Search

Objective Develop a cohort to evaluate a real-world experiment of the built environment on the incidence of disease, accounting for movement of people, postal codes, incidence of disease and changing demographics. Approach The City of Winnipeg, Canada built multi-use trails in 2010-2012. We used the administrative data housed at the Manitoba Centre for Health Policy to evaluate the effect of this change in the built environment on the incidence of cardiovascular disease events (CVDE) and risk factors (CVDRF) before (2000-2009) and after (2012-2019) the trails were built. The Manitoba Health Insurance Registry contains postal code and demographic information on nearly all city residents and was utilized for cohort creation and person-time exposure. Individuals’ residential postal codes were linked to geo-spatial data to determine proximity to built trails at 400m, 800m and 1200m. Semi-annual postal code changes accounted for movement within/outside city limits or exclusion from the cohort. Diagnoses from hospital abstract and physician visit data and outpatient prescription dispensations were used to access prevalence and incidence of a CVDE composite measure of congestive heart failure, ischemic heart disease and stroke and a CVDRF composite measure of diabetes, dyslipidemia and hypertension. Conclusions and Implications Leveraging a diverse set of administrative databases, we built a cohort evaluate the effect of building multi-use trails in Winnipeg on the reduction on CVDE and CVDRF. This demonstrates how administrative data can be used to evaluate natural, real-world experiments with minimal direct data measurement or public intrusion, resulting in actionable results to inform public policy.

Objectives SPECTRUM is a unique multi-disciplinary, cross-sector collaborative research partnership between academics, government, and community organizations in Manitoba working together to address complex social problems. In our first research project, partners identified the need for quantitative evidence around outcomes of children involved with Child Protection Services (CPS), using linked administrative data. Approach From the SPECTRUM partnership, a research team with CPS expertise was established, including government policy-makers, community organizations representing First Nations families, and academics from multiple disciplines. The research is guided by an Advisory Circle of First Nations Knowledge Keepers. Linking health, education, CPS, and justice data we developed a matched cohort identifying children involved with CPS (2007-2018) for whom there was discretion in the decision to 1) place them in out-of-home care (n=19,718), or 2) keep them in their family home while providing services (n=28,154). Instrumental Variable analysis, with CPS agency as the instrument, will be used to compare outcomes. Results Following the trajectories of these two groups of CPS-involved children over time, we will compare their mental and physical health, educational achievement, and justice system involvement while accounting for individual-level (e.g., age, sex, chronic health conditions) and family-level (e.g., family income, maternal mental health, number of siblings) factors that may contribute to these outcomes. Preliminary findings will be workshopped with the SPECTRUM partnership to facilitate discussions on framing the evidence for policy makers. The SPECTRUM policy team will then prepare policy recommendations for government to consider. The Advisory Circle and a youth advisory squad will facilitate contextualizing and mobilizing findings. Actions taken by government in response to material provided will be monitored and will inform the development of subsequent research projects conducted by SPECTRUM. Conclusion Government and community stakeholder involvement throughout bolsters the likelihood of evidence translating into program and policy changes. This is not a situation where academics are telling government how to do their jobs – this is government, community organizations, and academics working together with the shared goal of better outcomes for children.

Background Under the National Housing Strategy, the Canadian government will make historic investments in housing over the next decade. The Canadian Mortgage and Housing Corporation is leading a research strategy to evaluate the impact of these investments. As part of this initiative, the Manitoba Center for Health Policy is conducting a pilot study to determine whether administrative data can be used to assess impacts, specifically looking at health, education and involvement in the justice system. Approach Using administrative data we tested for changes in healthcare use and justice involvement in the two years before and after a cohort of individuals moved into public housing. Additionally, to determine if changes in the outcomes over time were unique to public housing, we included a matched comparison group of individuals who did not reside in public housing. GLM with generalized estimating equations tested for differences over time and between cohorts in the number of hospitalizations, inpatient days, emergency department visits, and contacts with the criminal justice system. The data were modeled using a Poisson distribution (rate ratio, RR). Results Compared to the matched cohort, individuals accepted into public housing showed a significant decline in number of hospitalizations (pre RR=1.58 (1.53, 1.63), post RR=1.23 (1.19, 1.27), days in hospital (pre RR=1.66 (1.64, 1.68), post RR=1.24 (1.23, 1.26) and visits to the emergency department (pre RR=1.57 (1.52, 1.62), post RR=1.42 (1.38, 1.47). A trend towards fewer involvements with the criminal justice system was also observed (pre RR=1.37 (1.32, 1.43), post RR=1.28 (1.22, 1.34). No significant differences were noted for total respiratory morbidity or high school grades. Conclusion Administrative data show good potential to be used for the evaluation of public housing impacts on a wide range of health and social outcomes. Additional indicator comparisons will be reported at the conference.

Introduction On their 18th birthday children in custody of provincial Child and Family Services (CFS) age out, and are adults in control of their own care. An additional extended transitional services program was introduced several years ago to address gaps in the provicion of, and access to, adult social services during this change. Objectives and Approach Using linked population based data from the Manitoba Population Research Data Repository, children in the custody of CFS who turned 18 during a 10 year study period were compared to children not in custody. For those in custody of CFS, we also compared individuals who participated in the extended transitional care services to those who opted out. Outcomes included use of health services and prescription drugs, social assistance, involvment with the justice system, living in social housing, and mental health outcomes. For most outcomes, the two year period prior to the 18th birthday and the two year period after were measured. Results During the study period, 4656 children in care of CFS turned 18 while in custody. There were 2811 permanent wards, of which 1663 participated in the extended transitional services program. An additional 1845 non-permanent wards also turned 18 during the study period. Permanent wards were much more likely to be long term wards (greater than six years, ~65\%) compared non permanent wards (~17\%). Opioid prescription rates more than doubled in the two years after their 18th birthday and were about 6 times greater than prescription rates for those not in care of CFS. Criminal accusation rates did not change after their 18th birthday, were about equal for permanent and non-permanent wards. For the majority of outcomes, the transitional services program appeared to have little impact. Conclusion/Implications Compared to children not in care of CFS, rates of most outcomes were considerably higher for wards. Not all outcomes demonstrated a significant change over the transition period. By linking data from so many different government departments, extra attention can be focused on areas likely to have the greatest impact.

Introduction Alcohol use disorders (AUDs) – mental and/or physical health diagnoses due to harmful alcohol consumption – are associated with compromised quality of life for the individual. Over the past two decades, pharmacotherapies have been developed to treat alcohol addiction and may help mitigate the harmful outcomes linked with excessive alcohol. Objectives and Approach The objectives were to examine the pharmacotherapy dispensation patterns among individuals with an AUD and their mental health comorbidities. We used ICD codes from medical claims and hospital discharge data to identify anyone with a physical / mental health diagnosis due to harmful alcohol consumption – AUD, April 1, 1996-March 31, 2015. We identified mental health comorbidities using administrative health records. Drug dispensation data were used to identify all first-time prescriptions for acamprosate, naltrexone, or disulfiram occurring after an initial AUD diagnosis. Generalized linear models tested for predictors of receiving a prescription and to identify differences in mental health comorbidities. Results We identified 53,556 treatment niave individuals with an AUD who were eligible to receive one of these three prescriptions. 493 of these received a prescription for acamprosate, naltrexone, or disulfiram. The majority of prescriptions came from general practitioners from urban centers. Those with a prescription were significantly more likely to have a comorbid mood or anxiety diagnosis. Those with a prescription were more likely to have a physician visit for a mental health issue a year to two years before diagnosis compared with those who did not have a prescription; and, they were more likely to be dispensed a selective serotonin, a sedative, and an anti-anxiety medication prior to receiving an AUD diagnosis. Conclusion/Implications Drug therapies to aid in the recovery from AUD are being underutilized. Diagnosis of and treatment for mental health disorders is more common among those dispensed these medications. Programs that study clinicians’ use of AUD-targeted drug therapies should be considered, while psychiatric services in addiction care require significant improvement.

ABSTRACT Objective The objective of this population health research is to identify the social burden associated with having an alcohol related diagnosis. Methods We used linkable population-based administrative data files held in the Population Health Research Data Repository to conduct our research, fiscal years 1990/91 to 2014/15. Data came from several domains including health, social services, justice, and the Canadian Census. We used ICD-9/10-CA codes from the hospital abstract database, medical claims data, and prescription drug data to identify individuals with an alcohol-related diagnosis. Individuals’ socioeconomic status was determined using neighbourhood-level income data from the Canadian Census. We matched, 3:1, diagnosed cases to individuals in our repository using on age, sex, income, and community area. We linked cases and matches to administrative data from justice and social services to identify social outcomes associated with having an alcohol-related diagnosis. Outcomes included receipt of income assistance, residence in publically funded social housing, having a child apprehended by child and family protective services, having a charge for driving under the influence recorded in the justice data, and having a charge for domestic violence recorded in the justice data. We modelled rates using generalized estimating equations from 5 years before date of diagnosis to a maximum of 20 years after date of diagnosis. Models tested for significant differences in rates between cases and matches both before and after diagnosis; as well, we tested for time trends in rates both before and after diagnosis. Results We identified 52,991 individuals with an alcohol related diagnosis between 1990/91 and 2014/15: 34,145 males and 18,846 females. 80.3% of cases had a mental-health related alcohol diagnosis. Diagnoses followed a socioeconomic gradient with the greatest number of cases coming from low-income neighbourhoods. Cases had a significant spike in rates from one year before to one year after diagnosis date, compared with matches, across all indicators. When we followed individuals for 20 years after diagnosis, we found a significantly elevated rate of social service use and involvement with the justice system, across all outcomes, for all years. Conclusion Receiving an alcohol-related diagnosis is associated with subsequent increased use of social services and contacts with the justice system. Upstream efforts to reduce alcohol-related diagnoses may result in reduced use of social services and justice contacts.

ABSTRACT Objective Fourteen years of social housing data (1995-2008) were acquired from the provincial government. This allowed for an unprecedented opportunity to describe the population of individuals living in social housing and, through data linkage, to compare them to the rest of the province on a number of health and social indicators. Approach Using data from the entire population of the province of Manitoba, Canada, cross-sectional comparison were made between those living in social housing and those not on 19 indicators of morbidity, mortality, health care utilization and social development. Regression models were developed to control for age, sex, region of residence, comorbidities, income and neighborhood level SES. Results 50% of the population in social housing are under the age 20, 75% are female and 50% of applicants receive some form of income assistance. As expected there are significant differences on most health status measures when compared to individuals not in social housing. However, after controlling for confounding factors most differences between the two groups disappear indicating that there is no independent effect of living in social housing. A few exceptions were noted on measures of total respiratory morbidity, mammography and high school completion rates, the later showing a very significant interaction with neighborhood level SES. Conclusion Despite overall poor health status, after controlling for income and other confounding factors individuals in social housing score no worse on many measures of health care utilization and prevention. High school completion rates, in particular, showed a very strong interaction with neighborhood level SES. Policy implications of this research are discussed.