Bias in machine learning applications to address non-communicable diseases at a population-level: a scoping review.

Background: Machine learning (ML) is increasingly used in population and public health to support epidemiological studies, surveillance, and evaluation. Our objective was to conduct a scoping review to identify studies that use ML in population health, with a focus on its use in non-communicable diseases (NCDs). We also examine potential algorithmic biases in model design, training, and implementation, as well as efforts to mitigate these biases.

Breaking Down Bias: A Methodological Primer on Identifying, Evaluating, and Mitigating Bias in Cardiovascular Research.

Systematic error, often referred to as bias is an inherent challenge in observational cardiovascular research, and has the potential to profoundly influence the design, conduct, and interpretation of study results. If not carefully considered and managed, bias can lead to spurious results, which can misinform clinical practice or public health initiatives and compromise patient outcomes. This methodological primer offers a concise introduction to the identification, evaluation, and mitigation of bias in observational cardiovascular research studies assessing the causal association of an exposure (or treatment) on an outcome.

Setting the balance of care for older adults at risk of hospitalization and delayed discharge: A mixed-methods research protocol.

Introduction: Delayed hospital discharge is a persistent care quality issue experienced across health systems worldwide and remains a priority area to be addressed in Canada. Often associated with a decrease in services while waiting to leave the hospital, delayed discharge from hospital can lead to increased frailty, physical and cognitive decline, and caregiver burnout. Optimizing availability of and timely access to community-based health and social care are avenues that could reduce initial admissions to the hospital and length of hospital stay, and facilitate hospital discharges.

Using a causal decomposition approach to estimate the contribution of employment to differences in mental health profiles between men and women.

Background: Mental health disorders are known to manifest differently in men and women, however our understanding of how gender interacts with mental health and well-being as a broader construct remains limited. Employment is a key determinant of mental health and there are historical differences in occupational roles among men and women that continue to influence working lives (Bonde, 2008; Cabezas-Rodríguez, Utzet, & Bacigalupe, 2021; Drolet, 2022; Gedikli, Miraglia, Connolly, Bryan, & Watson, 2023; Moyser, 2017; Niedhammer, Bertrais, & Witt, 2021; Stier & Yaish, 2014; Van der Doef & Maes, 1999). This study aims to explore differences in multidimensional mental health between men and women, and to quantify how these differences may change if women had the same employment characteristics as men.

High-cost users after sepsis: a population-based observational cohort study.

Background: High-cost users (HCU) represent important targets for health policy interventions. Sepsis is a life-threatening syndrome that is associated with high morbidity, mortality, and economic costs to the healthcare system. We sought to estimate the effect of sepsis on being a subsequent HCU.

An Organizational Case Study of Mental Models among Health System Leaders during Early-Stage Implementation of a Population Health Approach.

Purpose: As the COVID-19 pandemic recedes, the importance of population health has come into sharp focus, prompting many health systems to explore leveraging population health data (PHD) for operational planning. This approach requires that healthcare leaders embrace the dual priorities of maintaining excellence in patient care while promoting the overall health of populations. However, many leaders are new to population-based thinking, posing a threat to successful operationalization if mental models are not aligned.

Barriers and facilitators to designing, maintaining, and utilizing rare disease patient registries: a scoping review protocol.

Objective: The objectives of this review are to identify barriers/facilitators to designing, maintaining, and utilizing rare disease patient registries (RDPRs); determine whether and how these differ among patient partners, other knowledge users (KUs), and researchers; and chart definitions of rare diseases and RDPRs.

Introduction: RDPRs are vital to improving the understanding of the natural histories and predictors of outcomes for rare diseases, assessing interventions, and identifying potential participants for clinical trials. Currently, however, the functionality of RDPRs is not fully optimized.

Rising Prevalence of Gestational Diabetes Mellitus in Ontario: A Population-based Study.

Objectives: Gestational diabetes mellitus (GDM) is a common pregnancy complication. Studies have shown that the prevalence of GDM is rising worldwide. In this study we aimed to describe the prevalence of GDM in Ontario, Canada, between 2015 and 2021.

Quantifying the magnitude of the general contextual effect in a multilevel study of SARS-CoV-2 infection in Ontario, Canada: application of the median rate ratio in population health research.

Background: Regional variations in SARS-CoV-2 infection were observed in Canada and other countries. Studies have used multilevel analyses to examine how a context, such as a neighbourhood, can affect the SARS-CoV-2 infection rates of the people within it. However, few multilevel studies have quantified the magnitude of the general contextual effect (GCE) in SARS-CoV-2 infection rates and assessed how it may be associated with individual- and area-level characteristics.

Human factors methods in the design of digital decision support systems for population health: a scoping review.

Background: While Human Factors (HF) methods have been applied to the design of decision support systems (DSS) to aid clinical decision-making, the role of HF to improve decision-support for population health outcomes is less understood. We sought to comprehensively understand how HF methods have been used in designing digital population health DSS.

Materials And Methods: We searched English documents published in health sciences and engineering databases (Medline, Embase, PsychINFO, Scopus, Comendex, Inspec, IEEE Xplore) between January 1990 and September 2023 describing the development, validation or application of HF principles to decision support tools in population health.

Sense of belonging to community and avoidable hospitalization: a population-based cohort study of 456,415 Canadians.

A sense of belonging to a community is a dimension of subjective well-being that is of growing population health interest. We evaluated sex-stratified associations between community belonging and risk of avoidable hospitalization. Adult men and women from the Canadian Community Health Survey (2000-2014) were asked to rate their sense of community belonging (N = 456,415) and were also linked to acute inpatient hospitalizations to 31 March 2018.

Predictive Modelling of Diabetes Risk in Population Groups Defined by Socioeconomic and Lifestyle Factors in Canada: A Cross-Sectional Study.

Objectives: This study modelled diabetes risk for population groups in Canada defined by socioeconomic and lifestyle characteristics and investigated inequities in diabetes risk using a validated population risk prediction algorithm.

Methods: We defined population groups, informed by determinants of health frameworks. We applied the Diabetes Population Risk Tool (DPoRT) to 2017/2018 Canadian Community Health Survey data to predict 10-year diabetes risk and cases across population groups.

Integrating equity indicators for hospital reporting metrics.

Disparities in healthcare delivery and design are deeply-rooted within healthcare systems globally. Many researchers have developed methods to measure inequity; however, there currently exists no accepted measurement approach implemented consistently across health systems. We applied the model-based Relative Index of Inequality (RII) as a measure of inequity at one of Canada's largest health systems, Trillium Health Partners, across two service types: planned and outpatient.

Ascertainment and description of pedestrian and bicycling injuries and fatalities in Ontario from administrative health records 2003-2017: contributions of non-collision falls and crashes.

Introduction: Pedestrian and bicycling injuries may be less likely to be captured by traffic injury surveillance relying on police reports. Non-collision injuries, including pedestrian falls and single bicycle crashes, may be more likely than motor vehicle collisions to be missed. This study uses healthcare records to expand the ascertainment of active transportation injuries and evaluate their demographic and clinical features.

[Epidemiological signatures and surveillance bias in cancer].

Surveillance bias occurs when variations in cancer incidence are the result of changes in screening or diagnostic practices rather than increases in the true occurrence of cancer. This bias is linked to the issue of overdiagnosis and can be apprehended by looking at epidemiological signatures of cancer. We explain the concept of epidemiological signatures using the examples of melanoma and of lung and prostate cancer.

Antipsychotic Discontinuation and New Trazodone Use in Ontario Nursing Homes: Evidence of Medication Substitution.

Objectives: An unintended consequence of efforts to reduce antipsychotic medications in nursing homes is the increase in use of other psychotropic medications; however, evidence of substitution remains limited. Our objective was to measure individual-level prescribing patterns consistent with substitution of trazodone for antipsychotics.

Design: Retrospective cohort study.

The impact of different imputation methods on estimates and model performance: an example using a risk prediction model for premature mortality.

Objective: To compare how different imputation methods affect the estimates and performance of a prediction model for premature mortality.

Study Design And Setting: Sex-specific Weibull accelerated failure time survival models were run on four separate datasets using complete case, mode, single and multiple imputation to impute missing values. Six performance measures were compared to access predictive accuracy (Nagelkerke R, integrated brier score), discrimination (Harrell's c-index, discrimination slope) and calibration (calibration in the large, calibration slope).

The equitable impact of sugary drink taxation structures on sugary drink consumption among Canadians: a modelling study using the 2015 Canadian Community Health Survey-Nutrition.

Objective: Estimate the impact of 20 % flat-rate and tiered sugary drink tax structures on the consumption of sugary drinks, sugar-sweetened beverages and 100 % juice by age, sex and socio-economic position.

Design: We modelled the impact of price changes - for each tax structure - on the demand for sugary drinks by applying own- and cross-price elasticities to self-report sugary drink consumption measured using single-day 24-h dietary recalls from the cross-sectional, nationally representative 2015 Canadian Community Health Survey-Nutrition. For both 20 % flat-rate and tiered sugary drink tax scenarios, we used linear regression to estimate differences in mean energy intake and proportion of energy intake from sugary drinks by age, sex, education, food security and income.

Using unsupervised clustering approaches to identify common mental health profiles and associated mental health-care service-use patterns in Ontario, Canada.

Mental health is a complex, multidimensional concept that goes beyond clinical diagnoses, including psychological distress, life stress, and well-being. In this study, we aimed to use unsupervised clustering approaches to identify multidimensional mental health profiles that exist in the population, and their associated service-use patterns. The data source was the 2012 Canadian Community Health Survey-Mental Health, linked to administrative health-care data; all Ontario, Canada, adult respondents were included.

Exploring the use of experimental small area estimates to examine the relationship between individual-level and area-level community belonging and self-rated health.

Background: Small area estimation refers to statistical modelling procedures that leverage information or "borrow strength" from other sources or variables. This is done to enhance the reliability of estimates of characteristics or outcomes for areas that do not contain sufficient sample sizes to provide disaggregated estimates of adequate precision and reliability. There is growing interest in secondary research applications for small area estimates (SAEs).

External validation and application of the Diabetes Population Risk Tool (DPoRT) for prediction of type 2 diabetes onset in the US population.

Introduction: Characterizing diabetes risk in the population is important for population health assessment and diabetes prevention planning. We aimed to externally validate an existing 10-year population risk model for type 2 diabetes in the USA and model the population benefit of diabetes prevention approaches using population survey data.

Research Design And Methods: The Diabetes Population Risk Tool (DPoRT), originally derived and validated in Canada, was applied to an external validation cohort of 23 477 adults from the 2009 National Health Interview Survey (NHIS).