Spatial Epidemiology And Risk Factors Of Stroke On A Smaller Scale

[Background and objectives]Stroke, also known as cerebrovascular accident or brain attack, is a disease that affects the arteries leading to and within the brain. A stroke occurs when a blood vessel that carries oxygen and nutrients to the brain is either blocked by a clot or bursts (or ruptures) which results in part of the brain not functioning properly. Ischemic stroke (IS) and hemorrhagic stroke (HS) are two main subtypes of stroke. Risk factors for stroke can be divided into two levels including group level (such as geographical ecological factors, etc.) and individual level (such as smoking, alcohol consumption, hypertension, etc.). Researches indicated that there was a significant regional aggregation in morbidity and mortality of stroke worldwide. The incidence of stoke in China gradually increased from south to north. However, the large spatial scale and insufficiency of systematic analysis for these studies made it difficult to put forward a more targeted regional prevention strategy well as control measures. In actual work, we found that even in a small area (such as a county), the incidences of stroke were disparity, which hinted that spatial aggregation could exist on a small spatial scale. Specific geographic risk factors might contribute to this spatial aggregation. This study was conducted in Yiyuan County in Shandong Province, based on the Active Surveillance System for Chronic Diseases, under the framework of geographic information system, using spatial epidemiology and spatial statistical methods to systematically analyze the spatial structure and spatial heterogeneity on a smaller scale. Furthermore, this study used geographic weighted regression to investigate the common geographical risk factor of stroke and other diseases (myocardial infarction, cancer, infectious diseases, etc.). The purpose of this study was to provide scientific prevention and control measures of stroke on a smaller scale. At the individual exposure level, the occurrence and development of stroke is the result of constantly exposed to risk factors throughout their life course. Exposure during early life may have specific pathogenic effect on the onset of stroke in later life. Hence, it is important to explore the effect of early exposure on stroke in later life. Therefore, this study used population-based spatial matched case-control design to explore the early risk factors for stroke and its efficient model.[Design and methods]The data used to analyze the spatial distribution of stroke was from the Active Surveillance System for Chronic Diseases (8505 stroke cases, including 6736 IS and 1709 HS). The data not only included the main demographic and diagnostic information of stroke, but also had the long-term residence of individuals and the geographic coordinates of cases. Population-based matched case-control design was used to explore the effect of risk factors in early life on stroke in later life. Lifespan Risk Exposure Measurement Instrument was performed to collect the exposure of risk factors during lifetime through a retrospective method. Use conditional logistic regression model to explore risk factors of stroke and its efficient model.[Results]1. Incidence density and distribution of stroke and its subtypes during 2011-2014 in Yiyuan County:The average incidence density of stroke was 408.05 per 100,000 person-years. The average incidence density of IS (408.05 per 100,000 person-years) was higher than HS (81.97 per 100,000 person-years). For stroke and its subtypes, incidence density was highest among 60-year-old and older population. The average incidence density of stroke and its subtypes (IS and HS) in man were higher than women. For IS, the incidence density was higher during March and May (spring) than other months, however, for HS, the incidence density was higher during December and February (winter) than other months.2. Spatial analysis of stroke and its subtypes:(1) Spatial heterogeneity and structure coexisted for stroke and its subtypes. There are many high incidence density villages throughout the research area, and these high incidence density villages showed point shaped distribution.(2) After standardization, the spatial heterogeneity and structure of stroke and its subtype still remained. The high incidence density villages showed point shaped distribution.(3) Based on the spatial correlation between the adjacent geographical units (villages), local empirical Bayes smoothing was used to correct the spatial error. The high incidence density villages still showed point shaped distribution when spatial variability decreased.(4) The villages with excess hazard ratios (HER)>2 and≤0.5 of stroke and its subtypes were crossing distribution suggesting that the dangerous geographical factors and protective geographical factors in the whole county coexisted.(5) On a larger spatial scale, the result of local spatial autocorrelation analysis and spatial scan analysis indicated that the high incidence density villages showed point shaped distribution. The high incidence density villages of stroke and IS located in the northwestern of Yiyuan County, and the high incidence density villages of HS located in the southwest of Yiyuan County.(6) The results of geographical weighted regression indicated that spatial correlation between myocardial infarction, cancer and infectious diseases and stroke exist.1) The incidence density of myocardial infarction was positively associated with stroke and IS, except in the northern region, and the coefficients were gradually weakened from west to east.2) The incidence density of cancer was negatively associated with stroke and IS and the coefficients were gradually weakened from north to south.3) The incidence density of infectious diseases was positively associated with stroke and IS, except in the west region, and the coefficients were greater in central region than in west or east, and in central region, the coefficients were gradually weakened from north to south.4) The incidence density of infectious diseases and cancer were positively correlated with HS, and the coefficients were gradually weakened from east to west.3. Analysis of risk factors for stroke(1) Cigarette smoking during 23 years old to 42 years old was a risk factor of stroke in later life in men.(2) Alcohol drinking between 20-year-old to 66-year-old could increase the risk of stroke in later life in men.(3) Tea consumption between 32-year-old to 34-year-old and 42-year-old to 43-year-old was a protective factor of stroke in later life in women.(4) Fish (shrimp) intake during 18-year-old to 24-year-old and 35-year-old to 47-year-old in men and during 61-year-old to 66-year-old in women could decrease the risk of stroke in later life.(5) Coarse food grain consumption between 25-year-old to 42-year-old could increase the risk of stroke in later life in women.(6) Moderate consumption of peanut oil was a protective factor of stroke in later life in both men and women.(7) Physical activities after 52-year-old was a protective factor for stroke in men.[Conclusions]1. The incidence density of stroke in Yiyuan County was extremely high (408.05 per 100,000 person-years,438.56 per 100,000 person-years for men and 377.39 per 100,000 person-years for women). The incidence density of IS was higher in spring, and the incidence density of HS was higher in winter.2. Spatial heterogeneity and structure coexisted for stroke and its subtypes. There are many high incidence density villages throughout the research area, and these high incidence density villages showed point shaped distribution. Dangerous geographical factors and protective geographical factors in the whole county coexisted.3. The spatial correlation between myocardial infarction, cancer and infectious diseases and stroke exist.4. Tobacco, alcohol, tea exposure in early life could increase the risk of stroke in later life, however, physical activities in later life were protective factor for stroke in later life in men. In women, coarse as the staple food in early life was a risk factor of stoke in later life. Both men and women, moderate consumption of peanut oil throughout the life course was a protective factor of the occurrence of stroke.