Predictive risk modelling of high resource users under different prescription drug coverage policies in Ontario and Manitoba, Canada.

INTRODUCTION: Studying high resource users (HRUs) across jurisdictions is a challenge due to variation in data availability and health services coverage. In Canada, coverage for pharmaceuticals varies across provinces under a mix of public and private plans, which has implications for ascertaining HRUs. We examined sociodemographic and behavioural predictors of HRUs in the presence of different prescription drug coverages in the provinces of Manitoba and Ontario. METHODS: Linked Canadian Community Health Surveys were used to create two cohorts of respondents from Ontario (n = 58,617, cycles 2005-2008) and Manitoba (n = 10,504, cycles 2007-2010). HRUs (top 5%) were identified by calculating health care utilization 5 years following interview date and computing all costs in the linked administrative databases, with three approaches used to include drug costs: (1) costs paid for by the provincial payer under age-based coverage; (2) costs paid for by the provincial payer under income-based coverage; (3) total costs regardless of the payer (publicly insured, privately insured, and out-of-pocket). Logistic regression estimated the association between sociodemographic, health, and behavioral predictors on HRU risk. RESULTS: The strength of the association between age (≥ 80 vs. <30) and becoming an HRU were attenuated with the inclusion of broader drug data (age based: OR 37.29, CI: 30.08-46.24; income based: OR 27.34, CI: 18.53-40.33; all drug payees: OR 29.08, CI: 19.64-43.08). With broader drug coverage, the association between heavy smokers vs. non-smokers on odds of becoming an HRU strengthened (age based: OR 1.58, CI: 1.32-1.90; income based: OR 2.97, CI: 2.18-4.05; all drug payees: OR 3.12, CI: 2.29-4.25). Across the different drug coverage policies, there was persistence in higher odds of becoming an HRU in low income households vs. high income households and in those with a reported chronic condition vs. no chronic conditions. CONCLUSIONS: The study illustrates that jurisdictional differences in how HRUs are ascertained based on drug coverage policies can influence the relative importance of some behavioural risk factors on HRU status, but most observed associations with health and sociodemographic risk factors were persistent, demonstrating that predictive risk modelling of HRUs can occur effectively across jurisdictions, even with some differences in public drug coverage policies.

Predicting High Health Care Resource Utilization in a Single-payer Public Health Care System: Development and Validation of the High Resource User Population Risk Tool.

BACKGROUND: A large proportion of health care spending is incurred by a small proportion of the population. Population-based health planning tools that consider both the clinical and upstream determinants of high resource users (HRU) of the health system are lacking. OBJECTIVE: To develop and validate the High Resource User Population Risk Tool (HRUPoRT), a predictive model of adults that will become the top 5% of health care users over a 5-year period, based on self-reported clinical, sociodemographic, and health behavioral predictors in population survey data. RESEARCH DESIGN: The HRUPoRT model was developed in a prospective cohort design using the combined 2005 and 2007/2008 Canadian Community Health Surveys (CCHS) (N=58,617), and validated using the external 2009/2010 CCHS cohort (N=28,721). Health care utilization for each of the 5 years following CCHS interview date were determined by applying a person-centered costing algorithm to the linked health administrative databases. Discrimination and calibration of the model were assessed using c-statistic and Hosmer-Lemeshow (HL) χ statistic. RESULTS: The best prediction model for 5-year transition to HRU status included 12 predictors and had good discrimination (c-statistic=0.8213) and calibration (HL χ=18.71) in the development cohort. The model performed similarly in the validation cohort (c-statistic=0.8171; HL χ=19.95). The strongest predictors in the HRUPoRT model were age, perceived general health, and body mass index. CONCLUSIONS: HRUPoRT can accurately project the proportion of individuals in the population that will become a HRU over 5 years. HRUPoRT can be applied to inform health resource planning and prevention strategies at the community level.

Reassessing access to intensive care using an estimate of the population incidence of critical illness.

BACKGROUND: The consistently observed male predominance of patients in intensive care units (ICUs) has raised concerns about gender-based disparities in ICU access. Comparing rates of ICU admission requires choosing a normalizing factor (denominator), and the denominator usually used to compare such rates between subpopulations is the size of those subpopulations. However, the appropriate denominator is the number of people whose medical condition warranted ICU care. We devised an estimate of the number of critically ill people in the general population, and used it to compare rates of ICU admission by gender and income. METHODS: This population-based, retrospective analysis included all adults in the Canadian province of Manitoba, 2004-2015. We created an estimate for the number of critically ill people who warrant ICU care, and used it as the denominator to generate critical illness-normalized rates of ICU admission. These were compared to the usual population-normalized rates of ICU care. RESULTS: Men outnumbered women in ICUs for all age groups; population-normalized male:female rate ratios significantly exceed 0 for every age group, ranging from 1.15 to 2.10. Using critical-illness normalized rates, this male predominance largely disappeared; critically ill men and women aged 45-74 years were admitted in equivalent proportions (critical-illness normalized rate ratios 0.96-1.01). While population-normalized rates of ICU care were higher in lower income strata (p < 0.001), the gradient for critical illness-based rates was reversed (p < 0.001). CONCLUSIONS: Across a 30-year adult age span, the male predominance of ICU patients was accounted for by higher estimated rates of critical illness among men. People in lower income strata had lower critical-illness normalized rates of ICU admission. Our methods highlight that correct inferences about access to healthcare require calculating rates using denominators appropriate for this purpose.

Is the change in body mass index among children newly diagnosed with type 1 diabetes mellitus associated with obesity at transition from pediatric to adult care?

BACKGROUND: Insulin therapy is lifesaving treatment for individuals with type 1 diabetes (T1D). Its initiation maybe associated with significant weight gain because of change from a catabolic to an anabolic state. Excessive weight-gain increases the risk of obesity and is associated with chronic disease. OBJECTIVE: To examine if change in body mass index (BMI) among children in the 6 months after diagnosis with type 1 diabetes mellitus is associated with long-term obesity. METHODS: This was a population-based retrospective study of 377 children (aged 2-18 yr) with type 1 diabetes. Measured heights and weights were used to calculate BMI z-scores based on Centers for Disease Control and Prevention (CDC) cut-points. Generalized Linear Models using BMI group, and age group at diagnosis; postdiagnosis weight change; and sex were applied to assess associations between postdiagnosis weight change and BMI z-score at transition to adult care. RESULTS: Mean BMI z-score increased from 0.28 at diagnosis, to 0.53 at 6 months and 0.66 at transition to adult care. Change in BMI z-scores differed by initial BMI group and magnitude of postdiagnosis weight change. Younger children (<11 yr) had higher (p = 0.004) BMI z-scores at diagnosis but not at last visit (p = 0.1) than older (≥11 yr) children at diagnosis. BMI z-score at diagnosis, postdiagnosis weight change, female sex, and longer duration with TID were associated with higher BMI z-score at time of transition. CONCLUSIONS: BMI z-score at diagnosis was the strongest predictor of BMI z-score at time of transition to adult care, however; its effect was mediated by magnitude of weight change 6 months after diagnosis, sex, and age group at diagnosis.

Increased incidence of critical illness among patients with inflammatory bowel disease: a population-based study.

BACKGROUND & AIMS: Little is known about how often, and for what reasons, patients with inflammatory bowel diseases (IBD) are admitted to the intensive care unit (ICU). We compared incidences of ICU admission, characteristics of critical illness, and mortality after ICU admission between patients with IBD and the general population. METHODS: We identified all persons with IBD in the province of Manitoba using a validated administrative definition of IBD for the period from 1984 to 2010. Cases were considered incident for IBD if their first health system contact for IBD was in 1989 or later. We identified a population-based control group, matched by age, sex, and geography (based on postal code). Case and control cohorts were linked to the Manitoba ICU database. We compared outcomes between groups using age- and sex-standardized rates, Cox proportional hazards models, and logistic regression models, adjusting for age, sex, comorbidity, and socioeconomic status. RESULTS: There were 8224 prevalent and 4580 incident cases of IBD. After adjustment, the risk for ICU admission was higher for patients with IBD than controls (hazard ratio [HR], 1.79; 95% confidence interval [CI], 1.58-2.02). The risk of ICU admission was higher for patients with Crohn's disease (HR, 2.31; 95% CI, 1.95-2.75) than ulcerative colitis (HR, 1.37; 95% CI, 1.13-1.65). From 2000 through 2010, age- and sex-standardized annual incidence rates for ICU admission in the prevalent IBD cohort ranged from 0.55% to 1.12%. Compared with controls admitted to ICUs, 1 year after ICU admission, mortality was 32% among patients with IBD. CONCLUSIONS: Patients with IBD have a higher risk for admission to the ICU than the general population, and increased mortality 1 year after admission. These findings underscore the potential severity of IBD.

Epidemiology of critically ill patients in intensive care units: a population-based observational study.

INTRODUCTION: Epidemiologic assessment of critically ill people in Intensive Care Units (ICUs) is needed to ensure the health care system can meet current and future needs. However, few such studies have been published. METHODS: Population-based analysis of all adult ICU care in the Canadian province of Manitoba, 1999 to 2007, using administrative data. We calculated age-adjusted rates and trends of ICU care, overall and subdivided by age, sex and income. RESULTS: In 2007, Manitoba had a population of 1.2 million, 118 ICU beds in 21 ICUs, for 9.8 beds per 100,000 population. Approximately 0.72% of men and 0.47% of women were admitted to ICUs yearly. The age-adjusted, male:female rate ratio was 1.75 (95% CI 1.64 to 1.88). Mean age was 64.5 ± 16.4 years. Rates rose rapidly after age 40, peaked at age 75 to 80, and declined for the oldest age groups. Rates were higher among residents of lower income areas, for example declining from 7.9 to 4.4 per 100,000 population from the poorest to the wealthiest income quintiles (p <0.0001). Rates of ICU admission slowly declined over time, while cumulative yearly ICU bed-days slowly rose; changes were age-dependent, with faster declines in admission rates with older age. There was a high rate of recidivism; 16% of ICU patients had received ICU care previously. CONCLUSIONS: These temporal trends in ICU admission rates and cumulative bed-days used have significant implications for health system planning. The differences by age, sex and socioeconomic status, and the high rate of recidivism require further research to clarify their causes, and to devise strategies for reducing critical illness in high-risk groups.

Social gradients in surgical sterilization rates: opposing patterns for males and females.

OBJECTIVE: The purpose of this study was to analyze population-based rates of surgical sterilization by sex, income, age group, and location of residence (urban, rural, or remote) among a universally insured population. METHODS: The study included all 1.2 million residents of Manitoba. Data from comprehensive hospital and medical visit records were used to calculate rates of vasectomy and tubal ligation among the population aged 20 to 55 years. Analyses were carried out with generalized linear modelling, using administrative data housed at the Manitoba Centre for Health Policy. RESULTS: There were significant income-related gradients in sterilization rates for both sexes (P < 0.001), but the trends were in opposite directions for males and females. Tubal ligation rates were highest in low-income areas, and decreased with income, whereas vasectomy rates were lowest in low-income areas and increased with income. Rates also varied considerably by age and area of residence (urban, rural, or remote). CONCLUSION: These results are potentially worrying because tubal ligation is more invasive, associated with higher risks, and more costly than vasectomy. It would therefore be expected to be less common than vasectomy among all groups, not just those in higher income areas. The results contribute unique population-based procedure rates, which are not subject to biases that can affect survey-based studies. The findings may reflect underlying differences across income groups in decision-making regarding reproductive health issues.

Objectif : Cette étude avait pour but d'analyser les taux en population générale de stérilisation chirurgicale en fonction du sexe, du revenu, du groupe d'âge et du lieu de résidence (urbain, rural ou éloigné) au sein d'une population bénéficiant d'une assurance maladie universelle. Méthodes : L'étude s'est penchée sur les 1,2 million de résidents du Manitoba. Des données issues de dossiers exhaustifs de consultation hospitalière et médicale ont été utilisées pour calculer les taux de vasectomie et de ligature des trompes chez les membres de la population dont l'âge se situait entre 20 et 55 ans. Des analyses ont été menées au moyen d'une modélisation linéaire généralisée, en utilisant les données administratives du Manitoba Centre for Health Policy. Résultats : Les taux de stérilisation chez les deux sexes (P < 0,001) présentaient des gradients significatifs liés au revenu; toutefois, les tendances étaient opposées chez les hommes et les femmes. Les taux de ligature des trompes atteignaient leur apogée dans les régions à faible revenu et étaient inversement proportionnels au revenu, tandis que les taux de vasectomie atteignaient leur plus bas niveau dans les régions à faible revenu et étaient proportionnels au revenu. Ces taux variaient également de façon considérable en fonction de l'âge et du lieu de résidence (urbain, rural ou éloigné). Conclusion : Ces résultats sont potentiellement troublants, en raison du fait que la ligature des trompes est une intervention plus effractive, associée à plus de risques et plus coûteuse que la vasectomie. On serait donc en droit de s'attendre à ce qu'elle soit moins courante que la vasectomie dans tous les groupes, et non seulement dans les groupes des régions à revenu élevé. Ces résultats donnent des taux d'intervention en population générale uniques, lesquels ne sont pas assujettis aux biais qui peuvent affecter les études fondées sur des sondages. Nos constatations pourraient refléter des différences sous-jacentes, d'un groupe de revenu à l'autre, en ce qui concerne la prise de décision en matière de questions de santé génésique.

Switching to second line MS disease-modifying therapies is associated with decreased relapse rate.

Background and objectives: While randomized, controlled trials (RCTs) are the gold standard for determining treatment efficacy, they do not capture the effectiveness of treatment during real-world use. We aimed to evaluate the association between demographics and multiple sclerosis (MS) disease-modifying therapy (DMT) exposure, including treatment adherence and switches between different DMTs, on the risk of subsequent MS relapse. Methods: All persons with relapsing-onset MS (pwRMS) living in Manitoba between 1999 and 2014 were identified from provincial healthcare databases using a validated case definition. Use of DMTs was abstracted from the provincial drug database covering all residents of Manitoba, including use of any DMT, stopping/starting any DMT, switches between different DMTs and adherence as defined by cumulative medication possession ratios (CUMMPRs) of 50, 70, 80 and 90%. Time to first-treated relapse was used as the outcome of interest in logistic regression and Cox-proportional hazards regression models adjusting for demographic covariates including age and year of diagnosis, sex, socioeconomic status and number of medical comorbidities. Results: 1780 pwRMS were identified, including 1,510 who were on DMT at some point in the study period. While total DMT exposure was not associated with the time to subsequent treated relapse, individuals who switched between more than 2 DMTs had higher post-switch rates of relapse. Switching to second-line DMTs was associated with a longer time to treated relapse in comparison to those who remained on a first-line DMT (HR 0.44; 95%CI: 0.32-0.62, p < 0.0001). Discussion: Switching to high-efficacy DMTs reduces the rates of subsequent MS relapse at the population level.

Development and Validation of the Chronic Disease Population Risk Tool (CDPoRT) to Predict Incidence of Adult Chronic Disease.

Importance: Predicting chronic disease incidence for the population provides a comprehensive picture to health policy makers of their jurisdictions' overall future chronic disease burden. However, no population-based risk algorithm exists for estimating the risk of first major chronic disease. Objective: To develop and validate the Chronic Disease Population Risk Tool (CDPoRT), a population risk algorithm that predicts the 10-year incidence of the first major chronic disease in the adult population. Design, Setting, and Participants: In this cohort study, CDPoRT was developed and validated with 6 cycles of the Canadian Community Health Survey, linked to administrative data from January 2000 to December 2014. Development and internal validation (bootstrap and split sample) of CDPoRT occurred in Ontario, Canada, from June 2018 to April 2019 followed by external validation in Manitoba from May 2019 to July 2019. The study cohorts included 133 991 adults (≥20 years) representative of the Ontario and Manitoba populations who did not have a history of major chronic disease. Exposures: Predictors were routinely collected risk factors from the Canadian Community Health Survey, such as sociodemographic factors (eg, age), modifiable lifestyle risk factors (ie, alcohol consumption, cigarette smoking, unhealthy diet, and physical inactivity), and other health-related factors (eg, body mass index). Main Outcomes and Measures: Six major chronic diseases were considered, as follows: congestive heart failure, chronic obstructive pulmonary disease, diabetes, myocardial infarction, lung cancer, and stroke. Sex-specific CDPoRT algorithms were developed with a Weibull model. Model performance was evaluated with measures of overall predictive performance (eg, Brier score), discrimination (eg, Harrell C index), and calibration (eg, calibration curves). Results: The Ontario cohort (n = 118 747) was younger (mean [SD] age, 45.6 [16.1] vs 46.3 [16.4] years), had more immigrants (23 808 [20.0%] vs 1417 [10.7%]), and had a lower mean (SD) body mass index (26.9 [5.1] vs 27.7 [5.4]) than the Manitoba cohort (n = 13 244). During development, the full and parsimonious CDPoRT models had similar Brier scores (women, 0.087; men, 0.091), Harrell C index values (women, 0.779; men, 0.783), and calibration curves. A simple version consisting of cigarette smoking, age, and body mass index performed slightly worse than the other versions (eg, Brier score for women, 0.088; for men, 0.092). Internal validation showed consistent performance across models, and CDPoRT performed well during external validation. For example, the female parsimonious version had C index values for bootstrap, split sample, and external validation of 0.778, 0.776, and 0.752, respectively. Conclusions and Relevance: In this study, CDPoRT provided accurate, population-based risk estimates for the first major chronic disease.

How health status affects progress and performance in school: a population-based study.

OBJECTIVE: To assess the effects of health status at birth and health status in the preschool years on educational outcomes to age 9 in a population-based birth cohort. METHODS: Administrative data were used to follow all children born to Winnipeg mothers in 1990, and remaining in Manitoba until September 2004 (N = 5,873). A structural equation model was used, incorporating latent variables to represent Health Status at Birth, Major Illness and Minor Illness during the preschool years. The model also included the child's sex and exact age, along with a number of social, economic, and demographic characteristics of the child's family. The outcome was a combination of marks on Grade 3 Standards Tests and enrollment in the appropriate grade for age. RESULTS: Major Illness in the preschool years had a significant influence on progress and performance in school (p = 0.0003), predicting 1.26% of the variation in the outcome. Minor Illness was weaker but still significant (p < 0.01). Health Status at Birth was not directly related to the outcome; its effect was mediated by Major and Minor Illness in childhood. Overall, the strongest predictors were the child's age and the area-level income, followed by the mother's age, family receipt of income assistance, the sex of the child, breastfeeding initiation (all p < 0.0001), and Major Illness. CONCLUSIONS: Health status plays a statistically significant but substantively small role in explaining progress and performance in school among a population-based cohort. Major Illness was more important than Minor Illness, and these two factors completely mediated the influence of Health Status at Birth on the outcome. The strength of the social, economic, and demographic variables underscores the importance of the broader factors that affect both health and educational outcomes.

Validation of an algorithm to detect severe MS relapses in administrative health databases.

BACKGROUND: Severe relapses that required treatment were important outcomes in the sentinel trials of disease-modifying therapy (DMT). Identifying such relapses in administrative data would allow comparative-effectiveness studies of DMTs to be conducted in real-world clinical settings. METHODS: All relapsing-remitting (RRMS) and secondary-progressive (SPMS) patients living in Manitoba between 1999 and 2015 were identified using a validated case definition and linkage to the Manitoba MS Clinic database. All healthcare interactions potentially due to relapses were extracted from population-based administrative (hospital, physician claims and prescription) databases. These "relapse markers" included varying thresholds of outpatient prednisone scripts, day hospital or emergency room (ER) codes for intravenous (IV) methylprednisolone therapy, family physician, neurologist or ER physician billing codes and hospital admissions due to MS. Algorithms using combinations of these markers were compared with a reference standard of neurologist-defined relapses. The most useful algorithms were also examined on a biannual basis over the study period to assess for trends in the sensitivity of relapse detection. RESULTS: 1131 participants with RRMS or SPMS were linked to administrative databases. Analysis of potential relapse markers over the whole 1999-2015 time period was limited by inconsistent coding of same day or ER admissions for IV methylprednisolone administration. Widespread adoption of high-dose oral outpatient prednisone for relapses since 2009 resulted in a progressive improvement in relapse marker sensitivity. The best algorithm consisted of oral prednisone prescriptions >50mg/day for 3-60 days and same day hospital or ER assessment codes with MS as the most responsible diagnosis (sensitivity 70%, specificity 100%, PPV 100%, NPV 96%, kappa 0.8 in 2013-2015). CONCLUSIONS: Severe relapses can be identified from administrative datasets with reasonable accuracy. The trend since 2009 toward outpatient relapse treatment will improve the sensitivity of relapse detection with longitudinal follow-up of this cohort and will allow comparison of severe relapse rates between different DMTs, supporting future comparative effectiveness studies.

Not as skinny as we used to think: Body mass index in children and adolescents at diagnosis of type 1 diabetes mellitus.

This retrospective analysis of clinical data for children (2-18 years old) with incident T1D found surprisingly low (9%) prevalence of underweight, and high (15% overweight; 8% obesity) respectively at diagnosis. These results suggests a need to rethink the classic clinical teaching surrounding skinnier presentation at diagnosis, and importantly, the corresponding expectation of weight gain after insulin therapy initiation.

A population-based observational study of intensive care unit-related outcomes. With emphasis on post-hospital outcomes.

RATIONALE: Many studies of critical illness outcomes have been restricted to short-term outcomes, selected diagnoses, and patients in one or a few intensive care units (ICUs). OBJECTIVES: Evaluate a range of relevant outcomes in a population-based cohort of patients admitted to ICUs. METHODS: Among all adult residents of the Canadian province of Manitoba admitted to ICUs over a 9-year period, we assessed ICU, hospital, 30-day, and 180-day mortality rates; ICU and hospital lengths-of-stay; Post-hospital use of hospital care, ICU care, outpatient physician care, medications, and home care; and Post-hospital residence location. We explored data stratified by age, sex, and separate categories of geocoded income for urban and rural residents. For Post-hospital use variables we compared ICU patients with those admitted to hospitals without the need for ICU care. MEASUREMENTS AND MAIN RESULTS: After ICU admission there was a high initial death rate, which declined between 30 and 180 days and thereafter remained at the lower value. Hospital mortality was 19.0%, with 21.7% dying within 6 months of ICU admission. Women had higher hospital mortality than men (20.8 vs. 17.8%; P = 0.0008). Among urban residents there was a steady gradient of declining hospital mortality with rising income (P < 0.0001). Mean ICU length of stay was 3.96 days, increasing 0.11 d/yr over the study period (P = 0.001); median ICU length of stay was 2.33 days and did not change over time. In the year after ICU care, 41% were rehospitalized, 10% were readmitted to an ICU, 98% had outpatient physician visits, 96% used prescription medications, and 27% used home care services. Although most of these parameters were statistically higher than for hospitalizations not requiring ICU care, differences were generally small. Among hospital survivors, 2.7% were discharged to chronic care facilities, with 2.5% living in such facilities 3 months later. CONCLUSIONS: Post-hospital medical resource use among ICU survivors is substantial, although similar to that after non-ICU hospitalization. Although the fraction of survivors unable to live independently was small, a larger fraction required home care services. Identifying Post-hospital supports needed by ICU survivors can be useful for policy makers and others responsible for healthcare planning.

Surge capacity: analysis of census fluctuations to estimate the number of intensive care unit beds needed.

OBJECTIVE: To compare methods of characterizing intensive care unit (ICU) bed use and estimate the number of beds needed. STUDY SETTING: Three geographic regions in the Canadian province of Manitoba. STUDY DESIGN: Retrospective analysis of population-based data from April 1, 2000, to March 31, 2007. METHODS: We compared three methods to estimate ICU bed requirements. Method 1 analyzed yearly patient-days. Methods 2 and 3 analyzed day-to-day fluctuations in patient census; these differed by whether each hospital needed to independently fulfill its own demand or this resource was shared across hospitals. PRINCIPAL FINDINGS: Three main findings were as follows: (1) estimates based on yearly average usage generally underestimated the number of beds needed compared to analysis of fluctuations in census, especially in the smaller regions where underestimation ranged 25-58 percent; (2) 4-29 percent fewer beds were needed if it was acceptable for demand to exceed supply 18 days/year, versus 4 days/year; and (3) 13-36 percent fewer beds were needed if hospitals within a region could effectively share ICU beds. CONCLUSIONS: Compared to using yearly averages, analyzing day-to-day fluctuations in patient census gives a more accurate picture of ICU bed use. Failing to provide adequate "surge capacity" can lead to demand that frequently and severely exceeds supply.

Health care utilization before and after intensive care unit admission in multiple sclerosis.

OBJECTIVES: The incidence of intensive care unit (ICU) admission is elevated in the multiple sclerosis (MS) population but the reasons for this are incompletely understood, as are outcomes post-ICU admission. Among MS patients we examined the association between ICU admission and health care utilization in the year preceding admission, and compared health care utilization following ICU admission among persons with MS and persons from the general population. METHODS: We used population-based administrative data from Manitoba, Canada to identify 4237 MS cases of which 2547 were incident. We compared the incidence rates of ICU admission in the prevalent MS population according to health care utilization in the year before admission, adjusting for age, sex, comorbidity and socioeconomic status. Among incident cases of MS we compared rates of health care utilization after ICU admission to those in a matched general population cohort. We used generalized linear models adjusting for age, sex, socioeconomic status, region, comorbidity and utilization before admission. RESULTS: Of 4219 prevalent MS cases, 222 (5.3%) were admitted to the ICU. After adjustment, any hospitalization in the prior year conferred an 80% increased incidence, and physician visits in the highest tertile and prescription costs in the highest quartile in the prior year each conferred a more than two-fold increased incidence of admission. Among 2547 incident cases of MS, 109 (4.3%) were admitted to the ICU and 93 survived their admission. Thirty-eight percent of the MS population were re-hospitalized in the year following admission, similar to the matched population (33.8%). Seven percent of both populations were readmitted to the ICU. The MS population had more hospital days after ICU admission than the matched population (adjusted RR 3.11; 95% CI: 1.34-5.90). After adjustment the number of physician visits did not differ between populations. CONCLUSIONS: The incidence of ICU admission is higher among persons with MS who have higher prior health care utilization. Health care utilization remains high after ICU admission. Efforts to prevent ICU admission in this population are needed.

Distinct determinants of long-term and short-term survival in critical illness.

PURPOSE: To identify the determinants of short-term and long-term survival in adult patients admitted to intensive care units (ICUs). METHODS: This population-based, observational cohort study included all eleven adult ICUs in the Winnipeg Health Region of Manitoba, Canada, analyzing initial ICU admissions during the period 1999-2010 of all Manitobans ≥17 years old. Analysis included Kaplan-Meier survival curves and multivariable regression models of 30-day mortality and post-90-day survival among those who survived to day 90. We used likelihood ratios to compare the predictive power of clusters of variables in these models. RESULTS: After 33,324 initial ICU admissions, mortality rates within 30 and 90 days were 15.9 and 19.5 %, respectively. The survival curve demonstrated an early phase with a high rate of death, followed by a markedly lower death rate that was only clearly established after several months. 30-day mortality was predominantly determined by characteristics of the acute illness; with its relative contribution set at 1.00, the next largest contributors were age (0.19) and comorbidity (0.16). In contrast, post-90-day mortality was mainly determined by age (relative contribution 1.00) and comorbidity (0.95); the next largest contributor was characteristics of acute illness (0.28). CONCLUSIONS: We observed two phases of survival related to critical illness. Short-term mortality was mainly determined by the acute illness, but its effect decayed relatively rapidly. Mortality beyond 3 months, among those who survived to that point, was mainly determined by age and comorbidity. Recognition of these findings is relevant to discussions with patients and surrogates about achievable goals of care.

The rising prevalence of asthma: true increase, diagnostic exchange or diagnostic accuracy?

In the midst of frequent reports about "the asthma epidemic," results from a number of studies by the manitoba centre for health policy have shown stable or decreasing prevalence of an overall indicator of respiratory diseases which includes asthma. To resolve these apparently contrary findings, we conducted a time trend analysis using administrative data. results revealed significant potential for diagnostic exchange: asthma prevalence increased, but that of bronchitis decreased.

Dementia and depression with ischemic heart disease: a population-based longitudinal study comparing interventional approaches to medical management.

BACKGROUND: We compared the proportion of ischemic heart disease (IHD) patients newly diagnosed with dementia and depression across three treatment groups: percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG) and medical management alone (IHD-medical). METHODS AND FINDINGS: De-identified, individual-level administrative records of health service use for the population of Manitoba, Canada (approximately 1.1 million) were examined. From April 1, 1993 to March 31, 1998, patients were identified with a diagnosis of IHD (ICD-9-CM codes). Index events of CABG or PCI were identified from April 1, 1998 to March 31, 2003. Outcomes were depression or dementia after the index event. Patients were followed forward to March 31, 2006 or until censored. Proportional hazards regression analysis was undertaken. Independent variables examined were age, sex, diabetes, hypertension and income quintile, medical management alone for IHD, or intervention by PCI or CABG. Age, sex, diabetes, and presence of hypertension were all strongly associated with the diagnosis of depression and dementia. There was no association with income quintile. Dementia was less frequent with PCI compared to medical management; (HR = 0.65; p = 0.017). CABG did not provide the same protective effect compared to medical management (HR = 0.90; p = 0.372). New diagnosis depression was more frequent with interventional approaches: PCI (n = 626; hazard ratio = 1.25; p = 0.028) and CABG (n = 1124, HR = 1.32; p = 0.0001) than non-interventional patients (n = 34,508). Subsequent CABG was nearly 16-fold higher (p<0.0001) and subsequent PCI was 22-fold higher (p<0.0001) for PCI-managed than CABG-managed patients. CONCLUSIONS: Patients managed with PCI had the lowest likelihood of dementia-only 65% of the risk for medical management alone. Both interventional approaches were associated with a higher risk of new diagnosed depression compared to medical management. Long-term myocardial revascularization was superior with CABG. These findings suggest that PCI may confer a long-term protective effect from dementia. The mechanism(s) of dementia protection requires elucidation.

Age difference explains gender difference in cardiac intervention rates after acute myocardial infarction.

Many investigators have reported higher rates of cardiac procedures for males than females after acute myocardial infarction (AMI), suggesting that men are treated more aggressively than women. However, others have reported no significant differences after controlling for age, resulting in uncertainty about the existence of a true gender bias in cardiac care. In this study, a population-based cohort approach was used to calculate age-specific procedure rates by sex from administrative data. Chi-square tests and generalized linear modelling were used to assess gender differences and interactions. For all four procedures studied, rates were significantly higher for males than females (p<0.01). However, age-specific rates revealed few significant differences by gender and a sharp decrease in intervention rates with age for both males and females. Generalized linear modelling confirmed that patient age was a significant predictor of intervention rates, whereas sex was not. The significant gender difference in overall rates was completely confounded by the older age profile of female AMI patients compared to their male counterparts.

Are cervical cancer screening rates different for women with schizophrenia? A Manitoba population-based study.

CONTEXT: Barriers to cervical cancer screening (Pap tests) may exist for women experiencing schizophrenia. DESIGN: This study analyzed healthcare records of all women in the province of Manitoba, Canada to: (a) compare cervical cancer screening rates of women with and without schizophrenia; and (b) determine factors associated with screening uptake. SETTING: This study took place in Manitoba, Canada, utilizing anonymized universal administrative data in the Population Health Research Data Repository at the Manitoba Centre for Health Policy. PARTICIPANTS: All females aged 18-69 living in Manitoba December 31, 2002, excluding those diagnosed with invasive or in situ cervical cancer in the study period or previous 5 years. MAIN OUTCOME: To determine factors associated with Papanicolaou (Pap) test uptake (1+ Pap test in 3 years, 2001/02-2003/04), logistic regression modeling included: diagnosis of schizophrenia, age, region, average household income, continuity of care (COC), presence of major physical comorbidity. Good COC was defined as at least 50% of ambulatory physician visits from the same general/family practitioner within two years. RESULTS: Women with schizophrenia (n=3220) were less likely to have a Pap test (58.8% vs. 67.8%, p<.0001) compared to all other women (n=335 294). In the logistic regression, a diagnosis of schizophrenia (aOR=0.70, 95% CI 0.65-0.75); aged 50+, and living in a low-income area or the North decreased likelihood; good continuity of care (aOR 1.88, 95% CI 1.85-1.91) and greater physical comorbidity (1.21, 95% CI 1.04-1.41) increased likelihood. CONCLUSION: Women with schizophrenia are less likely to receive appropriate cervical cancer screening. Since good continuity of care by primary care physicians may mitigate this, psychiatrists should consider assisting in ensuring screening uptake.