A population-based study of emergency department presentations for asthma in regions of Alberta

Community EM

CJEM 2010;12(4):339-346

Abstract

Résumé

Introduction

Asthma is a common disease with a prevalence of approximately 7%–10% in adults and 15% in children.1 The prevalence of this disease is thought to be on the increase, perhaps because of gene–environmental factors, obesity and the hygiene hypothesis.2,3

Asthma control can be achieved by limiting exposure to triggering stimuli, by education and by employing anti-inflammatory agents alone or in combination with bronchodilators.4-6 Despite these interventions, asthma control can be elusive and exacerbations can frequently occur. Moderate to severe exacerbations are a common reason for presentation to the emergency department (ED), and some people with asthma return frequently. In Canada, the treatment of acute asthma accounts for close to one-quarter of the nearly $500 million spent annually on asthma.7,8

Rowe and colleagues9 and Rosychuk and coauthors10 have previously described the epidemiology of presentations to EDs made by adults9 and children10 in the province of Alberta, using large population-based administrative health databases. The objective of this study was to explore interregional differences in greater detail. Specifically, we wanted to compare the presentation rates, outcomes, ED visit durations and follow-up visits based on the hypothesis that the major urban regions of Edmonton and Calgary differ from the rest of the province.

Methods

Setting

As in other provinces, all residents of Alberta can access free health care, assuming they are eligible for and maintain health insurance through the Alberta Health Care Insurance Plan (AHCIP). In Alberta, some citizens still pay for their health care if they are self-employed, and subsidies exist for those with limited financial means.

Data sources

The Ambulatory Care Classification System (ACCS)11 was developed as a flexible and integrated system for tracking ambulatory care visits within government-funded facilities in Alberta. The only other ED database in Canada is the National Ambulatory Care Reporting System, which includes all ED data from Ontario and data from some EDs in British Columbia, Nova Scotia and the Yukon.12 All ED encounters at more than 100 provincial emergency care facilities are entered into computerized abstracts that constitute the majority of records within the ACCS system. Using a uniform protocol, trained medical records nosologists code each chart using ICD-9-CM (International Classification of Diseases, 9th Revision, Clinical Modification) diagnostic codes (before Apr. 1, 2002) or ICD-10-CA (International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canadian Enhancement) (Apr. 1, 2002, onward) at each ED in the province.

Each ACCS record represents a service and contains a personal health number (unique to each Albertan), start and end dates and times of the visit, diagnoses, disposition status and region of the ED facility where the service occurred. We obtained demographic data by linking the individuals in the ACCS to the annual AHCIP cumulative registry file based on the personal health number (true linkage). This file includes all people (99% of the population) registered under the AHCIP at any time in a given year. The demographic information obtained included age, sex and region of residence.

We also obtained data on subsequent visits to physicians in non-ED settings (follow-up visits) by linking the individuals in the ACCS to the individuals in the physician claims database. This database provides dates of follow-up visits, diagnoses, physician specialties and facility types.

Variable description

Diagnostic information in the ACCS consisted of a main ambulatory diagnosis field, and 5 and 9 additional diagnostic fields, for ICD-9-CM and ICD-10-CA codes, respectively. To be considered an asthma-related visit, the first or second diagnosis fields in the ACCS had to have diagnostic codes 493.x “Asthma all forms” or J45.x “Asthma all forms.” Previous studies using ACCS data indicate that the accuracy of the diagnosis is approximately 97%.13-15

The difference between the start and end dates and times provided the duration of an ED service in hours for each visit. Disposition from the ED is based on the manner in which patients are separated/released from the ambulatory service facility, and includes “discharged” (returning to previous place of residence) and “admitted” (admitted to hospital in a critical care or inpatient area in a health facility).

Ages were grouped into approximate 5-year categories and 2 subgroups were created: children (age < 18 yr) and adults (age = 18 yr). Until 2008, Alberta had 9 regional health authorities (RHAs) responsible for the delivery of health care services. Patients were assigned codes according to which RHA they lived in at the end of the fiscal year, and the RHA of the ED facility was also provided. The 2 most urban RHAs are Edmonton and Calgary (including surrounding areas). The remaining 7 RHAs are grouped together (non–major urban [NMU] regions) to divide the province into 3 regions with approximately similar populations.

The follow-up visit data include the date of visit, 3 ICD-9 diagnosis fields, physician specialty codes and facility types. The follow-up visits within 365 days of an ED visit were extracted and classified as asthma-related if the first or second diagnosis field was ICD-9 coded as 493.x. We used 11 physician specialty codes (e.g., general practice, respiratory medicine) and 3 facility types (active treatment hospital, and clinic and ambulatory care centre; practitioner’s office; other).

Data analysis

All ED visits including our asthma diagnostic codes between Apr. 1, 1999, and Mar. 31, 2005, were extracted. Numerical summaries (e.g., frequencies, medians, interquartile ranges [IQRs]) were calculated by region. Age group–specific ED visits per 1000 population were calculated for each sex. Directly standardized visit rates for ED visits and associated standard deviations16 were calculated and tested using the 1999/2000 Alberta population as the reference. Proportions and continuous values were compared using χ² and Kruskal–Wallis tests, respectively. Region refers to region of residence, except for duration and outcome summaries, for which the region is the region of the ED facility. The latter 2 variables represent ED-specific information and hence the region is specified as the ED facility.

For analyses with both ED and follow-up visits, a discharged data subset of “index” ED visits was created for the period from Nov. 1, 2003, to Oct. 31, 2004. If an individual patient had multiple ED visits during this time frame, one visit was randomly selected. This subset allowed summaries at 7 and 30 days after the index ED visit to be at the individual level, permitted a follow-up time of at least 6 months and avoided variability associated with different fiscal years. The times to the next ED visit and the first follow-up visit were calculated, with observations censored at Mar. 31, 2005, or the date of the next ED visit, respectively. SAS (version 9.1, SAS Institute Inc.) and S-PLUS (version 7.0, TIBCO Software Inc.) were used for analysis.

Ethics

The study was approved by the health research ethics board of the University of Alberta.

Results

General trends

During the 6-year period, there were 199 991 ED visits for asthma (Tables 1 and 2) from Alberta residents for whom regional information was available (region of residence was missing for 9 ED visits). These ED visits for asthma accounted for approximately 2% of the total number of ED visits for any reason. The number of ED visits for asthma in each region was relatively stable over the years, although the population increased to approximately one million residents in each of the 3 study regions by 2004/05. Crude visit rates were lower in the Calgary and Edmonton regions, compared with the NMU regions. By 2004/05, the crude visit rates per 1000 population for the Calgary, Edmonton and NMU regions were 6.5, 7.9 and 15.4, respectively.

Overall, the 199 991 ED visits were made by 93 146 patients during the 6-year study period. Most patients had only 1 visit during the study period: 65.2% for Edmonton, 65.7% for Calgary and 60.9% for the NMU regions. The mean number of visits per patient for the Edmonton, Calgary and NMU regions was 2.0, 1.9 and 2.3, respectively. The vast majority of visits (93.2%) were made to ED facilities within the same region as patients’ region of residence.

Age and sex

In all 3 regions, the overall sex split was similar (female sex 51.0% for Edmonton, 47.5% for Calgary and 51.8% for NMU regions). Until age 14, more male patients than female patients presented for asthma in each region (Fig. 1). Visit rates were highest among male patients in the 1- to 4-year age group (37.9/1000 for Edmonton, 45.5/1000 for Calgary, 57.9/1000 for NMU regions) compared with visit rates among female patients of 24.7/1000 for Edmonton, 26.5/1000 for Calgary and 29.2/1000 for the NMU regions. After age 14, the pattern reversed so that older female patients present more often than males. These patterns were similar for each year.

Over time, the sex and age group directly standardized visit rates (DSVRs) remained relatively stable (Fig. 2) for Edmonton and Calgary with 8.3/1000 (95% confidence interval [CI] 8.0–8.6) and 6.9/1000 (95% CI 6.6–7.0), respectively, in 2004/05. For the NMU regions, the rates were much higher than in the other 2 regions and declined from 21.0/1000 in 1999/2000 until around 2002/03, and remained relatively stable thereafter with 15.5/1000 (95% CI 15.1–15.9) in 2004/05. All regions had statistically different DSVRs (p < 0.001 for each year).

Outcomes

The vast majority of patients with asthma in each region were discharged from the ED (86.7%–92.6% in 2004/05, Table 1). In 2004/05, the Calgary region had the lowest proportion of discharges (86.7%; χ²2 = 12.7, p = 0.002). The proportion of patients who left without completion of care was small in all 3 regions, with Edmonton experiencing the highest proportion for 2004/05 (χ²2 = 114.2, p < 0.001).

Visit duration

The median length of stay was higher (p < 0.001) in Edmonton (n = 47 597; 2 h 32 min, IQR 1 h 27 min to 4 h 20 min) and Calgary (n = 35 732; 2 h 46 min, IQR 1 h 26 min to 4 h 49 min) than in the NMU regions (n = 95 036; 1 h, IQR 33 min to 2 h).

Admitted patients had longer median ED lengths of stay in Edmonton (6 h 33 min) and Calgary (6 h 48 min) compared with the NMU regions (1 h 55 min). Discharged patients had the following median lengths of stay: Edmonton (2 h 21 min), Calgary (2 h 31 min) and the NMU regions (1 h). For each region, admitted children had shorter median ED lengths of stay than admitted adults (p < 0.001 for each region, Fig. 3).

Discharged subset

Between Nov. 1, 2003, and Oct. 31, 2004, 19 946 patients with region of residence information were discharged from the ED. In the 7 days after the index ED visits, Calgary (45.5%) and Edmonton (36.6%) had higher asthma-related follow-up visits than the NMU regions (30.3%; χ²2 = 196.6, p < 0.001) by 7 days (Table 3). The NMU regions had a higher proportion of visits to general practitioners (78.9%) than the Edmonton (54.6%) and Calgary (55.9%) regions (χ²2 = 687.9, p < 0.001). The follow-up visits were more likely to occur in practitioners’ offices in Edmonton (65.9%) and Calgary (62.0%) than in the NMU regions (50.6%; χ²4 = 330.5, p < 0.001). The results at 30 days followed the same general patterns.

At 7 days after discharge from the ED, 4.4% of patients in Edmonton and 5.0% in Calgary had returned to the ED. In the NMU regions, 7.2% had returned. The estimated median time to the first follow-up visit was 21 days for Edmonton, 19 days for Calgary and 26 days for the NMU regions.

Discussion

This study explored presentations to EDs for asthma in 3 main regions of the province of Alberta during a 6-year period, using large provincial databases. The study demonstrated markedly higher rates of presentation in NMU regions in the province. Overall, a patient visits an ED in Edmonton, Calgary and the NMU regions every 64, 67 and 30 minutes, respectively. Patients with asthma in Edmonton, Calgary and the NMU regions are admitted to hospital every 2.1, 2.6 and 3.8 days.

Although it is important to put these visit rates in perspective, comparative data are sparse. Using a 1-year stratified sample of 16 hospitals in Ontario, the estimated age- and sex-standardized ED visit rates ranged from 1.7 to 10.1 per 1000 population for adults, again with marked variation.17 Our results demonstrate higher visit rates, especially in the NMU regions. Using a representative sample of EDs in the United States, an overall rate of 6.7 visits per 1000 population was reported, which had appeared to plateau despite marked variability.18 A higher proportion of patients with asthma are admitted to hospitals in the United States compared with Canada, a phenomenon which appears to be multifactorial.19

Our study identified new interregional differences and temporal trends. In each region studied, the number of asthma-related visits to EDs has remained relatively stable despite increasing populations. Explanations may include ED overcrowding that results in acute asthma being treated elsewhere, improved access to after-hours care in major centres and improved application of evidence-based management in primary care. Recent evidence has examined the association of air quality on asthma presentations.20,21 Nonurban air quality in Alberta may be influenced by proximity to cities and industrial pollution and other environmental allergens. Moreover, options for rural patients to seek care may be more limited because of the lack of available alternatives. Consequently, the severity of asthma in patients presenting to these sites may be lower.

Finally, the most important barrier to asthma visits in urban Alberta is overcrowding in urban EDs. Emergency department overcrowding has been a growing concern across Alberta, especially in major centres.22 As proof, patients with asthma who presented to Edmonton and Calgary region facilities experienced longer stays, whether they were discharged or admitted to hospital. In keeping with national trends, adults waited longer than children. The use of inhaled corticosteroids and combination agents has been increasing in Canada,23 and status Aboriginals and welfare recipients within the urban centres have higher rates of asthma-related ED visits.24 Further research is required to determine the relative contribution of each of these factors.

After an acute exacerbation requiring ED care, follow-up reassessment by the primary care provider is recommended in both pediatric and adult asthma guidelines;24,25 however, the timing of this follow-up is unclear. Most asthma follow-up rates from clinical trials are spuriously high. Approximately 30% of discharged asthmatics had a non-ED follow-up within 7 days and approximately 50% by 3 weeks. This interval suggests that clinical reassessment, reinforcement of asthma education messaging and medication adjustment are too infrequent for optimal asthma control. “Asthma centres” where patients receives education, reassessment and overall management have been proposed as an alternative model for chronic disease management, but there is limited evidence regarding their cost-effectiveness.

This study has several limitations. The employed databases are unable to capture all cases of acute asthma visits to the health care system and do not provide a “true” incidence of disease. Many individuals may present to non-ED settings such as walk-in clinics for acute asthma, so these data underrepresent rates of acute asthma visits. We are not able to ascertain why individuals presented at EDs outside their region of residence. The observed patterns may also be indicative of differences in emergency service delivery and not systematic differences in the distribution of the disease. Higher rates of ED visits may reflect a preference for emergency services over other medical services in some regions. Moreover, the clinical treatment received in the EDs across Alberta is likely to vary considerably, affecting admission and relapse rates. Despite these limitations, the ACCS data have been shown to be valid and reliable.13-15 Asthma is a common presenting problem in Alberta EDs, and the interregional variation requires further study to understand the factors associated with these patterns. The most impressive findings of this study are the rates of presentation and the disparities in presentations based on age and sex, and the low proportions of post-ED follow-up.

Disclaimer: This study is based in part on data provided by Alberta Health and Wellness. The interpretation and conclusions contained herein are those of the researchers and do not necessarily represent the views of the Government of Alberta. Neither the government nor Alberta Health and Wellness express any opinion in relation to this study.