The psychosocial work environment and incident diabetes in Ontario, Canada.

BACKGROUND: Relatively few longitudinal studies have explored the relationship between psychosocial work conditions and diabetes incidence. Given the increasing global burden of diabetes this is an important area for public health research. AIMS: To examine the relationships between dimensions of the psychosocial work environment on the subsequent incidence of diabetes among men and women in Ontario, Canada over a 9 year period. METHODS: We used data from Ontario respondents (35 to 60 years of age) to the 2000-01 Canadian Community Health Survey linked to the Ontario Health Insurance Plan database for physician services and the Canadian Institute for Health Information Discharge Abstract Database for hospital admissions. Our sample of actively employed labour market participants with no previous diagnoses for diabetes was followed for a 9 year period to ascertain incident diabetes. RESULTS: There were 7443 participants. Low levels of job control were associated with an increased risk of diabetes among women, but not among men. Counter to our hypotheses high levels of social support were also associated with increased diabetes risk among women, but not among men. No relationship was found between any psychosocial work measure and risk of diabetes among men. CONCLUSIONS: Given the increasing prevalence of diabetes worldwide, job control could potentially be an import ant modifiable risk factor to reduce the incidence of diabetes among female, but not among male, workers. More research is needed to understand the pathways through which low social support may protect against the development of diabetes.

Disability management outcomes in the Ontario long-term care sector.

INTRODUCTION: Optimal disability management practices supporting early and safe return-to-work involve the workplace adoption of formal policies and procedures to ensure the quality of disability management outcomes. In the Canadian province of Ontario, there are approximately 60,000 health care workers in 600 licensed facilities providing long-term residential care to approximately 75,000 elderly residents. Workers in this sector are exposed to high biomechanical demands arising from care-giving tasks and have a substantial risk of work-related disability. Over the period 2000-2006, many long-term care facilities in Ontario adopted disability management practices that encourage modified work arrangements. The objective of this study was to describe differences in modified work arrangements and disability outcomes in long-term care facilities in Ontario. METHODS: Measures of disability episode outcomes are described for a representative sample of 32 Ontario long-term care facilities for two consecutive years 2005 and 2006. Data were obtained from a questionnaire survey of facilities, a survey of a representative sample of caregivers and administrative records from the provincial workers' compensation agency. RESULTS: A total of 28,747 days of disability attributed to work-related conditions were experienced by 3,271 full-time equivalent staff in 2005 (28,034 days in 2006). Average total disability days were 922 per 100 full-time equivalent staff in 2005 and 889 per 100 full-time equivalent staff in 2006. Disability compensation expenditures, measured as wage replacement benefits received by disabled workers, were estimated to be $72,332 per 100 full-time equivalent staff in 2005 and $64,619 per 100 full-time equivalent staff in 2006. On average, approximately 60% of all disability days were managed by modified duty arrangements and the proportion of total disability days managed by modified duty arrangements for each facility was correlated between the two observation years. CONCLUSIONS: Across facilities, there was no evidence that modified duty arrangements were associated with lower disability compensation expenditures and there was mixed evidence that modified duty was associated with a lower burden of disability. In this setting, disability days managed by modified duty arrangements were not accurately documented in worker's compensation claim records.

Comparing the risk of work-related injuries between immigrants to Canada and Canadian-born labour market participants.

OBJECTIVES: To examine the burden of work-related injuries among immigrants to Canada compared to Canadian-born labour force participants. METHODS: Using data from the 2003 and 2005 Canadian Community Health Surveys (n = 99,115), two nationally representative population samples, we examined the risk of self-reported, activity limiting work-related injuries among immigrants with varying time periods since arrival in Canada. Models were adjusted for hours of work in the last 12 months as well as various demographic and work-related variables. RESULTS: Immigrant men in their first 5 years in Canada reported lower rates of activity limiting injuries compared to Canadian-born respondents. Surprisingly, the percentage of injuries that required medical attention was much higher among recent immigrants compared to Canadian-born respondents, resulting in an increased risk of activity limiting injuries requiring medical attention among immigrant men compared to Canadian-born labour force participants. No excess risk was found among female immigrants compared to Canadian-born female labour market participants. CONCLUSIONS: Immigrant men in their first 5 years in Canada are at increased risk of work-related injuries that require medical attention. A similar risk is not present among immigrant women. Further, given differences in the number of activity limiting injuries requiring medical attention across immigrant groups, we believe this excess risk among immigrant men may be underestimated in the current data source. Future research should attempt to fully capture the barriers faced by immigrants in obtaining safe employment, the number of injuries that are sustained by immigrants while working, and the consequences of these injuries.

Examining the relationships between job control and health status: a path analysis approach.

OBJECTIVE: To examine the pathways through which job control affects health status; to examine if the effects of job control on health status are attenuated by including other measures associated with lower socioeconomic status, and to examine if the relationship between job control and health status is consistent across socioeconomic status groups. DESIGN: A prospective observational cohort study over eight years (1994-2002). PARTICIPANTS: 4886 Respondents aged 25-60 years, who were non-self-employed labour force participants, working more than 20 hours per week, without physical or mental limitations restricting the type or amount of work they could do at baseline. After longitudinal attrition, the remaining study sample was 3411 (87% of the original study sample who did not die or become pregnant during the survey period). MAIN RESULTS: Low job control in 1994 was associated with worse than expected self-rated health in 2002, both directly and indirectly via a lower physical activity level in 1996. Adjustment for other factors associated with low socioeconomic status did not attenuate these relationships to a large extent. No differences were found in the effects of job control on physical activity or health status between socioeconomic groups (high and low education and high and low household income). CONCLUSIONS: The inclusion of other factors associated with lower socioeconomic status did not attenuate the direct and indirect effects of job control on health status. The finding that low job control is associated with lower physical activity levels deserves further investigation, given the increasing concern about rising levels of obesity in the developed world.

Age-social stratification designs had a negligible impact on income-mortality associations.

OBJECTIVES: Age-social stratification has been used to offset socioeconomic status (SES) misclassification due to cohort effects. This study was to evaluate whether age-income stratification designs generate comparable income-mortality associations as those whose income rankings are based on absolute thresholds. STUDY DESIGN AND SETTING: Using self-reported income as our SES variable, and mortality as our outcome measure, the impact of age-social stratification was examined in two distinct cohorts: one with acute myocardial infarction (AMI) (n=3,138), and the second free of cardiovascular disease (n=15,115). Age-adjusted income-mortality associations were compared between age-social stratification techniques, which used "age-relative" income thresholds and "absolute" income thresholds whose ranks were independent of patient age. RESULTS: In both cohorts, crude mortality inversely correlated with age and income. Techniques using "age-relative" income thresholds yielded similar adjusted odds ratio for mortality as did those that used "absolute" income threshold methods (differences in adjusted odds ratios [+/-95% confidence interval (CI)] between "absolute" and "age-relative" classifications for highest vs. lowest income tertiles: -0.05 [-0.24, 0.12] among patients with AMI and 0.05 [-0.03, 0.13] among patients without cardiovascular disease). CONCLUSION: More complex designs incorporating age-social stratification techniques generate similar income-mortality associations as more simplified approaches, which classified SES using absolute income thresholds.

Healthcare use before and after a workplace injury in British Columbia, Canada.

OBJECTIVES: There is growing evidence that occupational injuries influence workers' emotional and physical wellbeing, extending healthcare use beyond what is covered by the Workers' Compensation Board (WCB). METHODS: The authors used an administrative database that links individual publicly funded healthcare and WCB data for the population of British Columbia (BC), Canada. They examined change in service use, relative to one year before the injury, for workers who required time off for their injuries (lost time = LT) and compared them to other injured workers (no lost time = NLT) and individuals in the population who were not injured (non-injured = NI). RESULTS: LT workers increased physician visits (22%), hospital days (50%), and mental healthcare use (43% physician visits; and 70% hospital days) five years after the injury, relative to the year before the injury, at a higher rate than the NI group. For the NLT workers, the level of increased use following the injury was between that of these two groups. These patterns persisted when adjusting for registration in the BC Medical Service Plan (MSP) and several workplace characteristics. CONCLUSIONS: Although the WCB system is the primary mechanism for processing claims and providing information about workplace injury, it is clear that the consequences of workplace injury extend beyond what is covered by the WCB into the publicly funded healthcare system.

Childhood and early adult predictors of risk of incident back pain: Ontario Child Health Study 2001 follow-up.

Musculoskeletal disorders of the back and spine are a leading cause of disability in working-age populations. There is limited information on the potential consequences of childhood socioeconomic and health status on the risk of incident back pain in early adulthood. The authors describe factors associated with having had a first episode of back pain during the past year in the Ontario Child Health Study, a prospective cohort study of children who were aged 4-16 years at the time of enrollment in 1983 and were resurveyed in 2001. Respondents reporting a first episode of back pain (n=143) were compared with respondents who had never experienced back pain (n=896). The annual incidence of a first episode of back pain in this sample of young adults was 74.7/1,000. Following adjustment for age, sex, childhood conditions, childhood health status, and measures of early adult health, behavior, socioeconomic status, and work environment, the risk of incident back pain was associated with both low (odds ratio (OR)=1.86, 95% confidence interval (CI): 1.14, 3.03) and moderate/high (OR=1.85, 95% CI: 1.07, 3.02) levels of psychological distress, current heavy smoking (OR=1.85, 95% CI: 1.10, 3.10), lower levels of parental education in childhood (OR=1.72, 95% CI: 1.06, 2.80), and emotional or behavioral disorders in childhood (OR=1.87, 95% CI: 1.02, 3.41). The associations of low childhood socioeconomic status and childhood emotional and behavioral disorders with risk of incident back pain in early adulthood are important findings with implications for better understanding the etiology of soft-tissue disorders.

Examining the associations between physical work demands and work injury rates between men and women in Ontario, 1990-2000.

AIMS: To describe the decline in injury rates between 1990 and 2000 within occupations stratified across three levels of physical demands and gender, adjusting for industry, in Canada's largest province. METHODS: Records of injury compensation claims were obtained from the Ontario Workplace Safety & Insurance Board. The population likely to be insured by the Ontario Workplace Safety & Insurance Board was estimated from Statistics Canada's Labour Force Survey. Injury rates were calculated by three broad levels of physical demands, separately for men and women. RESULTS: Injury rates decreased across each grouping of lower physical demands at work for both men and women, with the largest absolute differences in manual occupational groups (high physical demands). Occupations classified as manual (high physical demands) and mixed (moderate physical demands) showed larger differences in injury rates between genders than did non-manual (low physical demands), although the directions of these differences were not always consistent across different natures of injury classification. CONCLUSIONS: The absolute reduction in injury rates in Ontario between 1990 and 2000 was dominated by the reduction in injury rates for men and women in manual and mixed occupations. However, not all types of injury have declined to the same extent. A large proportion of differences in injury rates between men and women can be attributed to the differential labour force participation across occupations and industries, as well as the differential tasks within occupational groups.

Gender differences in socioeconomic inequality in mortality.

OBJECTIVES: There is uncertainty about whether position in a socioeconomic hierarchy confers different mortality risks on men and women. The objective of this study was to conduct a systematic review of gender differences in socioeconomic inequality in risk of death. METHODS: This research systematically reviewed observational cohort studies describing all cause or cause specific mortality for populations aged 25-64 in developed countries. For inclusion in the review, mortality had to be reported stratified by gender and by one or more measures of socioeconomic status. For all eligible studies, five absolute and six relative measures of the socioeconomic inequality in mortality were computed for male and female populations separately. RESULTS: A total of 136 published papers were reviewed for eligibility, with 58 studies deemed eligible for inclusion. Of these eligible studies, 20 papers published data that permitted the computation of both absolute and relative measures of inequality. Absolute measures of socioeconomic mortality inequality for men and women generally agreed, with about 90% of studies indicating that male mortality was more unequal than female mortality across socioeconomic groups. In contrast, the pattern of relative inequality results across the 20 studies suggested that male and female socioeconomic inequality in mortality was equivalent. CONCLUSIONS: Inferences about gender differences in socioeconomic inequality in mortality are sensitive to the choice of inequality measure. Wider understanding of this methodological issue would improve the clarity of the reporting and synthesis of evidence on the magnitude of health inequalities in populations.

Socioeconomic status, drug insurance benefits, and new prescriptions for inhaled corticosteroids in schoolchildren with asthma.

BACKGROUND: Low-income children with asthma are less likely to receive inhaled corticosteroid prescriptions that can prevent asthma morbidity. OBJECTIVE: To determine whether the receipt of inhaled corticosteroids in children with asthma is related to household socioeconomic status and type of drug insurance. DESIGN: Using population-based prescription and health care data from Manitoba, a cohort study of the determinants of receiving new prescriptions for inhaled corticosteroids was conducted in children treated with asthma drugs. PARTICIPANTS: School-aged children (n = 12 481) receiving asthma prescriptions from January 1995 to March 1996 but no inhaled corticosteroid prescriptions in the initial 6-month period. MAIN OUTCOME MEASURES: Household socioeconomic and drug insurance predictors of the probability of receiving a new inhaled corticosteroid prescription from July 1995 to March 1998, following adjustment for disease and health care utilization factors. RESULTS: In comparison with higher-income children insured through a provincial cost-sharing drug plan, the adjusted likelihood ratio for a new inhaled corticosteroid prescription was 0.88 (95% confidence interval, 0.80-0.97) in low-income children insured through the same drug plan and 0.82 (95% confidence interval, 0.76-0.88) in children receiving prescriptions at no charge through provincial income assistance or First Nations benefits programs (Winnipeg, Manitoba). CONCLUSION: Independent of asthma severity, type of drug insurance, or health care utilization patterns, low-income children with asthma are significantly less likely to receive inhaled corticosteroid prescriptions.

Income-based drug benefit policy: impact on receipt of inhaled corticosteroid prescriptions by Manitoba children with asthma.

BACKGROUND: Drug benefit policies are an important determinant of a population's use of prescription drugs. This study was undertaken to determine whether a change in a provincial drug benefit policy, from a fixed deductible and copayment system to an income-based deductible system, resulted in changes in receipt of prescriptions for inhaled corticosteroids by Manitoba children with asthma. METHODS: Using Manitoba's health care administrative databases, we identified a population-based cohort of 10,703 school-aged children who met our case definition for asthma treatment before and after the province's drug benefit policy was changed in April 1996. The effects of the program change on the probability of receiving a prescription for an inhaled corticosteroid and on the mean number of inhaled corticosteroid doses dispensed were compared between a group of children insured under other drug programs (the comparison group) and 2 groups of children insured under the deductible program: those living in low-income neighbourhoods and those living in higher-income neighbourhoods. All analyses were adjusted for a measure of asthma severity. RESULTS: For higher-income children with severe asthma who were covered by the deductible program, the probability of receiving an inhaled corticosteroid prescription and the mean annual number of inhaled corticosteroid doses declined after the change to the drug policy. A trend toward a decrease in receipt of prescriptions was also observed for low-income children, but receipt of prescriptions was unaltered in the comparison group. Before the policy change, among children with severe asthma, the mean annual number of inhaled corticosteroid doses was lowest for low-income children, and this pattern persisted after the change. Among children with mild to moderate asthma, those covered by the deductible program (both low income and higher income) were less likely to receive prescriptions for inhaled corticosteroids than those in the comparison group, and this difference was statistically significant for the higher-income children. INTERPRETATION: The change to an income-based drug benefit policy was associated with a decrease in the use of inhaled corticosteroids by higher-income children with severe asthma and did not improve use of these drugs by low-income children.

Work-related population health indicators.

OBJECTIVE: To provide evidence-based recommendations for work-related population health indicators. METHODS: Drawing on a framework of work-related experiences, we systematically reviewed studies that assess the association between these experiences and health and reviewed related measures at the population level that could be used as indicators. RESULTS: We recommend (and grade the strength of evidence supporting our recommendation for) the following indicators for which data are already routinely collected: unemployment rate (strong), long-term unemployment rate (limited), and permanent lay-off rate (limited). As well, we recommend and grade our support for the following new indicators: insecurity associated with pending job loss (limited), with possible major organizational change (limited), and with actual major organizational change (limited); and job strain (medium). CONCLUSION: These evidence-based indicators can be used to monitor work-related determinants of health and thus to inform the conceptualization, development, and evaluation of policies and programs related to these determinants.

Undiagnosed diabetes: does it matter?

BACKGROUND: The 1998 Canadian clinical practice guidelines for the management of diabetes lowered the cutoff point for diagnosing diabetes mellitus from a fasting plasma glucose (FPG) level of 7.8 to 7.0 mmol/L. We studied the prevalence and clinical outcomes of undiagnosed and diagnosed diabetes within specific ranges of FPG among a cohort of subjects recruited in 1990. METHODS: In 1990 a representative sample of 2792 adult residents of Manitoba participated in the Manitoba Heart Health Survey, which included measurement of FPG and a question about each participant's past history of diabetes. Individuals who would now be classified as having undiagnosed diabetes under the new criteria were not considered as such in 1990. Through data linkage with the provincial health care utilization database, the use of health care by these individuals was tracked and compared with that of individuals whose diabetes had been diagnosed and with that of normoglycemic individuals over an 8-year period subsequent to the survey. RESULTS: The prevalence of undiagnosed diabetes in the adult population of Manitoba was 2.2%. Undiagnosed cases accounted for about one-third of all diabetes cases. Individuals with undiagnosed diabetes had an unfavourable lipid profile and higher blood pressure and obesity indices than normoglycemic individuals. Individuals who satisfied the new criteria for diabetes but remained undiagnosed had an additional 1.35 physician visits per year (95% confidence interval [95% CI] 0.93-1.96) and were more likely to be admitted to hospital at least once (odds ratio 1.23, 95% CI 0.40-3.79), compared with normoglycemic individuals. INTERPRETATION: Undiagnosed cases represent the unseen but clinically important burden of diabetes, with significant concurrent metabolic derangements and a long-term impact on health care use.

Factors associated with nursing-home entry for elders in Manitoba, Canada.

BACKGROUND: As the population ages, a greater demand for long-term care services and, in particular, nursing homes is expected. Policy analysts continue to search for alternative, less costly forms of care for the elderly and have attempted to develop programs to delay or prevent nursing-home entry. Health care administrators require information for planning the future demand for nursing-home services. This study assesses the relative importance of predisposing, enabling, and need characteristics in predicting and understanding nursing-home entry. METHODS: Proportional hazard models, incorporating changes in needs over time, are used to estimate the hazard of nursing-home entry over a 5-year period, using health and sociodemographic characteristics of a representative sample of elderly residents from Manitoba, Canada. RESULTS: After age, need factors have the greatest impact on nursing-home entry. Specific medical conditions have at least as great a contribution as functional limitations. The presence of a spouse significantly reduces the hazard of entry for males only. CONCLUSIONS: The results suggest that the greatest gains in preventing or delaying nursing-home entry can be achieved through intervention programs targeted at specific medical conditions such as Alzheimer's disease, musculoskeletal disorders, and stroke.

Widening regional inequality in premature mortality rates in Manitoba.

OBJECTIVE: To describe regional trends in premature mortality in Manitoba. DESIGN: Comparison of all-cause and cause-specific mortality of persons less than age 75 in 11 Regional Health Authority populations over two time periods: 1985-89 and 1990-94. RESULTS: The provincial premature mortality rate declined over the two time periods (4.00/1,000 to 3.72/1,000). Declines were also observed in 9 of 11 regional populations. Premature mortality increased, however, in the 2 regional populations with the highest mortality rates in the first observation period. CONCLUSION: Declining premature mortality in low mortality populations and rising premature mortality in high mortality populations has resulted in a widening of regional mortality rates in Manitoba. Recent policy initiatives in many provinces, including the devolution of authority for the management and delivery of health services and the implementation of population need-based funding formulas to share health care resources among regional health authorities, if implemented, have the potential to partially mitigate the processes producing these widening regional health inequalities.

Assessing ecologic proxies for household income: a comparison of household and neighbourhood level income measures in the study of population health status.

This paper examines the validity of using ecologic measures of socioeconomic status as proxies for individual-level measures in the study of population health. Based on a representative 5% sample of households in a Canadian province, the study integrated three sources of information: administrative records of individual health care utilization, records of deaths and 1986 census records which contained information on household income and average neighbourhood income. Thirteen measures of health status were developed from these sources of information. The hypothesis that risk estimates derived from ecologic income measures will be attenuated relative to estimates obtained from household income was not supported. These results provide evidence for the use of ecologic-level measures of income in studies which do not have access to individual-level income measures.

Validation of an electronic, population-based prescription database.

BACKGROUND: The Drug Programs Information Network (DPIN), Manitoba's (Canada) new electronic prescription database, is a valuable data source for pharmacoepidemiologic research. Pharmacies are required to submit to the DPIN all prescriptions for Pharmacare, the province's drug insurance plan, but submission of prescriptions for social assistance recipients and treaty status Indians is discretionary. OBJECTIVE: The completeness of the DPIN prescription database was assessed to determine whether treaty status Indians and social assistance recipients were underrepresented. DESIGN: Prescriptions dispensed during March 13-17, 1995, in a stratified sample of Manitoba pharmacies were linked to DPIN by prescription number to determine the proportions submitted for Indian Affairs, Social Services, and Pharmacare recipients. Pharmacare records in the DPIN were compared with original pharmacy records to evaluate data accuracy. RESULTS: Of 2196 Indian Affairs and 1879 Social Services prescriptions dispensed in 58 pharmacies, a corresponding prescription was found in the DPIN for 79.7% (98% CI 78.0% to 81.4%) and 90.1% (98% CI 88.8% to 91.4%) of prescriptions, respectively. These proportions were significantly lower than the estimated proportion of Pharmacare prescriptions submitted (93%, 98% CI 92.4% to 93.6%). Ninety-two percent of 8012 DPIN Pharmacare prescriptions matched the original prescription on the drug name, quantity, and days' supply. CONCLUSIONS: This study established that the DPIN is a valid and reliable data source for studying prescription use among the majority of Manitoban residents. However, the DPIN database has differential validity and underrepresents prescriptions dispensed for the aboriginal population.

A profile of health care utilization of the disabled population in Manitoba.

OBJECTIVES: This study profiled health care utilization by disabled and nondisabled individuals in the Canadian province of Manitoba to evaluate the association between health care utilization and disability. METHODS: Age-standardized annualized utilization rates were calculated according to sex using longitudinal data on individual encounters with the Manitoba health care system from 1983 to 1990. Associations between severity of disability, number of prior chronic conditions, and prospective utilization were examined using multivariate regressions. RESULTS: Utilization patterns of the mildly disabled and the nondisabled differed only slightly. Severely disabled individuals had much higher rates of contact and consumed more resources, even after controlling for chronic conditions. The severely disabled accounted for 3% of the population and consumed 16% of hospital days and 7% of physician costs annually. CONCLUSIONS: The findings emphasize the importance of incorporating measures of disability in health services research. Both the severity of disability and the number of chronic conditions had independent value in predicting health care utilization. This has important implications for data collection and for the allocation of health care resources for research, which has traditionally been targeted toward fatal chronic conditions.

Sex differences in the use of health care services.

BACKGROUND: Sex differences in the use of health care services can be substantial at several stages of life. However, the extent to which differences in reproductive biology and mortality affect the use of health care services is unclear. METHODS: We studied age- and sex-specific per capita use of health care resources for a one-year period during 1994 and 1995 in the Canadian province of Manitoba, where there is universal insurance for a comprehensive range of health care services. Using information obtained from administrative records of physicians' services and acute hospital care, we tabulated the use of health care resources by male and female subjects in three categories: care for conditions specific to sex, care provided to persons who died during the study year, and care provided for all other conditions. RESULTS: The crude annual per capita use of health care resources (in Canadian dollars) was greater for female subjects ($1,164) than for male subjects ($918). Approximately 22 percent of health care expenditures for female subjects was associated with conditions specific to sex, including pregnancy and childbirth, as compared with 3 percent of expenditures for male subjects. An estimated 14 percent of health care expenditures for male subjects was consumed by persons who died during the study period, as compared with 10 percent of expenditures for female subjects. After adjustment for the use of health care associated with sex-specific conditions and differences in mortality, the female:male ratio in health care expenditures was reduced from 1.3 to 1.0. CONCLUSIONS: Expenditures for health care are similar for male and female subjects after differences in reproductive biology and higher age-specific mortality rates among men have been accounted for.

Emergency department use as a component of total ambulatory care: a population perspective.

OBJECTIVES: (a) To describe the overall proportion of ambulatory care provided in emergency departments for a complete urban population, (b) to describe the variation across small geographic areas in the overall proportion of ambulatory care provided in emergency departments and (c) to identify attributes of small-area populations that are related to the provision of high proportions of total ambulatory care in emergency departments. DESIGN: Cross-sectional ecologic study combining 4 sources of secondary data on health service utilization and socioeconomic status. SETTING: Winnipeg. PARTICIPANTS: A total of 657,871 residents of metropolitan Winnipeg in the period April 1991 to March 1992, grouped into 112 neighbourhoods. MAIN OUTCOME MEASURE: A proportion calculated, for each neighbourhood population, from the estimated count of emergency department visits divided by the population's use of total ambulatory care for a sample of 55 days in the study period. RESULTS: The overall proportion of ambulatory care provided in emergency departments was 4.9% (range 2.6% to 10.8%), representing 35.5 emergency department visits per 100 person-years. Neighbourhoods with a higher proportion of total ambulatory care provided in emergency departments were characterized by lower mean household income, a higher proportion of emergency department visits for mental illness and a higher proportion of residents with treaty Indian status. Measures of need for medical care for were not consistently associated with the proportion of ambulatory care received in emergency departments. CONCLUSIONS: In a health care system with an adequate supply of primary care physicians and universal insurance, this study has documented significant variation across small geographic areas in the proportion of total ambulatory care received in emergency departments. In the absence of strong evidence that this variation was associated with underlying need, the results suggest that attention be paid to the accessibility of conventional primary care.