Manitoba Research Articles

Inflammatory Bowel Disease in Children with Elevated Serum Gamma Glutamyltransferase Levels

To assess the characteristics of inflammatory bowel disease and disease prognosis among children with elevated gamma glutamyltransferase (GGT) and primary sclerosing cholangitis (PSC)-ulcerative colitis (UC). Our longitudinal, population-based cohort comprised all children and young adults diagnosed with UC in the Canadian province of Manitoba between 2011 and 2018. Diagnosis of PSC was confirmed based on a combination of cholestatic biochemical markers and cholangiographic features. The Fisher exact test with Bonferroni correction was used to examine the relationship between categorical variables. We enrolled 95 children with UC/Inflammatory bowel disease-unclassified with a median age at diagnosis of 14years (IQR: 10.4-15.9years) and 1399 person-years follow-up. Among them, 9 children developed PSC-UC, with an incidence rate of 6.43 new cases per 1000 person-years. In this cohort, 8 (72.7%) of 11 children with high baseline serum GGT levels developed PSC-UC in comparison with 1 (1.2%) of 84 children with normal serum GGT levels at baseline (P<.001). All children with high serum GGT levels at diagnosis had pancolitis in comparison with 63.9% of children with normal serum GGT levels (P=.01). Children with high serum GGT levels were more likely to be perinuclear neutrophil antibodies-positive than those with normal levels (90.9% vs 52.0%, P=.01). Our findings indicated that pediatric patients with UC and with even mild elevations of serum GGT levels, especially at baseline, might be predisposed to develop PSC.

Fracture prediction from self-reported falls in routine clinical practice: a registry-based cohort study.

There is conflicting evidence from research cohorts that falls independently increase fracture risk. We examined the independent effects of falls on subsequent fractures in a large clinical registry of bone mineral density (BMD) results for the Province of Manitoba, Canada that has been systematically collecting self-reported falls information since September 1, 2012. The study population consisted of 24,943 women and men aged 40years and older (mean age 65.5 ± 10.2years) with fracture probability assessment (FRAX), self-reported falls for the previous year (categorized as none, 1, 2, or >3) and fracture outcomes. Adjusted hazard ratios (HR) with 95 confidence intervals (CI) for time to fracture were estimated using Cox proportional hazards models. During mean observation time of 2.7 ± 1.0years, 863 (3.5%) sustained one or more major osteoporotic fractures (MOF), 212 (0.8%) sustained a hip fracture, and 1210 (4.9%) sustained any incident fracture. Compared with no falls in the previous year (referent), there was a gradient of increasing risk for fracture with increasing number of falls (all P < 0.001). Results showed minimal attenuation with covariate adjustment. When adjusted for baseline fracture probability (FRAX score with BMD) the HR for MOF increased from 1.49 (95% CI 1.25-1.78) for one fall to 1.74 (1.33-2.27) for two falls to 2.62 (2.06-3.34) for ≥ 3 falls. HRs were similar for any incident fracture and slightly greater for prediction of hip fracture, reaching 3.41 (95% CI 2.19-5.31) for ≥ 3 previous falls. Self-report number of falls in the previous year is strongly associated with incident fracture risk in the routine clinical practice setting, and this risk is independent of age, sex, BMD, and baseline fracture probability. Moreover, there is dose-response with multiple falls (up to a maximum of 3) conferring greater risk than a single fall.

Field-Scale Crop Seeding Date Estimation from MODIS Data and Growing Degree Days in Manitoba, Canada

Information on crop seeding date is required in many applications; such as crop management and yield forecasting. This study presents a novel method to estimate crop seeding date at the field level from time-series 250-m Moderate Resolution Imaging Spectroradiometer (MODIS) data and growing degree days (GDD; base 5 °C; °C-days). The start of growing season (SOS) was first derived from time-series EVI2 (two-band Enhanced Vegetation Index) calculated from a MODIS 8-day composite surface reflectance product (MOD09Q1; Collection 6). Based on GDD; calculated from the Daymet gridded estimates of daily weather parameters; a simple model was developed to establish a linkage between the observed seeding date and the SOS. Calibration and validation of the model was conducted on three major crops; spring wheat; canola and oats; in the Province of Manitoba; Canada. The estimated SOS had a strong linear correlation with the observed seeding date; with a deviation of a few days depending on the year. The seeding date of the three crops can be calculated from the SOS by adjusting the number of days needed to accumulate GDD (AGDD) for emergence. The overall root-mean-square-difference (RMSD) of the estimated seeding date was less than 10 days. Validation showed that the accuracy of the estimated seeding date was crop-type independent. The developed method is useful for estimating the historical crop seeding date from remote sensing data in Canada; to support studies of the interactions among seeding date; crop management and crop yield under climate change. It is anticipated that this method can be adapted to other crops in other locations using the same or different satellite data.

Trends of Utilization of Tumor Necrosis Factor Antagonists in Children With Inflammatory Bowel Disease: A Canadian Population-Based Study.

Population-based studies examining the prevalence of anti-tumor necrosis factor (anti-TNF) antagonist utilization in children and young adults with inflammatory bowel disease (IBD) are lacking. We aimed to describe the trend of anti-TNF utilization in pediatric IBD over time. Survival analyses were performed for all patients diagnosed with IBD before age 18 years in the province of Manitoba to determine the time from diagnosis to first anti-TNF prescription in different time eras (2005-2008, 2008-2012, 2012-2016). There were 291 persons diagnosed with IBD (157 with Crohn's disease [CD] and 134 with ulcerative colitis [UC]) over the study period. The likelihood of being initiated on an anti-TNF by 1, 2, and 5 years postdiagnosis was 18.4%, 30.5%, and 42.6%, respectively. The proportion of persons aged <18 years utilizing anti-TNFs rose over time; in 2010, 13.0% of CD and 4.9% of UC; by 2016, 60.0% of CD and 25.5% of UC. For those diagnosed after 2012, 42.5% of CD and 28.4% of UC patients had been prescribed an anti-TNF antagonist within 12 months of IBD diagnosis. Initiating an anti-TNF without prior exposure to an immunosuppressive agent increased over time (before 2008: 0%; 2008-2012: 18.2%; 2012-2016: 42.8%; P < 0.001). There was a significant reduction in median cumulative dose of corticosteroids (CS) in the year before anti-TNF initiation (2005-2008: 4360 mg; 2008-2012: 2010 mg; 2012-2016: 1395 mg prednisone equivalents; P < 0.001). Over a period of 11 years, anti-TNFs are being used earlier in the course of pediatric IBD, with a parallel reduction in the cumulative CS dose.

A deep learning framework for land-use/land-cover mapping and analysis using multispectral satellite imagery

In this article, we present an approach to land-use and land-cover (LULC) mapping from multispectral satellite images using deep learning methods. The terms satellite image classification and map production, although used interchangeably have specific meanings in the field of remote sensing. Satellite image classification describes assignment of global labels to entire scenes, whereas LULC map production involves producing maps by assigning a class to each pixel. We show that by classifying each pixel in a satellite image into a number of LULC categories we are able to successfully produce LULC maps. This process of LULC mapping is achieved using deep neural networks pre-trained on the ImageNet large-scale visual recognition competition datasets and fine-tuned on our target dataset, which consists of Landsat 5/7 multispectral satellite images taken of the Province of Manitoba in Canada. This approach resulted in 88% global accuracy. Performance was further improved by considering the state-of-the-art generative adversarial architecture and context module integrated with the original networks. The result is an automated deep learning framework that can produce highly accurate LULC maps images significantly faster than current semi-automated methods. The contribution of this article includes extensive experimentation of different FCN architectures with extensions on a unique dataset, high classification accuracy of 90.46%, and a thorough analysis and accuracy assessment of our results.

Latent tuberculosis treatment completion rates from prescription drug administrative data.

In the province of Manitoba, Canada, given that latent tuberculosis infection (LTBI) treatment is provided at no cost to the patient, treatment completion rates should be optimal. The objective of this study was to estimate LTBI treatment completion using prescription drug administrative data and identify patient characteristics associated with completion. Prescription drug data (1999-2014) were used to identify individuals dispensed isoniazid (INH) or rifampin (RIF) monotherapy. Treatment completion was defined as being dispensed INH for ≥ 180days (INH180) or ≥ 270days (INH270) or RIF for ≥ 120days (RIF120). Logistic regression models tested socio-demographic and comorbidity characteristics associated with treatment completion. The study cohort comprised 4985 (90.4%) persons dispensed INH and 529 (9.6%) RIF. Overall treatment completion was 60.2% and improved from 43.1% in 1999-2003 to 67.3% in 2009-2014. INH180 showed the highest completion (63.8%) versus INH270 (40.4%) and RIF120 (27.0%). INH180 completion was higher among those aged 0-18years (68.5%) compared with those aged 19+ (61.0%). Sex, geography, First Nations status, income quintile, and comorbidities were not associated with completion. Benchmark 80% treatment completion rates were not achieved in Manitoba. Factors associated with non-completion were older age, INH270, and RIF120. Access to shorter LTBI treatments, such as rifapentine/INH, may improve treatment completion.

Tuberculosis among northern Manitoba First Nations, 2008-2012: program performance on- and off-reserve.

The objectives of this study were to:(1) report tuberculosis (TB) program performance for northern First Nations in the province of Manitoba; (2) present methods for TB program performance measurement using routinely collected surveillance data; and (3) advance dialogue on performance measurement of Canadian TB programs. Data on a retrospective cohort of people diagnosed with TB in Manitoba between January 1, 2008 and December 31, 2010, and their contacts,were extracted from theManitoba TB Registry. Performance measures based on US-CDC were analyzed. Adjusted probability ratios (aPR) and 95% confidence intervals (CIs) werereported with comparisons between on-/off-reserve First Nations, adjusted for age, sex, and treatment history. A cohort of n = 149 people diagnosed with TB and n = 3560 contactswereidentified. Comparisons off-/on-reserve: Treatment completion (aPR = 1.03; 95% CI 0.995-1.07); early detection (aPR = 0.87; 95% CI 0.57-1.33); HIV testing and reporting (aPR = 0.42; 95% CI 0.21-0.83); pediatric TB (age < 15years) (aPR = 1.20; 95% CI 0.47-3.06); retreatment for TB (aPR = 0.93; 95% CI 0.89-0.97); contact elicitation (aPR = 0.94; 95% CI 0.84-1.05); contact assessment (aPR = 0.69; 95% CI 0.50-0.94). Pediatric (ages < 15years) TB incidence in northern Manitoba was 37.1 per 100,000/year. TB program performance varies depending on residence in a reserve or non-reserve community. Action is urgently needed to address TB program performance in terms of contact investigation and HIV testing/reporting for First Nations off-reserve and to address high rates of pediatric TB in northern Manitoba. First Nations collaboration and models of care should be considered both on- and off-reserve to improve TB program performance.

Differential item functioning of the SF-12 in a population-based regional joint replacement registry

BackgroundJoint replacement, an increasingly common procedure amongst older adults, can substantially improve health-related quality of life (HRQoL). However, differential item functioning (DIF) may affect the accurate interpretation of differences in HRQoL amongst patients with different demographic and health status characteristics but the same underlying (i.e., latent) level of the investigated construct. This study tested for DIF in pre-operative SF-12 physical health (PH) and mental health (MH) sub-scale items amongst patients undergoing total hip arthroplasty (THA) and total knee arthroplasty (TKA).MethodsData were from a population-based joint replacement registry from the Canadian province of Manitoba. TKA and THA patients who had surgery between 2009 and 2015 and completed a pre-operative assessment were included. DIF was tested using the multiple indicators multiple causes (MIMIC) method with sex, age group, body weight status, and presence of multiple comorbid conditions (i.e., multimorbidity) as covariates. Analyses were stratified by joint type.ResultsThe study cohort included 8820 patients; 42.1% underwent THA, 57.3% were female, 32.7% were 70+ years, and 52.8% were obese. For each sub-scale, four of the six items exhibited DIF in both THA and TKA groups. Differences in the covariate effect estimates for DIF and No-DIF models on the MH latent variable were largest for age and body weight status for the THA group, and for sex and multimorbidity for the TKA group. All of the differences were small for PH. Multimorbidity had the strongest association with PH and age and sex had the strongest association with MH in the DIF models.ConclusionsDemographic and health status characteristics influenced SF-12 PH and MH item responses in joint replacement populations, although the size of the effects were not large for PH. We recommend testing and adjusting for DIF effects to ensure comparability of HRQoL measures in joint replacement populations.

PF282 THE EFFECT OF MODERN THERAPEUTICS IN ADULT ACUTE MYELOID LEUKEMIA (AML): POPULATION‐BASED ANALYSIS OF LONG‐TERM OUTCOMES FROM THE MANITOBA CANCER REGISTRY

Background:While the prognosis for adults with acute myeloid leukemia (AML) is historically poor, advances in supportive care and new therapeutics such as blood and marrow transplant (BMT) and hypomethylating agents may have improved overall outcomes. In the Canadian province of Manitoba, all patients with AML are managed at a single cancer centre with support from a community cancer program. Clinical and demographic data are collected in a provincial cancer registry. This registry captures all cases of AML, increasing the reliability of population‐based data as case ascertainment and follow‐up follow a uniform approach.Aims:We sought to describe the clinical characteristics, incidende and relative survival for patients with AML in the province of Manitoba, with attention to temporal trends and outcomes in rural patients, where access to care may be challenging.Methods:We collected data on consecutively diagnosed adult cases of AML that were reported to the Manitoba Cancer Registry (MCR) from 1990 ‐ 2014. Life tables from Statistics Canada were used as a reference for expected survival in the general population. Patient demographics were stratified by age, time period and residence at diagnosis. Two‐year age‐standardized relative survival estimates of patients in Manitoba with AML were estimated using the cohort method within various subcategories, including time period of diagnosis (in 5 year increments), sex and urban (Winnipeg) versus non‐Winnipeg location.Results:A total of 910 patients from the ages of 15 to 99 were diagnosed from 1990 ‐ 2014 with available survival data to 2016. Of the 910 patients, 517 were men and 393 were women and 511 (56%) were residents of Winnipeg at diagnosis. Two‐year crude relative survival estimates by time period and age group demonstrated improving relative survival for each age quintile category within each 5‐year time period, with the most notable increase over time in patients aged 15–44 and 45–54. Total two‐year age‐standardized relative survival for all patients was 20.37% (CI 95% 17.81–23.05%). Two‐year age‐standardized relative survival in males was 19.22% (CI 95% 15.95–22.73%) and 21.93% (CI 95% 17.88–26.25%) in females. Two‐year age‐standardized relative survival by residence at diagnosis was 20.32% (CI 95%. 16.92–23.94%) for urban (Winnipeg) residents, and 20.54% (CI 95% 16.74–24.62%) for those outside of Winnipeg.Summary/Conclusion:We demonstrate improved relative survival in adult AML over time, most notably in adults aged 15–54, where BMT has played an increasingly important therapeutic role. Advanced supportive care and hypomethylating agents may also have contributed to improved outcomes, especially in older patients. Similar relative survival for urban and rural patients supports the benefits of a centralized and coordinated leukemia centre with its closely linked community cancer program.

Some observations on BWIM data collected in Manitoba

Three highway bridges in the Canadian province of Manitoba are being monitored continuously not only for their long-term performance but also for bridge weighing-in-motion (BWIM). Data collected for the BWIM study has led to some observations that have far-reaching consequences about the design and evaluation loads for highway bridges. This paper presents the well-known concept of equivalent base length, Bm, as a useful tool for comparing trucks with different axle weight and spacing configurations as they influence load effects in all bridges. It is discussed that the statistics of gross vehicle weights (GVWs), W, collected over a one-month period is not significantly different from that for the GVW data collected over a longer period. A rational method concludes that the value of W for the CL-W Truck, the design live load specified by the Canadian Highway Bridge Design Code, is 555 kN for Manitoba. The observed truck data in Manitoba presented on the W–Bm space is found to be similar to that collected in the Canadian province of Ontario more than four decades ago. It was also found that the multi-presence factors, accounting for the presence of side-by-side trucks in two-lane bridges, specified in North American bridge design and evaluation codes are somewhat conservative.

Incidence and outcome of interval breast cancer among women participating in the provincial population based screening program in Manitoba, Canada.

6595 Background: The province of Manitoba, Canada has an organized population based biennial mammographic screening program. Here we report outcomes of women diagnosed with Interval Cancers (IC), defined as cancer diagnosed within 24 months of a normal screening mammogram and before the next screening mammogram, compared to Screen Detected (SD) cancers. Methods: The Manitoba Cancer Registry was used to obtain data about tumor and host characteristics and cause-specific mortality for women 52 to 64 years of age diagnosed with invasive breast cancer from January 2004 to December 2010. Lead time bias in SD cancers was adjusted based on Duffy's correction factor. Competing risk analysis was used to examine the risk of death by cancer detection method. To examine the relationship between breast cancer detection type and personal and tumour characteristics, we performed multinomial logistic regression analysis with age, area-level income quintile, tumour grade, hormone receptor, and HER2 receptor as independent variables. Results: There were 5,884 women diagnosed with invasive breast cancer during the study period of which 1,338 were SD, 362 were IC, and the remainder were diagnosed outside the screening program or were non-compliant to screening. IC were more likely than SD cancers to be high grade [Odds Ratio (OR) 3.8, 95% Confidence Interval (CI): 2.1-6.8], and ER negative [OR 1.7, 95% CI: 1.02-3.12]. At data cut-off date of June 30, 2012, risk of death from breast cancer was significantly higher for IC compared to SD cancers [Hazard Ratio (HR) 4.18, 95% CI: 1.97-8.87)], for sojorn time (period when tumour is asymptomatic but screen detectable) of 2 years adjusting for area-level average income quintile and age. Sensitivity analyses with sojourn times of 1, 3, and 4 years showed similar results. Risk of non-breast cancer death was not increased with IC compared to SD cancers (HR 1.33, 95% CI: 0.43-4.15). Conclusions: Among women who participated in a systematic population-based screening program, IC occurred frequently; breast cancer-related death for IC was 4-times that of SD cancers. These results highlight the discordance between the principles underpinning population based breast cancer screening and natural history of these more lethal breast cancers.

Sa1829 – Diet Stability in Persons with Ibd: the Manitoba Living with IBD Study

Sa1829 – Diet Stability in Persons with Ibd: the Manitoba Living with IBD Study

Gliadin genotypes worldwide for spring wheats (Triticum aestivum L.) 2. Strong differentiation of polymorphism between countries and regions of origin

Genotypes of 290 cultivars from ten countries (four continents) of spring wheat Triticum aestivum, were analyzed through the identification of alleles at the gliadin-encoding (Gli) loci. The group of cultivars bred in one country during the 20th century might be characterized by its level of genetic diversity (H) and a specific set of alleles. Strong differences between genotypes of cultivars bred in different countries were confirmed by cluster analysis of genetic distances. Genotypes of cultivars bred in regions of the same country might differ, but they always composed a single cluster and strongly differed from genotypes from other countries or regions. Therefore, world polymorphism of spring wheat germplasm of the 20th century is highly structured and differentiated between countries and their regions. This differentiation seems to be mainly a result of natural selection acting in different ways in different eco-climatic conditions of breeding and propagation. At a regional scale, statistically significant temporal changes in frequency of some gliadin alleles and a tendency to temporal decrease of overall genetic diversity were discovered in cultivars bred in each of seven regions studied in Australia, Canada and the former USSR. In regions of Canada (provinces of Manitoba and Ontario) and of the former USSR (Saratov), a statistically significant decay of the genetic diversity of cultivars bred during the 20th century was discovered. However, the influence of breeders' selection for end-use wheat quality on the wheat genetic diversity was the least, if any.

Which is the preferred site for bone mineral density monitoring as an indicator of treatment-related anti-fracture effect in routine clinical practice? A registry-based cohort study.

The role of monitoring bone mineral density (BMD) as an indicator of an anti-fracture effect is controversial. Discordance between the spine and hip BMD is common and creates uncertainty in clinical practice. Using a population-based BMD Registry for the Province of Manitoba, Canada, we compared change in the spine and hip BMD as an indicator of treatment-related fracture risk reduction. The study cohort included 6093 women age >40years initiating osteoporosis treatment with two consecutive dual-energy X-ray absorptiometry (DXA) scans (mean interval 4.7years). We computed change in the spine, total hip, and femur neck BMD between the first and second DXA scans as categorical (categorized as stable, detectable decrease, or detectable increase) and continuous measures. We modeled time to first incident fracture, ascertained from health services data, using Cox regression adjusted for baseline fracture probability. During a mean follow-up of 12.1years, 995 women developed incident major osteoporotic fractures (MOF) including 246 with hip fractures and 301 with clinical vertebral fractures. Women with a detectable decrease in total hip BMD compared with stable BMD experienced an increase in MOF (adjusted hazard ratio [aHR] 1.46, 95% confidence interval [CI] 1.25-1.70) while those with a detectable increase in total hip BMD experienced a decrease in MOF (aHR 0.71, 95% CI 0.61-0.83), and these results were not attenuated when adjusted for change in spine BMD. Similar results were seen for hip and clinical vertebral fracture outcomes, when BMD change was assessed as a continuous measure, and when femur neck BMD monitoring was used instead of total hip BMD monitoring. Treatment-related increases in total hip BMD are associated with lower MOF, hip, and clinical vertebral fracture risk compared with stable BMD, while BMD decreases are associated with higher fracture risk. In contrast, spine BMD change is not independently associated with fracture risk.

Primary care reform in Manitoba, Canada, 2011–15: Balancing accountability and acceptability

Primary care reform cannot succeed without substantive change on the part of providers. In Canada, these are mostly fee-for-service physicians, who tend to regard themselves as independent professionals and not under managerial sway. Hence, policymakers must balance two conflicting imperatives: ensuring the acceptability of renewal efforts to these physicians while enforcing their accountability for defined actions or outcomes. In its 2011–15 strategy to improve access to primary care, the province of Manitoba introduced several linked initiatives, each striving to blend acceptability- and accountability-promoting elements. Clearly delimited initiatives that directly promoted a specific observable behaviour (accountability) through financial or non-financial support (acceptability) were most successfully implemented. System-wide initiatives with complicated designs (notably a primary care network model that established formal partnership among clinics and regional health authorities) encountered greater difficulties in recruiting and sustaining physician participation. Although such initiatives offered physicians considerable decision-making latitude (acceptability), many physicians questioned the meaningfulness of opportunities for voice within a predetermined structure (accountability). Moreover, policymakers struggled to enhance the acceptability of such initiatives without sacrificing strong accountability mechanisms. Policymakers must carefully consider how acceptability and accountability elements may interact, and design them in such a way as to minimize the risk of mutual interference.

Trends in Population-Based Incidence and Prevalence of Juvenile Idiopathic Arthritis in Manitoba, Canada.

To estimate the incidence and prevalence of juvenile idiopathic arthritis (JIA) in children ages <16 years in the province of Manitoba, Canada, and to determine changes in estimates between 2000 and 2012. JIA cases were ascertained from the administrative health data of Manitoba, using a validated case-finding algorithm. Annual incidence and prevalence rates were estimated using a generalized linear model with generalized estimating equations (GEEs), adjusting for sociodemographic characteristics. Changes in estimates were tested using piecewise regression models. A total of 455 cases of JIA prevalence met the inclusion criteria. Sex- and age-adjusted incidence estimates were 14.01 (95% confidence interval [95% CI] 13.52, 14.53) in 2000/2001 and 9.18 (95% CI 8.56, 9.85) in 2010/2011. Prevalence estimates were 65.33 (95% CI 63.87, 66.83) in 2000/2001 and 59.61 (95% CI 58.17, 61.08) in 2010/2011. A linear piecewise model provided the best fit to the data. There was a significant decrease in prevalence over the study period (-0.18 [95% CI -0.35, -0.02]; P = 0.0292) but no statistically significant change in incidence (-0.46 [95% CI -0.94, 0.01]; P = 0.0571). Sex-stratified models revealed a decrease for males in both prevalence (estimate -0.54 [95% CI -0.84, -10.25]; P = 0.0003) and incidence (estimate -1.02 [95% CI -2.02, -0.04]; P = 0.0439); there were no changes for females. Few population-based longitudinal epidemiologic studies of JIA have been conducted. Our findings suggested a decrease in overall JIA prevalence, and in incidence and prevalence in men. Further research to validate these findings in other cohorts and to explore factors that contribute to this change will benefit future health care planning for JIA.

Real-world outcomes among patients (pts) treated with gemcitabine (GEM)-based therapy post-FOLFIRINOX (FFOX) failure in advanced pancreatic cancer (APC).

303 Background: Limited evidence exists for the selection of chemotherapy in APC after first-line (1stL) FFOX. Gemcitabine/nab-paclitaxel (GEMNAB) is publicly funded for second-line (2ndL) use in the provinces of Alberta (AB) and Manitoba (MB), but is not covered in British Columbia (BC). We compared population-based outcomes by region to examine the utility of 2ndL GEMNAB versus GEM alone. Methods: We identified pts treated with 1stL FFOX between 2013-2015 across BC, AB, and MB. Baseline characteristics and treatment regimens were compared between AB/MB and BC. Survival outcomes were assessed by the Kaplan-Meier, and compared with log-rank test. Results: 370 pts treated with 1stL FFOX were identified (145 AB/MB, 225 BC), with a median age of 61y, 42% female, and 68% with metastatic disease (similar in both groups). Receipt of 2ndL therapy was 49% AB/MB vs 44% BC ( p = 0.35), and time from diagnosis to 2ndL therapy measured 7.6 mos AB/MB versus 9.4 mos BC ( p = 0.1). The distribution of 2ndL gemcitabine use was: 72% GEMNAB, 23% GEM in AB/MB versus 27% GEMNAB, 66% GEM in BC ( p &lt; 0.001). Median overall survival (OS) from diagnosis was similar: 12.4 mos in AB/MB versus 10.9 mos in BC ( p = 0.75). On Cox regression analysis, region was not significant. A secondary survival analysis by 2ndL regimen demonstrated a median OS of 18.0 mos with GEMNAB versus 14.3 mos GEM ( p &lt; 0.01). Conclusions: In our population-based comparison of APC pts treated with 1stL FFOX, survival outcomes were comparable regardless of publicly funded access to 2ndL GEMNAB versus GEM. OS by regimen favored 2ndL GEMNAB, but patient selection may be largely responsible for this difference. Randomized trials are needed to demonstrate the benefit of GEMNAB post-FFOX in APC.

Automated LULC map production using deep neural networks

ABSTRACTThis article presents an approach to automating the creation of land-use/land-cover classification (LULC) maps from satellite images using deep neural networks that were developed to perform semantic segmentation of natural images. This work is important since the production of accurate and timely LULC maps is becoming essential to government and private companies that rely on them for large-scale monitoring of land resource changes. In this work, deep neural networks are trained to classify each pixel of a satellite image into one of a number of LULC classes. The presented deep neural networks are all pre-trained using the ImageNet Large-Scale Visual Recognition Competition (ILSVRC) datasets and then fine-tuned using approximately 19,000 Landsat 5/7 satellite images of resolution taken of the Province of Manitoba in Canada. The result is an automated solution that can produce LULC maps images significantly faster than current semi-automated methods. The contributions of this article are the observation that deep neural networks developed for semantic segmentation can be used to automate the task of producing LULC maps; the use of these networks to produce LULC maps; and a comparison of several popular semantic segmentation architectures for solving the problem of automated LULC map production.

Fancy Schools for Fancy People: Risks and Rewards in Fieldwork Research Among the Low German Mennonites of Canada and Mexico

In the 1920s, conflict over schooling prompted the exodus of nearly 8000 Mennonites from the Canadian prairie provinces of Manitoba and Saskatchewan to Mexico and Paraguay; this is the largest voluntary exodus of a single people group in Canadian history. Mennonites—whose roots are found in the 1520s Reformation—are an Anabaptist, pacifist, isolationist ethnic, and religious minority group, and victims of a fledgling Canada’s nation-building efforts. It is estimated that approximately 80,000 descendants of the original emigrants have subsequently returned to Canada, where tensions over schooling have persisted. The tensions—then, as now—are rooted in a fundamentally different understanding of the purposes of education—and it is this tension that interests me as an ethnographer and education researcher. My research is concerned with assessing attitudes towards education within the Low German Mennonite (LGM) community in both Canada and Mexico. Too often academic research is presented as a tidy finished product, with little insight shed into the messy, highly iterative process of data collection. The purpose of this article is to pull back the curtain and discuss the messiness of the process, including security risks involved with methodology, site selection, research participants, and gaining access to the community.

Validation of a one year fracture prediction tool for absolute hip fracture risk in long term care residents

BackgroundFrail older adults living in long term care (LTC) homes have a high fracture risk, which can result in reduced quality of life, pain and death. The Fracture Risk Scale (FRS) was designed for fracture risk assessment in LTC, to optimize targeting of services in those at highest risk. This study aims to examine the construct validity and discriminative properties of the FRS in three Canadian provinces at 1-year follow up.MethodsLTC residents were included if they were: 1) Adults admitted to LTC homes in Ontario (ON), British Columbia (BC) and Manitoba (MB) Canada; and 2) Received a Resident Assessment Instrument Minimum Data Set Version 2.0. After admission to LTC, one-year hip fracture risk was evaluated for all the included residents using the FRS (an eight-level risk scale, level 8 represents the highest fracture risk). Multiple logistic regressions were used to determine the differences in incident hip or all clinical fractures across the provinces and FRS risk levels. We examined the differences in incident hip or all clinical fracture for each FRS level across the three provinces (adjusted for age, BMI, gender, fallers and previous fractures). We used the C-statistic to assess the discriminative properties of the FRS for each province.ResultsDescriptive statistics on the LTC populations in ON (n = 29,848), BC (n = 3129), and MB (n = 2293) are: mean (SD) age 82 (10), 83 (10), and 84 (9), gender (female %) 66, 64, and 70% respectively. The incident hip fractures and all clinical fractures for FRS risk level were similar among the three provinces and ranged from 0.5 to 19.2% and 1 to 19.2% respectively. The overall discriminative properties of the FRS were similar between ON (C-statistic = 0.673), BC (C-statistic = 0.644) and MB (C-statistic = 0.649) samples.ConclusionFRS is a valid tool for identifying LTC residents at different risk levels for hip or all clinical fractures in three provinces. Having a fracture risk assessment tool that is tailored to the LTC context and embedded within the routine clinical assessment may have significant implications for policy, service delivery and care planning, and may improve care for LTC residents across Canada.