Programing During COVID-19 at Fred Victor

Fred Victor is an organization that supports those experiencing poverty and homelessness in Toronto. As a practicum student in the Health Promotions department at Fred Victor, I gained experience working on health promotion projects and was able to work directly with the community. Throughout the practicum, I worked on several projects to adapt Fred Victor’s services during COVID-19. First, I worked to develop a resilience toolkit for Fred Victor staff. COVID-19 has led to higher levels of stress. This prompted Fred Victor to develop tools to support their staff. I designed a toolkit that instructs managers on how to promote resilience in their supervision sessions and team meetings. This toolkit provided information on what resilience is, as well as practical actions that managers can take to promote resilience in staff. This project involved knowledge translation to convey the research on resilience to Fred Victor staff in an accessible way. Additionally, I worked to support the development of online peer support groups. Typically, Fred Victor runs weekly in-person peer support groups for community members. However, due to COVID-19, these groups had to move to an online format. I helped facilitate this transition by developing a guide for facilitating online group programming. This guide included information on the best platforms to run online programming, how to create a safety agreement, and best practices for facilitating the group. I then conducted outreach to community members to ask for their input on the format and content of the groups. These projects are important to public health as they work to meet the public health goal to improve quality of life by promoting and encouraging healthy behaviours. These projects played an important role in promoting the health of Fred Victor staff and clients during COVID-19 by providing them with support and tools to manage their mental health.

The impact of ignoring Interval censoring in progression-free survival in cancer trials: a systematic review

Introduction & Objective: From statistical literature, the bias in treatment effect from ignoring interval censoring in Progression-free survival (PFS) is demonstrated. However, the impact on estimators caused by interval censoring is not carefully took account and investigated by researchers in practice. The objective of this study is to examine the impact of accounting for interval censoring in practice among RCTs used to support FDA approvals anti-cancer drugs between the years 2005 and 2019 that used PFS as an endpoint. Methods: In this systematic review, the differences of hazard ratios between two methods: considering and ignoring interval censoring, are visualized by Kaplan-Meier survival curves and estimated from a Cox proportional hazard model of 87 RCTs. With assumption that these differences and mean differences (bias) follow a normal distribution, limits of agreement of differences and confidence interval of bias are used to represent agreement of two methods. Results: Limits of agreement of difference range from -0.044 to 0.0615, while confidence intervals for the bias range from 0.0026 to 0.0145, which does not include zero, resulting in estimated treatment effect differs for two methods. Conclusion: In general, bias caused by interval censoring in treatment effect exists with large sample studies. Focusing on individual clinical trials, limits of agreement can provide more information for researchers to make decision on how to account for interval censoring.

Impact of COVID-19 on displaced populations and migrants around the world

My practicum placement was completed with the Dalla Lana School of Public Health Centre for Global Health. I have contributed to the work of a team of student and faculty members developing a review of the literature and environmental scan to explore the impact of the COVID-19 pandemic on migrant populations. I worked with colleagues to design and run a search strategy on the Medline (OVID) and Scopus bibliographic databases. The findings showed that crises including the COVID-19 pandemic act as magnifying lens and expose existing inequities within society as the impact of the pandemic is not equally felt by all population groups. Migrant populations are particularly impacted due to their intersectional identities that marginalize and disempower them and severely impact their health outcomes. Even though migration is the engine of the globalized economy and migrant workers make significant contribution to agricultural and economic prosperity, their precarious living conditions have worsened during the pandemic and they are being excluded from relief packages and income support. Furthermore, racism and xenophobia are fuelling hostility and prejudice towards migrants as governments are controlling the movement of migrants by closing their borders to asylum seekers and existing refugee camps are having outbreaks due to cramped and overcrowded living conditions and limited healthcare access. It is evident that migrant populations are very diverse groups that are facing unique challenges and thus, require distinct forms of protection particularly during this pandemic. The results of this work are currently being summarized in a manuscript that recognizes how determinants of health impact the health and well-being of migrants, the need to develop a road map for recovery using a health equity lens, and inform health policies. To eradicate COVID-19, it is imperative to leave no one behind including migrant populations and re-evaluate how inequities are addressed globally.

Exploring Frontline Healthcare Worker's Stress and Recovery Off-Shift during the COVID-19 Pandemic

I completed my practicum with 4YouandMe, a non-profit created to aid individuals who are interested in sharing health-related data using smartphones and other wearable devices so that they can better understand and navigate health conditions. The Stress and Recovery Study used the Oura ring and smartphones to track and understand the multidimensional components of stress and recovery off-shift in frontline healthcare workers during the current COVID-19 pandemic. My role in this study was actively working as a clinical research coordinator and digital participant engagement expert. This role consisted of calling participants and asking them about their overall study experience, details regarding their stress triggers, their home and work environments, and use of their Oura ring. I was responsible for maintaining contact with about 70 participants and creating contact logs after each phone call. The purpose of these phone calls is to provide support and encourage participant adherence to the study tasks. In addition to this primary role, I also completed an emerging COVID-19 hotspot map that was used in the recruitment process of the study. I outlined regions in the U.S that may become hotspots for COVID cases and may subsequently translate to a higher stressed group of healthcare workers in those areas. Additionally, I contributed to developing adherence tracking frameworks and other study materials used by team members. This study is contributing to the public health literature by using novel methodologies including digital approaches to understanding stress. Looking at digital stress responses and biometric data as signals to predict infection may inform other tools to aid in early detection. Finally, the study aims to determine whether resiliency factors and some social determinants of health modify stress and recovery.