Gonorrhea cluster detection in Manitoba, Canada: Spatial, temporal, and spatio-temporal analysis.

In Canada, Gonorrhea infection ranks as the second most prevalent sexually transmitted infection. In 2018, Manitoba reported an incidence rate three times greater than the national average. This study aims to investigate the spatial, temporal, and spatio-temporal patterns of Gonorrhea infection in Manitoba, using individual-level laboratory-confirmed administrative data provided by Manitoba Health from 2000 to 2016. Age and sex patterns indicate that females are affected by infections at younger ages compared to males. Moreover, there is an increase in repeated infections in 2016, accounting for 16% of the total infections. Spatial analysis at the 96 Manitoba regional health authority districts highlights significant positive spatial autocorrelation, demonstrating a clustered distribution of the infection. Northern districts of Manitoba and central Winnipeg were identified as significant clusters. Temporal analysis shows seasonal patterns, with higher infections in late summer and fall. Additionally, spatio-temporal analysis reveals clusters during high-risk periods, with the most likely cluster in the northern districts of Manitoba from January 2006 to June 2014, and a secondary cluster in central Winnipeg from June 2004 to November 2012. This study identifies that Gonorrhea infection transmission in Manitoba has temporal, spatial, and spatio-temporal variations. The findings provide vital insights for public health and Manitoba Health by revealing high-risk clusters and emphasizing the need for focused and localized prevention, control measures, and resource allocation.

The effective reproductive number: Modeling and prediction with application to the multi-wave Covid-19 pandemic.

Classical compartmental models of infectious disease assume that spread occurs through a homogeneous population. This produces poor fits to real data, because individuals vary in their number of epidemiologically-relevant contacts, and hence in their ability to transmit disease. In particular, network theory suggests that super-spreading events tend to happen more often at the beginning of an epidemic, which is inconsistent with the homogeneity assumption. In this paper we argue that a flexible decay shape for the effective reproductive number (Rt) indexed by the susceptible fraction (St) is a theory-informed modeling choice, which better captures the progression of disease incidence over human populations. This, in turn, produces better retrospective fits, as well as more accurate prospective predictions of observed epidemic curves. We extend this framework to fit multi-wave epidemics, and to accommodate public health restrictions on mobility. We demonstrate the performance of this model by doing a prediction study over two years of the SARS-CoV2 pandemic.

A discrete-time susceptible-infectious-recovered-susceptible model for the analysis of influenza data.

We develop a discrete time compartmental model to describe the spread of seasonal influenza virus. As time and disease state variables are assumed to be discrete, this model is considered to be a discrete time, stochastic, Susceptible-Infectious-Recovered-Susceptible (DT-SIRS) model, where weekly counts of disease are assumed to follow a Poisson distribution. We allow the disease transmission rate to also vary over time, and the disease can only be reintroduced after extinction if there is a contact with infected individuals from other host populations. To capture the variability of influenza activities from one season to the next, we define the seasonality with a 4-week period effect that may change over years. We examine three different transmission rates and compare their performance to that of existing approaches. Even though there is limited information for susceptible and recovered individuals, we demonstrate that the simple models for transmission rates effectively capture the behaviour of the disease dynamics. We use a Bayesian approach for inference. The framework is applied in an analysis of the temporal spread of influenza in the province of Manitoba, Canada, 2012-2015.

Analyzing COVID-19 data in the Canadian province of Manitoba: A new approach.

The basic homogeneous SEIR (susceptible-exposed-infected-removed) model is a commonly used compartmental model for analysing infectious diseases such as influenza and COVID-19. However, in the homogeneous SEIR model, it is assumed that the population of study is homogeneous and, one cannot incorporate individual-level information (e.g., location of infected people, distance between susceptible and infected individuals, vaccination status) which may be important in predicting new disease cases. Recently, a geographically-dependent individual-level model (GD-ILM) within an SEIR framework was developed for when both regional and individual-level spatial data are available. In this paper, we propose to use an SEIR GD-ILM for each health region of Manitoba (central Canadian province) population to analyse the COVID-19 data. As different health regions of the population under study may act differently, we assume that each health region has its own corresponding parameters determined by a homogeneous SEIR model (such as contact rate, latent period, infectious period). A Monte Carlo Expectation Conditional Maximization (MCECM) algorithm is used for inference. Using estimated parameters we predict the infection rate at each health region of Manitoba over time to identify highly risk local geographical areas. Performance of the proposed approach is also evaluated through simulation studies.

A time-heterogeneous D-vine copula model for unbalanced and unequally spaced longitudinal data.

In many longitudinal studies, the number and timing of measurements differ across study subjects. Statistical analysis of such data requires accounting for both the unbalanced study design and the unequal spacing of repeated measurements. This paper proposes a time-heterogeneous D-vine copula model that allows for time adjustment in the dependence structure of unequally spaced and potentially unbalanced longitudinal data. The proposed approach not only offers flexibility over its time-homogeneous counterparts but also allows for parsimonious model specifications at the tree or vine level for a given D-vine structure. It further provides a robust strategy to specify the joint distribution of non-Gaussian longitudinal data. The performance of the time-heterogeneous D-vine copula models are evaluated through simulation studies and by a real data application. Our findings suggest improved predictive performance of the proposed approach over the linear mixed-effects model and time-homogeneous D-vine copula model.

Modeling the spatio­temporal spread of COVID­19 cases, recoveries and deaths and effects of partial and full vaccination coverage in Canada.

The purposes of our study are to map high-risk areas in Canada as well as quantifying the effects of vaccination intervention and socio-demographic factors on the transmission rates of infection, recovery, and death related to COVID-19. The data of this research included weekly number of COVID­19 cases, recovered, and dead individuals from 2020 through 2021 in Canada at health region and provincial levels. These data were associated with cumulative rates of partial and full vaccination and socio-demographic factors. We applied the spatio-temporal Susceptible-Exposed-Infected-Removed (SEIR), and Susceptible-Exposed-Infected-Removed-Vaccinated (SEIRV) models. The results indicated the partial vaccination rate has a greater effect compared with full vaccination rate on decreasing the rate of infectious cases (risk ratio (RR) = 0.18; 95%CrI: 0.16-0.2; RR = 0.60; 95%CrI: 0.55-0.65, respectively) and increasing the rate of recovered cases (RR = 1.39; 95%CrI: 1.28-1.51; RR = 1.21; 95%CrI: 1.23-1.29, respectively). However, for mortality risk reduction, only increasing full vaccination rate was significantly associated (RR = 0.09; 95%CrI: 0.05-0.14). In addition, our results showed that regions with higher rates of elderly and aboriginal individuals, higher population density, and lower socioeconomic status (SES) contribute more to the risk of infection transmission. Rates of elderly and aboriginal individuals and SES of regions were significantly associated with recovery rate. However, elderly individuals rate of regions was only a significant predictor of mortality risk. Based on the results, protection against mild and severe COVID-19 infection after the primary vaccination series decreased.

A descriptive analysis of food pantries in twelve American states: hours of operation, faith-based affiliation, and location.

BACKGROUND: Our objectives were to describe both the development, and content, of a charitable food dataset that includes geographic information for food pantries in 12 American states. METHODS: Food pantries were identified from the foodpantries.org website for 12 states, which were linked to state-, county-, and census-level demographic information. The publicly available 2015 Food Access Research Atlas and the 2010 US Census of Population and Housing were used to obtain demographic information of each study state. We conducted a descriptive analysis and chi-square tests were used to test for differences in patterns of food pantries according to various factors. RESULTS: We identified 3777 food pantries in 12 US states, providing an estimated 4.84 food pantries per 100,000 people, but ranged from 2.60 to 7.76 within individual states. The majority of counties (61.2%) had at least one food pantry. In contrast, only 15.7% of all census tracts in the study states had at least one food pantry. A higher proportion of urban census tracts had food pantries compared to rural tracts. We identified 2388 (63.2%) as being faith-based food pantries. More than a third (34.4%) of food pantries did not have information on their days of operation available. Among the food pantries displaying days of operation, 78.1% were open at least once per week. Only 13.6% of food pantries were open ≤1 day per month. CONCLUSIONS: The dataset developed in this study may be linked to food access and food environment data to further examine associations between food pantries and other aspects of the consumer food system (e.g. food deserts) and population health from a systems perspective. Additional linkage with the U.S. Religion Census Data may be useful to examine associations between church communities and the spatial distribution of food pantries.

Social determinants and spatio-temporal variation of Ischemic Heart Disease in Manitoba.

INTRODUCTION: The aim was to study any spatial and/or temporal patterns of ischemic heart disease (IHD) prevalence and measure the effects of selected social determinants on these spatial and space-time patterns. METHODS: Data were obtained from the Population Research Data Repository housed at the Manitoba Centre for Health Policy to identify persons who were diagnosed with IHD between 1995 and 2018. These persons were geocoded to 96 geographic regions of Manitoba. An area-level socioeconomic factor index (SEFI-2) and the proportion of the population who was Indigenous were calculated for each geographic region using the 2016 Canadian Census data. Associations between these factors and IHD prevalence were measured using Bayesian spatial Poisson regression models. Temporal trends and spatio-temporal trends were measured using Bayesian spatio-temporal Poisson regression models. RESULTS: Univariable models showed a significant association with increased regional Indigenous population proportion associated with a higher prevalence of IHD (RR: 0.07, 95% CredInt: (0.05, 0.10)) and for SEFI-2 (RR: 0.17, 95% CredInt: (0.11, 0.23)). Using a multivariable model, after accounting for the proportion of the population that was Indigenous, there was no evidence of an association between IHD prevalence and area-level socioeconomic factor. Spatio-temporal models showed no significant overall temporal trend in IHD prevalence, but there were significant spatially varying temporal trends within the 96 regions. CONCLUSIONS: Association between Indigenous population proportion and IHD is consistent with previous research. No significant overall temporal trend was measured. However, regions with significantly increasing trends and significantly decreasing trends in IHD prevalence were identified.

Geographical variation and factors associated with gastric cancer in Manitoba.

OBJECTIVES: We investigated the spatial disparities and factors associated with gastric cancer (GC) Incidence in Manitoba. METHODS: We combined information from Manitoba Cancer registry and Census data to obtain an age-sex adjusted relative risk (IRR) of GC incidence. We geocoded the IRR to the 96 regional health authority districts (RHADs) using the postal code conversion file (PCCF). Bayesian spatial and spatio-temporal Poisson regression models were used for the analysis. RESULTS: Adjusting for the effect of socio-economic score index (SESI), Indigenous, and immigrant population, 25 districts with high overall GC risk were identified. One unit increase in SESI was associated with reduced risk of cardia GC (CGC) by 14% (IRR = 0.859; 95% CI: 0.780-0.947) and the risk of non-cardia GC (NCGC) by approximately 10% (IRR = 0.898; 95% CI: 0.812-0.995); 1% increase in regional Indigenous population proportion reduced the risk of CGC by 1.4% (IRR = 0.986; 95% CI: 0.978-0.994). In the analysis stratified by sex, one unit increase in SESI reduced the risk of CGC among women by 26.2% (IRR = 0.738; 95% CI: 0.618-0.879), and a 1% increase in Indigenous population proportion reduced the risk of CGC among women by 1.9% (IRR = 0.981; 95% CI: 0.966-0.996). CONCLUSION: Our results support a significant association between SESI and NCGC. We report regional variation of GC IRR and a varying temporal pattern across the RHADs. These results could be used to prioritize interventions for regions with high and progressive risk of GC.

Spatial modeling of individual-level infectious disease transmission: Tuberculosis data in Manitoba, Canada.

Geographically dependent individual level models (GD-ILMs) are a class of statistical models that can be used to study the spread of infectious disease through a population in discrete-time in which covariates can be measured both at individual and area levels. The typical ILMs to illustrate spatial data are based on the distance between susceptible and infectious individuals. A key feature of GD-ILMs is that they take into account the spatial location of the individuals in addition to the distance between susceptible and infectious individuals. As a motivation of this article, we consider tuberculosis (TB) data which is an infectious disease which can be transmitted through individuals. It is also known that certain areas/demographics/communities have higher prevalent of TB (see Section 4 for more details). It is also of interest of policy makers to identify those areas with higher infectivity rate of TB for possible preventions. Therefore, we need to analyze this data properly to address those concerns. In this article, the expectation conditional maximization algorithm is proposed for estimating the parameters of GD-ILMs to be able to predict the areas with the highest average infectivity rates of TB. We also evaluate the performance of our proposed approach through some simulations. Our simulation results indicate that the proposed method provides reliable estimates of parameters which confirms accuracy of the infectivity rates.

Managing mental health: why we need to redress the balance between healthcare spending and social spending.

BACKGROUND: Mental health outcomes vary widely among high-income countries, although mental health problems represent an increasing proportion of the burden of disease for all countries. This has led to increased demand for healthcare services, but mental health outcomes may also be particularly sensitive to the availability of social services. This paper examines the variation in the absolute and relative amounts that high-income countries spend on healthcare and social services to determine whether increased expenditure on social services relative to healthcare expenditure might be associated with better mental health outcomes. METHODS: This paper estimates the association between patterns of government spending and population mental health, as measured by the death rate resulting from mental and behavioural disorders, across member countries of the Organisation for Economic Cooperation and Development (OECD). We use country-level repeated measures multivariable modelling for the period from 1995 to 2016 with region and time effects, adjusted for total spending and demographic and economic characteristics. Healthcare spending includes all curative services, long-term care, ancillary services, medical goods, preventative care and administration whilst social spending consists of all transfer payments made to individuals and families as part of the welfare state. RESULTS: We find that a higher ratio of social to healthcare expenditure is associated with significantly better mental health outcomes for OECD populations, as measured by the death rate resulting from mental and behavioural disorders. We also find that there is no statistically significant association between healthcare spending and population mental health when we do not control for social spending. CONCLUSION: This study suggests that OECD countries can have a significant impact on population mental health by investing a greater proportion of total expenditure in social services.

Geographical Variation and Factors Associated With Inflammatory Bowel Disease in a Central Canadian Province.

BACKGROUND: We investigated temporal trends, geographical variation, and geographical risk factors for incidence of inflammatory bowel disease (IBD). METHODS: We used the University of Manitoba IBD Epidemiology Database to identify incident IBD cases diagnosed between 1990 and 2012, which were then geocoded to 296 small geographic areas (SGAs). Sociodemographic characteristics of the SGAs (proportions of immigrants, visible minorities, Indigenous people, and average household income) were obtained from the 2006 Canadian Census. The geographical variation of IBD incidence was modeled using a Bayesian spatial Poisson model. Time trends of IBD incidence were plotted using Joinpoint regression. RESULTS: The incidence of IBD decreased over the study years from 23.6 (per 100,000 population) in 1990 to 16.3 (per 100,000 population) in 2012. For both Crohn's disease (CD) and ulcerative colitis (UC), the highest incidence was in Winnipeg and the southern and central regions of Manitoba, whereas most of northern Manitoba had lower incidence. There was no effect of sociodemographic characteristics of SGAs, other than the proportion of Indigenous people, which was associated with lower IBD incidence. CONCLUSIONS: Although the incidence of IBD in Manitoba is decreasing over time, we have identified geographic areas with persistently higher IBD incidence that warrant further study for etiologic clues.

Health-related quality of life and economic burden to smoking behaviour among Canadians.

OBJECTIVES: The objectives of this study were to: (1) examine whether the smoking status of the Canadian population is associated with a reduction in health-related quality of life (HRQoL); (2) calculate the overall economic burden of loss in HRQoL using a commonly accepted $100,000 willingness-to-pay (WTP) threshold to gain one quality-adjusted life year (QALY); and (3) calculate the loss of HRQoL over a lifetime. METHODS: We used the 2015 Canadian Community Health Survey. The variations in HRQoL were estimated using a multivariable generalized linear model. Total expected lifetime QALYs lost due to smoking were calculated by compounding the annual adjusted health utility loss associated with smoking across a respondent's remaining years of life expectancy stratified by age. A discount rate of 1.5% was applied to the analysis based on recent analysis of the costs of borrowing in Canada. RESULTS: Smoking is significantly associated with HRQoL loss. This study demonstrated that smoking is associated with a 0.05 and 0.01 reduction in Health Utilities Index Mark 3 (HUI3) score for current and former smokers, which also corresponds to a loss of 0.66 quality-adjusted life years in average, and also is associated with substantial individual and societal economic cost. The total lifetime economic burden of HUI3 loss per smoker was $65,935, yielding in the aggregate a societal burden of $1068.88 billion in the study population. CONCLUSION: Tobacco control, prevention and intervention not only will improve HRQoL but also will generate social returns on investment.

Exploring dental hygiene decision making: A quantitative study of potential organizational explanations.

BACKGROUND: To improve public access to oral health care, dental hygienists have been identified for practice expansion, and, therefore, they must demonstrate decision-making capacity. This study aimed to identify and test potentially influential factors in dental hygiene decision making. Organizational and gender factors were hypothesized to be most influential and focused the study. METHODS: A 2-phase mixed methods approach was used. In Phase I, a qualitative decision-making model was developed and subsequently published in 2012. Phase II tested aspects of the model through an electronic survey instrument and key informant interviews. This article reports on the statistical results of the quantitative survey. A third article will report on the qualitative thematic analyses and merged interpretation. RESULTS: The Phase I qualitative model guided the development of the survey instrument. The survey had a 38% response rate; moderate to weak correlations between predictor variables (structural and individual) and clinical decision making were shown. The final statistical model demonstrated that individual characteristics and graduating from a 3-year dental hygiene program were together significantly associated with decision-making capacity. DISCUSSION AND CONCLUSIONS: Individual characteristics and longer education were together shown to be associated with increased decision-making capacity. These findings did not show the organization or gender to be important in influencing decision-making capacity. However, the merging of the quantitative survey and qualitative key informant data will potentially inform how the organization influences the individual dental hygienist.

CONTEXTE: Afin d'améliorer l'accès de la population aux soins de santé buccodentaire, les hygiénistes dentaires ont été désignés pour une expansion de la pratique et doivent par conséquent démontrer une capacité décisionnelle. La présente étude visait à cerner et à vérifier les facteurs influents potentiels dans la prise de décision en hygiène dentaire. L'étude était axée sur l'hypothèse que les facteurs organisationnels et de sexe étaient les plus influents. MÉTHODOLOGIE: Une approche méthodologique mixte en 2 phases a été utilisée. Dans la phase I, un modèle décisionnel qualitatif a été conçu et publié par la suite en 2012. La phase II a évalué des aspects du modèle au moyen d'un outil de sondage électronique et des entrevues d'intervenants clés. Cet article présente les résultats statistiques de ce sondage quantitatif. Un troisième article fera part des analyses thématiques qualitatives et des interprétations fusionnées. RÉSULTATS: La phase I du modèle qualitatif a guidé la conception de l'outil de sondage. Le sondage avait un taux de réponse de 38 %, et des corrélations modérées à faibles entre les variables indépendantes (structurelles et individuelles) et la prise de décision clinique étaient démontrées. Le modèle statistique final a démontré que les caractéristiques individuelles, ainsi que l'obtention d'un diplôme d'un programme d'hygiène dentaire de 3 ans étaient fortement associées à la capacité décisionnelle. DISCUSSION ET CONCLUSIONS: Les caractéristiques individuelles et des études plus longues étaient ensemble associées à une meilleure capacité décisionnelle. Ces résultats n'ont pas montré que l'organisation ou le sexe étaient des facteurs d'influence importants dans la capacité décisionnelle. Cependant, la fusion du sondage quantitatif et des données qualitatives des intervenants clés pourrait éclaircir la façon dont l'organisation influence chaque hygiéniste dentaire.

The effects of cigarette price and the amount of pocket money on youth smoking initiation and intensity in Canada.

OBJECTIVES: To investigate the price and income elasticities of adolescent smoking initiation and intensity to determine the extent to which increased pocket money leads to greater smoking among youth, and whether higher taxes can mitigate this effect. METHODS: We used the 2012/2013 Canadian Youth Smoking Survey including students in grades 7-12. The multivariable logistic regression was used to examine the probability of smoking initiation, and a linear regression to examine the smoking intensity determined by province-level prices of cigarettes, pocket money, and a vector of individual characteristics, including age, sex, race, and school-related and psychosocial factors. RESULTS: Of respondents, 28.8% have tried cigarette smoking. More than 90% of these initiated smoking between age 9 and 17. Male smokers consumed a higher average number of whole cigarettes daily than did females. The price elasticity of smoking initiation and intensity for youth in the full sample were - 1.13 and - 1.02, respectively, which means that a 10% increase in price leads to an 11.3% reduction in initiation and a 10.2% reduction in intensity. The income elasticity of smoking initiation and intensity for youth in the full sample were 0.07 and 0.06, respectively, which means that a 10% increase in income leads to a 0.7% increase in initiation and a 0.6% increase in intensity. CONCLUSION: Economic measures such as taxation that raise the price of cigarettes may be a useful policy tool to limit smoking initiation and intensity.

Examining the impact of the number of regions used in cluster detection methods: An application to childhood asthma visits to a hospital in Manitoba, Canada.

The level of spatial aggregation is a major concern in cluster investigations. Combining regions to protect privacy may result in a loss of power and thus, can limit the information researchers can obtain. The impact of spatial aggregation on the ability to detect clusters is examined in this study, which shows the importance of choosing the correct level of spatial aggregation in cluster investigations. We applied the circular spatial scan statistic (CSS), flexible spatial scan statistic (FSS) and Bayesian disease mapping (BYM) approaches to a dataset containing childhood asthma visits to a hospital in Manitoba, Canada, using three different levels of spatial aggregation. Specifically, we used 56, 67 and 220 regions in the analysis, respectively. It is expected that the three scenarios will yield different results and will highlight the importance of using the right level of spatial aggregation. The three methods (CSS, FSS, BYM) examined in this study performed similarly when detecting potential clusters. However, for different levels of spatial aggregation, the potential clusters identified were different. As the number of regions used in the analysis increased, the total area identified in the cluster decreased. In general, potential clusters were identified in the central and northern parts of Manitoba. Overall, it is crucial to identify the appropriate number of regions to study spatial patterns of disease as it directly affects the results and consequently the conclusions. Additional investigation through future work is needed to determine which scenario of spatial aggregation is best.

Lifetime Pattern of Atrial Fibrillation and the Risks of Stroke and Death in a Population-based Cohort of Men (from The Manitoba Follow-Up Study).

Atrial fibrillation (AF) is associated with stroke and mortality. The arrhythmia can be sustained or intermittent. Previous studies that have used fixed covariates and short-time horizons to examine the relation between the pattern of AF and the occurrence of events have produced conflicting results. The Manitoba Follow-Up Study includes 3,983 originally healthy men who have been followed with routine examinations since 1948. AF status during each visit was classified into the following patterns: free of AF, newly diagnosed; intermittent AF-in sinus;intermittent AF-in AF; sustained AF. We created adjusted Cox proportional hazards models with time-dependent covariates to estimate risks for stroke and death according to AF pattern. After 167,982 person-years of follow-up and 66,297 electrocardiograms (ECGs), 548 men had at least 1 ECG documenting AF, 799 had a stroke, and 3173 died. Relative to men free of AF, sustained and newly diagnosed AF were associated with stroke (hazard ratio [HR] 1.85, 95% confidence interval [CI] 1.33 to 2.59 and HR 1.71, 95% CI 1.10 to 2.66, respectively) and death (HR 2.48, 95% CI 2.11 to 2.92 and HR 2.03, 95% CI 1.64 to 2.52, respectively). Intermittent AF was associated with death (HR 2.41 95% CI 1.58 to 3.68 in AF and HR 1.71 95% CI 1.44 to 2.03 in sinus), but not with stroke (HR 0.68, 95% CI 0.22 to 2.13 in AF and HR 1.02 95% CI 0.72 to 1.45 in sinus). Antithrombotic therapy was associated with a reduced risk of the outcomes. In conclusion, longitudinal analysis of patterns of AF evolving over time provided evidence that the associated risks of stroke and death vary considerably with rhythm classification on serial ECGs.

Modeling the random effects covariance matrix for longitudinal data with covariates measurement error.

Longitudinal data occur frequently in practice such as medical studies and life sciences. Generalized linear mixed models (GLMMs) are commonly used to analyze such data. It is typically assumed that the random effects covariance matrix is constant among subjects in these models. In many situations, however, the correlation structure may differ among subjects and ignoring this heterogeneity can lead to biases in model parameters estimate. Recently, Lee et al developed a heterogeneous random effects covariance matrix for GLMMs for error-free covariates. Covariates measured with error also happen frequently in the longitudinal data set-up (eg, blood pressure and cholesterol level). Ignoring this issue in the data may produce bias in model parameters estimate and lead to wrong conclusions. In this paper, we propose an approach to properly model the random effects covariance matrix based on covariates in the class of GLMMs, where we also have covariates measured with error. The resulting parameters from the decomposition of random effects covariance matrix have a sensible interpretation and can be easily modeled without the concern of positive definiteness of the resulting estimator. The performance of the proposed approach is evaluated through simulation studies, which show that the proposed method performs very well in terms of bias, mean squared error, and coverage rate. An application of the proposed method is also provided using a longitudinal data from Manitoba follow-up study.

Geographical Variation and Factors Associated with Non-Small Cell Lung Cancer in Manitoba.

BACKGROUND: Screening decreases non-small cell lung cancer (NSCLC) deaths and is recommended by the Canadian Task Force on Preventive Health Care. We investigated risk factor prevalence and NSCLC incidence at a small region level to inform resource allocation for lung cancer screening. METHODS: NSCLC diagnoses were obtained from the Canadian Cancer Registry, then geocoded to 283 small geographic areas (SGAs) in Manitoba. Sociodemographic characteristics of SGAs were obtained from the 2006 Canadian Census and Canadian Community Health Survey. Geographical variation was modelled using a Bayesian spatial Poisson model. RESULTS: NSCLC incidence in SGAs ranged from 1 to 343 cases per 100,000 population per year. The highest incidence rates were in the Southeastern, Southwestern, and Central regions of Manitoba, while most of Northern Manitoba had lower rates. Poisson regression suggested areas with higher proportions of Aboriginal people and higher average income, and immigrants had lower NSCLC incidence whereas areas with higher proportions of smokers had higher incidence. CONCLUSION: On an SGA level, smoking rates remain the most significant factor driving NSCLC incidence. Socioeconomic status and proportions of immigrants or Aboriginal peoples independently impact NSCLC rates. We have identified SGAs in Manitoba to target in policy and infrastructure planning for lung cancer screening.

Time and spatial trends in lymphoid leukemia and lymphoma incidence and survival among children and adolescents in Manitoba, Canada: 1984-2013.

OBJECTIVES: To test for time and spatial trends in lymphoid malignancies, including lymphoid leukemia (LL), Hodgkin lymphoma (HL), and non-Hodgkin lymphoma (NHL), in children and adolescents in the province of Manitoba, Canada. METHODS: Incident cases diagnosed between 1984 and 2013 were identified from the Manitoba Cancer Registry. We assessed time trends in age-standardized incidence rates using joinpoint regression and in 5-year relative survival using Poisson regression model. Kulldorff's scan method was used to assess spatial variation and clustering. RESULTS: Age-standardized incidence rates (per million person-years) in males and females were 34.0 (95% confidence interval [CI] 28.9-39.1) and 26.2 (95% CI 21.5-30.7) for LL, 10.5 (95% CI 7.7-13.3) and 12.5 (95% CI 9.4-15.7) for HL, 12.5 (95% CI 9.3-15.4) and 7.7 (95% CI 5.2-10.2) for NHL (except for Burkitt lymphomas), and 3.2 (95% CI 1.6-4.7) and 1.5 (95% CI 0.4-2.5) for Burkitt lymphomas. Age- and sex- standardized LL incidence rate increased 1.4% (95% CI 0.3%-2.5%) per year, while the changes for HL and NHL incidence rates were not statistically significant. There were geographic differences in age-standardized incidence rates for LL, HL, and NHL and spatial clusters were detected in southern part of the province. Five-year relative survival has improved over time and there was no difference between rural and urban areas. CONCLUSIONS: Lymphoid leukemia incidence rate increased over time and varied by geographic area. Further research should examine the factors contributing to these trends.