Suchergebnisse

IntroductionAvoidable hospitalizations refer to acute care use for conditions that should normally be managed inprimary care settings. Lower socioeconomic status that is often measured using area-based indicators(e.g. median household income) has been shown to increase risk of avoidable hospitalizations.However, both area- and individual-level socioeconomic status can contribute to hospitalization risk,but previous data limitations have prevented separate analyses. Further, the joint effect of individualand neighbourhood socioeconomic status has not been established in the Canadian population. Toaddress this, this study links individual-level household income and neighbourhood-level materialdeprivation data within a population-based Canadian cohort.
ObjectivesTo determine the individual and joint effect of individual-level household income and neighbourhood-level material deprivation on risk of hospitalization for a set of chronic ambulatory care sensitiveconditions using linked health survey, hospital discharge, and census-derived data.
MethodsA pooled cohort was created by linking sociodemographic and health information from eight cycles ofthe Canadian Community Health Survey (2000/2001 - 2010) to hospital discharge records and Cana-dian Marginalization Indices (2001, 2006) (N = 354,595). The primary outcome variable was riskof index hospitalization with a primary diagnosis of angina, asthma, congestive heart failure, chronicobstructive pulmonary disease, diabetes, epilepsy, or hypertension. The primary exposure variablewas joint individual-level national household income quintile and neighbourhood-level material de-privation quintile. Relative risk (RR) was estimated by constructing modified Poisson regressionmodels with robust error variance.
ResultsIn fully adjusted models with income and deprivation considered separately, individuals in the lowesthousehold income quintile and highest material deprivation quintile were at increased risk of hospi-talization (Income RR: 1.82 (95% CI 1.56-2.13) Deprivation RR: 1.67 (1.44-1.95)). When incomeand deprivation were jointly considered, those with low individual income living in high deprivationneighbourhoods were at greatest risk of hospitalization (RR 1.83 (95% CI 1.63 - 2.05)).
ConclusionsBoth individual income and neighbourhood deprivation separately and jointly increase risk of avoid-able hospitalizations. Additional research is needed to understand their mechanisms of action.However, both levels should be considered when designing effective policies and interventions toreduce avoidable hospitalizations.

IntroductionDepression has consistently been associated with an increased risk of diabetes in recent meta-analyses. However, depression is a highly heterogeneous construct and people with specific symptoms of depression, such as weight gain and increased sleep, may be at a higher risk of diabetes.
Objectives and ApproachThis work will compare incident diabetes in Ontario adults with recent depressive episodes that included symptoms of weight gain, weight loss, or no weight change and in those with no recent depressive episodes. Participants will be drawn from several waves of the Canadian Community Health Survey and the National Population Health Survey. Past 12-month depressive episodes and weight change during most recent or worst episodes was measured using the CIDI/CIDI short form. Time to incident diabetes will be ascertained through linkage with the Ontario Diabetes Database. Cox proportional hazards regression will assess diabetes incidence by depression and weight change characteristics.
ResultsThis study will include 106 084 Ontario adults who participated in the Canadian Community Health Survey (2000/2001, 2002, 2003, 2012) and the National Population Health Survey (1996). Follow-up time will range from 4 to 19 years (until March 2017). Study covariates will include demographic and lifestyle factors, comorbidities, and health care use and will be extracted from the surveys above and from administrative health data. The dataset for this study is currently being prepared by the Institute for Clinical Evaluative Sciences (ICES) and the findings of this analysis will be presented at this conference.
Conclusion/ImplicationsThe results of this work will provide insight into who, among those with depression, is at highest risk of new-onset diabetes. These results will be relevant to the development of both personalized and population-level diabetes screening and prevention strategies.

IntroductionGeographic disparities in health indicators, such as premature mortality, may reveal area-level weaknesses in health system performance. Monitoring geographic trends can therefore have powerful implications for system evaluation and planning. However, attempts to understand patterns of population health can be complicated by underlying regional differences in demographics and behaviours.
Objectives and ApproachThis study aimed to identify regional disparities in premature mortality (defined as death before age 75), and to investigate how fully these disparities can be explained by population-level characteristics. Ontario's 76 administrative Local Health Integration Network (LHIN) sub-regions, which vary in geographic and population size, were analyzed using linked population-level data from the Institute for Clinical Evaluative Sciences and Cancer Care Ontario.
Spatially structured, sex-stratified Bayesian hierarchical models were used to estimate standardized mortality ratios (SMRs) for each LHIN sub-region in the 2011-2015 period. Models were adjusted for key population-level demographic and behavioural risk factors.
ResultsLarge disparities in premature mortality presented at the sub-region level in males and females. Low premature mortality clustered around large, urban population centers in Ottawa and Toronto. Premature mortality was comparatively higher throughout the rest of the province, particularly in northern and southeast Ontario.
Higher prevalence of material deprivation, overweight and obesity, sedentary behaviour, and smoking were all significantly (α=0.05) associated with elevated premature mortality risk, while increased alcohol consumption and immigrant population were associated with decreased risk.
Adjusting for model covariates reduced variance of sub-region SMR estimates by 87% in males and 89% in females. Population-level characteristics thus explain a large proportion of geographic inequality in premature mortality. However, residual spatial variation suggests that systematic regional differences in premature mortality extend beyond population-level traits.
Conclusion/Implications
This study represents a novel application of small-area analytic techniques to Ontario mortality data, made possible by comprehensive linkage of vital statistics. The findings highlight the importance of population composition to geographic disparities in health. Future work should investigate the influence of system-level factors in areas with elevated premature mortality.

To prevent exponential spread of COVID-19, many governments restricted economic activity through lockdowns. We model these restrictions as shocks to productivity by sector and trace total equilibrium effects across the economy using techniques from production network economics. We combine this economic model with an epidemiological model of income shocks to long-term health. On both long-run health and economic grounds, it is better to keep upstream sectors such as transportation, manufacturing, and wholesale open than consumer-facing sectors such as retail and restaurants.

ObjectivesLocal neighbourhoods have great potential to foster community belonging which can improve health and well-being. In partnership with the City of Toronto, we prioritize and assess the importance of physical and social environmental attributes to community belonging. Using new data linkages, we contribute to Toronto's community safety and well-being plan, SafeTO.
ApproachWe leverage an individual-level record linkage of respondents from multiple cycles of the Canadian Community Health Survey (CCHS, 2000 to 2017), the Canadian Vital Statistics Death Database (CVSD), and the Discharge Abstract Database (DAD) from the Centre for Population Health Data at Statistics Canada. Environmental data sources include the Census, administrative data, and open-source data. Using postal code information from the CCHS, we connected each respondent with their neighbourhood's attributes, which include but are not limited to proximity to amenities (such as childcare, libraries and public transit), the Canadian Active Living Environment measure, green space, and air quality.
ResultsOf the 74,000 CCHS respondents from the Toronto census metropolitan area (representing a population of 4 to 5 million annually), an overall rate of 86% agreed to link and share their data, which renders an estimated linked sample of approximately 63,600 respondents. Across the entire linkage, 54% of respondents are linked to a record in the DAD between 1999/00 and 2017/18, and 10% are linked to a death record in the CVSD between 2000 and 2017. In consultation with municipal agencies and community stakeholders, we will prioritize the environmental attributes that are most relevant to community belonging and well-being, and comprehensively model the relationship between these attributes, community belonging and downstream health outcomes using multivariable regression and time-to-event models.
ConclusionThis on-going work contributes to the development of innovative approaches for using multi-sector data to inform decision making, as per the goals of SafeTO. Results from the analyses will be used to identify environmental attributes potentially important for community belonging and will support municipal city planning and resource allocation.

IntroductionAlthough RCTs remain the gold standard for generating clinical evidence, follow up of participants to study long-term effects is limited by cost and other logistical considerations. Linkage of participant information to routinely collected data potentially offers a cost-effective solution to achieving long-term follow-up of treatment effects
Objectives and ApproachThis scoping review aimed to identify RCTs that had been extended by record linkage, and characterize these in terms of nationality, numbers of trials, disease areas and outcomes, types of data, linkage modes and duration of follow-up. We followed published guidelines for the conduct of scoping reviews, with a registered protocol and comprehensive literature search. Criterion-based selection of studies and extraction of date were performed in duplicate. Descriptive statistics were used to summarise the characteristics of eligible studies.
ResultsOne hundred thirteen RCTs had been extended by record linkage. Fifty-six were conducted in Nordic countries, 26 in the USA and 24 in the UK. Types of linkage data used were: vital statistics 36, adimistrative data 31, cancer registry 28, special registries 13 and others 11. The literature spanned 45 years, but 66 (58%) were published between 2010 and 2016. Linkage methods were reported as: deterministic 33, probabilistic 16 and unspecified 64. In 44 studies researchers reported ethics approval for linkage; this was not obtained in 39 cases and was absent in 30 reports. The overall follow up times achieved by record linkage were: 1-4 y (6 studies), 5-9y (34), 10-19y (48), 20-29y (21), 30-39y(4) and over 50y (1).
Conclusion/ImplicationsAlthough we uncovered over 100 RCTs that were extended by record linkage this is tiny compared with the number of trials that have been undertaken. Linkage to routinely collected data seems to be a feasible but under-used approach to extending the follow-up of clinical trial participants for very long periods.

IntroductionHomicide is an important cause of death for older youth and adult Canadians; however, little is known about health care use prior to death among this population.
ObjectivesTo characterise health care use for mental health and addictions (MHA) and serious assault (herein referred to assault) one year prior to death among individuals who died by homicide in Ontario, Canada using linked mortality and health care utilisation data.
MethodsWe report rates of health care use for MHA and assault in the year prior to death among all individuals 16 years and older in Ontario, Canada, who died by homicide from April 2003 to December 2012 (N=1,541). Health care use for MHA included inpatient stays, emergency department (ED) visits and outpatient visits, whereas health care use for assault included only hospital-based care (ED visits and inpatient stays). Sociodemographic characteristics and health care utilisation were examined across homicide deaths, stratified by sex.
ResultsOverall, 28.5% and 5.9% of homicide victims sought MHA and assault care in the year prior to death, respectively. A greater proportion of females accessed care for MHA, whereas a greater proportion of males accessed assault-related health care. Males were more likely to be hospitalised following an ED visit for a MHA or assault related reason, in comparison to females. The most common reason for a MHA hospital visit was for substance-related disorders. We found an increase over time for hospital-based visits for assault prior to death, a trend that was not observed for MHA-related visits.
ConclusionsA large proportion of homicide victims interacted with the health care system for MHA or assault in the year prior to death. An increase in hospital-based visits for assault-related reasons prior to death was observed. These trends may offer insight into avenues for support and prevention for victims of homicide.

IntroductionData from population health surveys, administrative health records and environmental monitoring are increasingly being linked at the individual level. As these data become available to health researchers, there is an increasing need for methods which can make sense of large, noisy and heterogeneous data and can model complex relationships. Using these data, machine learning methods have the potential to produce population health risk algorithms with better performance than those developed with traditional statistical approaches.
Objectives and ApproachThe objective of this work is to explore the use of machine learning methods for the development, validation and implementation of predictive risk algorithms designed specifically for population health planning purposes. Algorithms to predict risk of dementia and avoidable hospitalizations are in development using the Canadian Community Health Survey, geographic sociodemographic information, administrative health care utilization data and vital statistics. Methods being explored include naïve Bayes, gradient boosting, support vector machines and neural networks.
ResultsRisk algorithms for population health should generally prioritize calibration over discrimination due to implications for resource allocation decisions. Approaches to minimize the risk of overfitting should be used and reweighting of unbalanced data avoided as it distorts the population-level nature of the data. It is important to be aware of propagating underlying bias in the data or exacerbating existing health inequities, which can be evaluated in part through assessment of calibration across relevant population subgroups. Approaches that consider multi-level data structures are needed to appropriately incorporate neighbourhood-level measures with individual-level information. To maximize population health impact and acceptability, model transparency and interpretability should be prioritized.
ConclusionThere is tremendous potential for machine learning approaches to leverage large volumes of linked population data to produce predictive risk algorithms that will inform population health decision-making. Future work will explore use of complex environmental remote sensing and built environment data.

IntroductionHypertension is leading risk factor for cardiovascular disease and mortality. Low socioeconomic position (e.g., income or high material deprivation) is an important risk factor for hypertension. However, there is limited evidence monitoring the extent to which socioeconomic inequalities in hypertension exist and are changing over time in Ontario.
Objectives and ApproachThe study objective was to estimate socioeconomic trends in prevalent hypertension by household income and material deprivation in Ontario from 2000 to 2012. A pooled cross-sectional study was conducted using data from 6 Canadian Community Health Surveys linked to the Discharge Abstract Database and Ontario Health Insurance Plan data (n=121,390 over 35 years, 54\% female). Relative-weighted Poisson regression models were used to estimate hypertension rates (adjusted for age, sex, ethnicity and immigration) across quintiles of equivalized household income and area-level material deprivation. Socioeconomic inequalities were estimated using the slope index of inequality (SII) and relative index of inequality (RII).
ResultsSocioeconomic inequalities in hypertension were observed across income quintiles on both absolute (SII: 1428 per 10,000, 95\%CI:1126,1730) and relative (RII:1.74, 95\%CI:1.53,1.94) scales in 2000, decreasing by 2012 (SII:297 per 10,000, 95%CI: -82,676; RII:1.19, 95%CI:0.93,1.45). A similar pattern was observed across material deprivation quintiles, however with smaller inequalities in 2000 (SII:595 per 10,000, 95%CI:306,884; RII:1.25, 95%CI:1.11,1.39) and 2012 (SII:389 per 10,000, 95%CI:17,761; RII:1.24, 95%CI:0.99,1.49).
Conclusion/ImplicationsSocioeconomic inequalities in hypertension were observed in Ontario, with decreasing trends between 2000 and 2012. Area-level material deprivation underestimated individual-level socioeconomic inequalities in hypertension.

IntroductionWe report on key performance indicators to highlight quality and variation in health care. Given Ontario's diverse geography, we have prioritized improving measurement across the rural-urban continuum. This will improve our ability to discern the impact of geography on health care and health status to inform planning and decision making.
Objectives and ApproachBuilding on previous work to advance measurement of equity in health care, we struck a technical working group of experts to review methods for stratifying health system performance data by geographic location in the Ontario context. These methods were applied to a set of key performance indicators. The working group's review of the results of this analysis will lead to recommendations for the best method to refine and standardize how geographic location is measured and stratified. This will improve our ability to discern the impact of geography on health system performance and health status for our suite of public-reporting products.
ResultsThe technical working group identified three methodologies for consideration that used linked postal code data: Population Centre (POPCTR), Statistical Area Classification (SAC) and a hybrid POPCTR/SAC methodology. These methods were tested against a set of key performance indicators across dimensions of quality including timeliness, effectiveness, population health and health outcomes. The results show that, in the health system performance dimensions of effectiveness and timeliness, as well as for a subset of health outcomes, there is variation in performance across the urban-rural continuum, though not always in a linear way. This may reflect differences in health care access, health risk factors, sociodemographic or socioeconomic characteristics across the urban-rural continuum. More definitive conclusions and recommendations will be available when the working group meets to review the results.
Conclusion/ImplicationsIdentifying a robust methodology for measuring performance across geographic locations will improve our ability to discern the impact of geography on health care including where geography may impact access and effectiveness of services as well as health outcomes. This information will enable better health system planning and decision-making.

IntroductionLow socioeconomic position (SEP) is an important risk factor for ischemic heart disease (IHD). Current surveillance methods use area-based SEP measures to monitor trends in socioeconomic inequalities in IHD. The extent to which these methods underestimate individual-level socioeconomic inequalities in IHD is unclear.
Objectives and ApproachThe study objective was to estimate socioeconomic trends in IHD by household income and material deprivation in Ontario from 2000 to 2012. A pooled cross-sectional study was conducted using data from 6 Canadian Community Health Survey (CCHS) cycles (2000-2012) linked to the Discharge Abstract Database (n= 119,529 over 35 years of age, 55% female). Relative-weighted Poisson regression models were used to estimate IHD prevalence rates (adjusted for age, sex, ethnicity and immigration) across quintiles of equivalized household income and area-level material deprivation. Socioeconomic inequalities were estimated using the slope index of inequality (SII) and relative index of inequality (RII).
ResultsSocioeconomic inequalities in IHD were observed across income and material deprivation quintiles. Measured using the SII, adjusted IHD rates were 345 per 10,000 (95%CI: 207,483) higher at the bottom of the income distribution than the top in 2000, decreasing to 167 per 10,000 (95%CI: 40,293) by 2012. These differences represented 2.52 (95%CI: 1.58,3.46) times higher IHD rates in 2000, an increased risk that remained in 2012 (RII: 1.80, 95%CI: 0.97,2.63). A similar pattern was observed across material deprivation quintiles, however with smaller absolute and relative inequalities observed in 2000 (SII:195 per 10,000, 95%CI:79,312; RII:1.64, 95%CI:1.16,2.11) and 2012 (SII:142 per 10,000, 95%CI:16,268; RII:1.54, 95%CI:0.94,2.14).
Conclusion/ImplicationsConsistent socioeconomic inequalities in IHD were observed in Ontario, with an absolute reduction between 2000 and 2012. Area-level material deprivation underestimated individual-level socioeconomic inequalities in IHD.

Funding This project was funded as an Applied Health Research Question (AHRQ), a process by which government-funded research organisations are funded to answer questions from relevant knowledge users. The Institute for Clinical Evaluative Sciences (ICES), which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care (MOHLTC), was funded to carry out this AHRQ research question on behalf of Public Health Ontario, Ontario's expert technical and scientific public health organisation. ; Background: Historically, women have lower all-cause mortality than men. It is less understood that sex differences have been converging, particularly among certain subgroups and causes. This has implications for public health and health system planning. Our objective was to analyse contemporary sex differences over a 20-year period. Methods: We analysed data from a population-based death registry, the Ontario Registrar's General Death file, which includes all deaths recorded in Canada's most populous province, from 1992 to 2012 (N=1 710 080 deaths). We calculated absolute and relative mortality sex differences for all-cause and cause-specific mortality, age-adjusted and age-specific, including the following causes: circulatory, cancers, respiratory and injuries. We used negative-binomial regression of mortality on socioeconomic status with direct age adjustment for the overall population. Results: In the 20-year period, age-adjusted mortality dropped 39.2% and 29.8%, respectively, among men and women. The age-adjusted male-to-female mortality ratio dropped 41.4%, falling from 1.47 to 1.28. From 2000 onwards, all-cause mortality rates of high-income men were lower than those seen among low-income women. Relative mortality declines were greater among men than women for cancer, respiratory and injury-related deaths. The absolute decline in circulatory deaths was greater among men, although relative deciles were similar to women. The largest absolute mortality gains were seen among men over the age of 85 years. Conclusions: The large decline in mortality sex ratios in a Canadian province with universal healthcare over two decades signals an important population shift. These narrowing trends varied according to cause of death and age. In addition, persistent social inequalities in mortality exist and differentially affect men and women. The observed change in sex ratios has implications for healthcare and social systems. ; Publisher PDF ; Peer reviewed

Background: People experiencing homelessness have diverse patterns of healthcare use. This study examined the distribution and determinants of healthcare encounters among adults with a history of homelessness. Methods: Administrative healthcare records were linked with survey data for a general cohort of adults with a history of homelessness and a cohort of homeless adults with mental illness. Binary and count models were used to identify factors associated with hospital admissions, emergency department visits and physician visits for comparison across the 2 cohorts. Results: During the 1-year follow-up period, a higher proportion of people in the cohort with a mental illness used any inpatient (27% vs 14%), emergency (63% vs 53%), or physician services (90% vs 76%) compared to the general homeless cohort. People from racialized groups were less likely use nearly all health services, most notably physician services. Other factors, such as reporting of a regular source of care, poor perceived general health, and diagnosed chronic conditions were associated with higher use of all health services except psychiatric inpatient care Conclusion: When implementing interventions for patients with the greatest health needs, we must consider the unique factors that contribute to higher healthcare use, as well as the barriers to healthcare access.