Alka B. Patel

Is there an association between spatial access to parks/green space and childhood overweight/obesity in Calgary, Canada?

BackgroundThe recent increase in childhood obesity is expected to add significantly to the prevalence of chronic diseases. We used multivariate multilevel analysis to examine associations between parks/green space and childhood overweight/obesity across communities in Calgary, Canada, a city characterized by intensified urban sprawl and high car use.MethodsBody Mass Index was calculated from measured height and weight data obtained from 6,772 children (mean age = 4.95 years) attending public health clinics for pre-school vaccinations. Each childs home postal code was geocoded using ESRI ArcGIS 9.2. We examined four measures of spatial access to parks/green space (based on Geographic Information Systems): 1) the number of parks/green spaces per 10,000 residents, 2) the area of parks/green space as a proportion of the total area within a community, 3) average distance to a park/green space, and 4) the proportion of parks/green space service area as a proportion of the total area within a community. Analyses were adjusted for dissemination area median family income (as a proxy for an individual childs family income) community-level education, and community-level proportion of visible minorities.ResultsIn general, parks/green space at the community level was not associated with overweight/obesity in Calgary, with the exception of a marginally significant effect whereby a moderate number of parks/green spaces per 10,000 residents was associated with lower odds of overweight/obesity. This effect was non-significant in adjusted analyses.ConclusionOur null findings may reflect the popularity of car travel in Calgary, Canada and suggest that the role built environment characteristics play in explaining health outcomes may differ depending on the type of urban environment being studied.

Postoperative complications and emergent readmission in children and adults with inflammatory bowel disease who undergo intestinal resection: a population-based study.

Background:Although the nature and frequency of postoperative complications after intestinal resection in patients with inflammatory bowel disease have been previously described, short-term readmission has not been characterized in population-based studies. We therefore assessed the risk of postoperative complications and emergent readmissions after discharge from an intestinal resection. Methods:We used a Canadian provincial-wide inpatient hospitalization database to identify 2638 Crohns disease (CD) and 559 ulcerative colitis (UC) admissions with intestinal resection from 2002 to 2011. We identified the cumulative risk of in-hospital complication and emergent readmission within 90 days after discharge along with predictors for both outcomes using a Poisson regression for binary outcomes. Results:The cumulative risks of in-hospital postoperative complications and 90-day emergent readmission were 23.8% and 12.6%, respectively in CD and 33.3% and 11.1%, respectively in UC. The predictors for in-hospital postoperative complications for CD and UC included older age, comorbidities, and open laparatomy for CD, additional predictors included emergent admission, stoma surgery, and concurrent resection of both small and large bowel. The predictors for 90-day readmission for CD included a postoperative complication (risk ratio, 1.61; 95% confidence interval, 1.30–2.01), emergent admission (risk ratio, 1.39; 95% confidence interval, 1.12–1.73), and stoma formation (risk ratio, 1.49; 95% confidence interval, 1.15–1.93) at the hospitalization requiring surgery. Conclusions:Readmission and postoperative complications are common after intestinal resection in CD and UC. Clinicians should closely monitor surgical patients who required emergent admission, undergo surgery with stoma formation, or develop in-hospital postoperative complications to anticipate need for readmission or interventions to prevent readmission.

Determining geographic areas and populations with timely access to cardiac catheterization facilities for acute myocardial infarction care in Alberta, Canada

BackgroundThis study uses geographic information systems (GIS) as a tool to evaluate and visualize the general accessibility of areas within the province of Alberta (Canada) to cardiac catheterization facilities. Current American and European guidelines suggest performing catheterization within 90 minutes of the first medical contact. For this reason, this study evaluates the populated places that are within a 90 minute transfer time to a city with a catheterization facility. The three modes of transport considered in this study are ground ambulance, rotary wing air ambulance and fixed wing air ambulance.MethodsReference data from the Alberta Chart of Call were interpolated into continuous travel time surfaces. These continuous surfaces allowed for the delineation of isochrones: lines that connect areas of equal time. Using Dissemination Area (DA) centroids to represent the adult population, the population numbers were extracted from the isochrones using Statistics Canada census data.ResultsBy extracting the adult population from within isochrones for each emergency transport mode analyzed, it was found that roughly 70% of the adult population of Alberta had access within 90 minutes to catheterization facilities by ground, roughly 66% of the adult population had access by rotary wing air ambulance and that no population had access within 90 minutes using the fixed wing air ambulance. An overall understanding of the nature of air vs. ground emergency travel was also uncovered; zones were revealed where the use of one mode would be faster than the others for reaching a facility.ConclusionCatheter intervention for acute myocardial infarction is a time sensitive procedure. This study revealed that although a relatively small area of the province had access within the 90 minute time constraint, this area represented a large proportion of the population. Within Alberta, fixed wing air ambulance is not an effective means of transporting patients to a catheterization facility within the 90 minute time frame, though it becomes advantageous as a means of transportation for larger distances when there is less urgency.

A validation of ground ambulance pre-hospital times modeled using geographic information systems

BackgroundEvaluating geographic access to health services often requires determining the patient travel time to a specified service. For urgent care, many research studies have modeled patient pre-hospital time by ground emergency medical services (EMS) using geographic information systems (GIS). The purpose of this study was to determine if the modeling assumptions proposed through prior United States (US) studies are valid in a non-US context, and to use the resulting information to provide revised recommendations for modeling travel time using GIS in the absence of actual EMS trip data.MethodsThe study sample contained all emergency adult patient trips within the Calgary area for 2006. Each record included four components of pre-hospital time (activation, response, on-scene and transport interval). The actual activation and on-scene intervals were compared with those used in published models. The transport interval was calculated within GIS using the Network Analyst extension of Esri ArcGIS 10.0 and the response interval was derived using previously established methods. These GIS derived transport and response intervals were compared with the actual times using descriptive methods. We used the information acquired through the analysis of the EMS trip data to create an updated model that could be used to estimate travel time in the absence of actual EMS trip records.ResultsThere were 29,765 complete EMS records for scene locations inside the city and 529 outside. The actual median on-scene intervals were longer than the average previously reported by 7–8 minutes. Actual EMS pre-hospital times across our study area were significantly higher than the estimated times modeled using GIS and the original travel time assumptions. Our updated model, although still underestimating the total pre-hospital time, more accurately represents the true pre-hospital time in our study area.ConclusionsThe widespread use of generalized EMS pre-hospital time assumptions based on US data may not be appropriate in a non-US context. The preference for researchers should be to use actual EMS trip records from the proposed research study area. In the absence of EMS trip data researchers should determine which modeling assumptions more accurately reflect the EMS protocols across their study area.

Validity and utility of ICD-10 administrative health data for identifying ST- and non-ST-elevation myocardial infarction based on physician chart review

BACKGROUND Health care administrative databases are useful for assessing the population-level burden of disease and examining issues related to access, costs and quality of care. In these databases, the diagnoses and procedures are coded with the use of the World Health Organization International Classification of Diseases (ICD). We examined the validity of 2 ICD-10 coding definitions for categorizing patients with acute myocardial infarction (MI) as having ST-elevation MI (STEMI) or non-ST-elevation MI (non-STEMI). METHODS Charts of patients with acute MI discharged between April and June 2007 from 3 hospitals in Edmonton, were reviewed to define the acute MI subtype (i.e., STEMI v. non-STEMI). The agreement between clinician chart review and STEMI/non-STEMI classification based on the standard (ICD-10 I21.x) and the supplementary electrocardiogram (ECG) codes (R94.3x) was determined. We assessed the effect of these alternative definitions on in-hospital mortality estimates by applying them to the data for all patients with acute MI admitted to hospital in the province from April 2007 to March 2010. RESULTS Of the 297 patients, 49.2% were identified as having STEMI based on chart review, 44.4% using the standard definition, and 44.1% using the ECG definition. Both the standard and ECG definitions provided high agreement (92% for STEMI and 100% for non-STEMI) with the chart review classification. In the larger population-level cohort (n = 15 148), use of the standard definition or the ECG definition did not affect in-hospital mortality estimates for patients with STEMI and those with non-STEMI. INTERPRETATION The standard definition appears equivalent to the definition using supplementary ECG codes to subcategorize patients with acute MI as having STEMI or non-STEMI. These findings may be relevant for the development of later versions of ICD codes.

Association Between Diabetic Ketoacidosis Hospitalizations and Driving Distance to Outpatient Diabetes Centres in Adults with Type 1 Diabetes Mellitus

OBJECTIVE The purpose of this study was to assess the relationship between diabetic ketoacidosis (DKA) hospitalization and driving distance from home to outpatient diabetes care in adults with type 1 diabetes mellitus. METHODS We identified adults with type 1 diabetes using clinical and administrative databases living in Calgary, Alberta. The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, codes were used to identify DKA hospitalizations, and geographic information systems were used to obtain road distance. Multivariate logistic regression was used to assess the association between driving distance (exposure) to diabetes care sites and the outcome of DKA hospitalization. RESULTS We identified 1467 patients (151 patients with DKA) with type 1 diabetes. Patients with DKA hospitalizations were younger (35.6 vs. 41.0 years), had shorter duration of diabetes (13.6 vs. 18.7 years) and higher glycated hemoglobin (9.2% vs. 8.4%). Driving distance from home to diabetes centre 1 (adjusted odds ratio 1.02 per 1 km; 95% confidence interval, 0.96 to 1.07), diabetes centre 2 (adjusted odds ratio 1.01; 95% confidence interval, 0.99 to 1.04) or closest general practitioner (adjusted odds ratio 0.9; 95% confidence interval, 0.63 to 1.25) was not associated with DKA hospitalization. Driving distance was also not associated with glycemic control. CONCLUSIONS Within a large urban city, driving distance to diabetes centres does not appear to be protective of DKA hospitalization. However, this work does not preclude the role of local travel distance and diabetes outcomes. More research is required to explore the role of other individual, neighbourhood and community factors that influence DKA hospitalization.

Using Geographic Information Systems for Health Research

The connection between public health and geography can be traced back to Hippocrates (c. 400 BC) who deduced that spatially varying factors such as climate, elevation, environmental toxins, ethnicity and race contributed to the spatial patterns of illness (Parchman et al., 2002). The observations of Hippocrates still hold true today and these relationships between geography and disease have allowed geospatial methods to become valuable within the field of public health. Maps have long been a useful tool for visualizing patterns in health care. One of the earliest and most commonly cited examples is from the mid 1800s when Dr. John Snow deduced the source of a cholera outbreak in London based on a simple visualization of the incidents of cholera in relation to water pumps (Johnson, 2007). Although visualizing data geographically is still very valuable for uncovering patterns and associations over space, geospatial analysis has become more sophisticated over time.

Drip and ship versus direct to endovascular thrombectomy: The impact of treatment times on transport decision-making:

Introduction In ischaemic stroke care, fast reperfusion is essential for disability free survival. It is unknown if bypassing thrombolysis centres in favour of endovascular thrombectomy (mothership) outweighs transport to the nearest thrombolysis centre for alteplase and then transfer for endovascular thrombectomy (drip-and-ship). We use conditional probability modelling to determine the impact of treatment times on transport decision-making for acute ischaemic stroke. Materials and methods Probability of good outcome was modelled using a previously published framework, data from the Irish National Stroke Register, and an endovascular thrombectomy registry at a tertiary referral centre in Ireland. Ireland was divided into 139 regions, transport times between each region and hospital were estimated using Google’s Distance Matrix Application Program Interface. Results were mapped using ArcGIS 10.3. Results Using current treatment times, drip-and-ship rarely predicts best outcomes. However, if door to needle times are reduced to 30 min, drip-and-ship becomes more favourable; even more so if turnaround time (time from thrombolysis to departure for the endovascular thrombectomy centre) is also reduced. Reducing door to groin puncture times predicts better outcomes with the mothership model. Discussion This is the first case study modelling pre-hospital transport for ischaemic stroke utilising real treatment times in a defined geographic area. A moderate improvement in treatment times results in significant predicted changes to the optimisation of a national acute stroke patient transport strategy. Conclusions Modelling patient transport for system-level planning is sensitive to treatment times at both thrombolysis and thrombectomy centres and has important implications for the future planning of thrombectomy services.

Variation in Maternal Co-morbidities and Obstetric Interventions across Area-Level Socio-economic Status: A Cross-Sectional Study

BACKGROUND Multiple studies indicate a significant association between area-level socio-economic status (SES) and adverse maternal health outcomes; however, the impact of area-level SES on maternal co-morbidities and obstetric interventions has not been examined. OBJECTIVE To examine the variation in maternal co-morbidities and obstetric interventions across area-level SES. METHODS This study used data from the Discharge Abstract Database that comprised birth data in Alberta between 2005-2007 (n = 120 285). Co-morbidities and obstetric interventions were identified using validated case-definitions. Material deprivation index was obtained for each dissemination area through linkage of hospitalisation and census data. Multilevel logistic regression was used to analyse the data adjusting for potential confounding variables. RESULTS The prevalence of any co-morbidity varied across area-level SES. Drug abuse odds ratio (OR) 2.5 (95% confidence interval (CI) 1.8, 3.5), pre-existing diabetes OR 1.7 (95% CI 1.1, 2.6), and prolonged hospital stay OR 1.5 (95% CI 1.4, 1.6) were significantly more likely to occur in the most deprived areas compared to the least deprived areas. In contrast, caesarean delivery OR 0.9 (95% CI 0.8, 0.9) was less likely to occur in the most deprived areas compared to the least deprived areas. Area-level deprivation explained area-level variance of drug abuse, HIV, and other mental diseases only. CONCLUSION Many co-morbidities and obstetric interventions vary at the area-level, but only some are associated with area-level SES, and few of them vary due to the area-level SES. This indicates that other area-level factors, in addition to area-level SES, need to be considered when investigating maternal health and use of health interventions.

Geographic Clustering of Acute Complications and Sociodemographic Factors in Adults with Type 1 Diabetes

OBJECTIVES To assess the geographic distribution of acute complications in patients with type 1 diabetes in a large urban centre; and to assess the association between acute complications and community-level sociodemographic factors. METHODS Adults (aged ≥18 years old) with type 1 diabetes and acute complications were identified between 2004 and 2008 by using a diabetes centre clinical database or discharge abstracts for acute complications (diabetic ketoacidosis or hypoglycemia). Using a geographic information system, hot-spot analysis was used to identify spatial clusters of acute complications in a large urban centre. The association between acute complications and community-level sociodemographic factors were assessed by Spearman rank correlation. RESULTS We identified 1779 patients with type 1 diabetes, of whom 456 had been hospitalized for acute complications. The mean age of patients was 40.9±16.0 years, and men were more likely to have acute complications (59.2% vs. 52.3%; p<0.01). Spatial clusters of high values and low values were identified. Higher median family income (r=-0.36; p<0.0001) and higher education levels (r=-0.30; p<0.0001) were associated with lower rates of acute complications. CONCLUSIONS This study demonstrated geographic clusters of hospitalizations for acute complications and important community sociodemographic factors. Prevention strategies and interventions targeting these geographic and sociodemographic disparities need to be explored as a means of minimizing hospitalizations for acute complications.