Socio-Demographic Factors and Diabetes Risk Analysis PDF

Socio-Demographic Factors and Diabetes Risk: Findings from a Cross-Sectional Analysis Ana Vilhena (107978), Beatriz Monteiro (110285), Bruna Morim (108843), Joana Rosa (127935), Matilde Tavares (108132). Master’s in medical Statistics, Aveiro University Abstract...

Socio-Demographic Factors and Diabetes Risk: Findings from a Cross-Sectional Analysis Ana Vilhena (107978), Beatriz Monteiro (110285), Bruna Morim (108843), Joana Rosa (127935), Matilde Tavares (108132). Master’s in medical Statistics, Aveiro University Abstract The primary objective of this study is to analyze how demographic and socioeconomic variables, sex, age, education, and income, impact the risk of developing diabetes, based on a database of 253,680 observations. Additionally, factors such as physical activity and body mass index were also evaluated through descriptive analysis, inferential statistical tests, and multinomial logistic regression, all conducted in R Studio. Results show higher diabetes prevalence among older age groups, with slightly increased incidence in males, individuals with lower educational attainment, and those with lower income levels. Age was identified as the most significant predictor of diabetes risk, followed by income, sex, and education. Furthermore, lower levels of education contribute to unhealthy lifestyle behaviors and social disparities, which are associated with an increased risk of obesity and Type 2 diabetes. Low income is often linked to lifestyle factors like unhealthy diets, lack of exercise, and higher smoking rates, all of which increase the risk of diabetes. Low-income individuals have also limited access to healthcare, making it hard to get needed medical care due to high costs. The use of multiple variables in the cross-sectional study allowed us to understand how these factors are associated with diabetes risk. However, the study design does not allow the establishment of causal relationships and the lack of distinction between Type 1 and Type 2 diabetes limits the applicability of the findings to specific health interventions designed to reduce the risk of each type of diabetes. Keywords: Prevalence; Insulin; Risk factors; United States; Public health. Introduction According to the World Health Organization (WHO), diabetes mellitus is a chronic metabolic disease that disrupts the body’s ability to regulate blood sugar levels. Normally, the pancreas produces insulin, a hormone that helps glucose enter the body's cells to provide energy. Persistent high glucose levels over time can lead to serious complications, including heart disease, vision impairment, kidney damage, and nerve issues. With advances in treatment for diabetes and its associated complications, people with diabetes are living longer with their condition. Type 1 diabetes, an autoimmune disease often diagnosed in childhood or adolescence, occurs when the immune system attacks and destroys insulin-producing beta cells in the pancreas, leading to little or no insulin production. In diabetes, either the pancreas doesn’t produce enough insulin or the body’s cells become resistant to its effects, resulting in high blood sugar (hyperglycemia). It requires daily insulin administration and is influenced by genetic and environmental factors. Type 2 diabetes typically develops in adults but is increasingly seen in younger individuals due to rising obesity rates. It is characterized by insulin resistance and lifestyle factors such as poor diet, lack of exercise are major contributors. While type 1 is managed with insulin therapy (insulin injections), type 2 can be controlled through lifestyle changes and medications. (World Health Organization, 2016, p.11). 1  Over the past three decades, the number of people with diabetes mellitus has more than doubled globally, making it one of the most important public health challenges to all nations (Chen et al., 2011). The International Diabetes Federation has predicted that the number of people with diabetes worldwide will increase almost by 50% from 366 million in 2011 to 552 million by 2030, with diabetes being the seventh leading cause of death in the United States. The global expenditure on diabetes-related healthcare is an estimated US$760 billion a year (Whiting et al., 2011). In the United States, more than 25 million adults suffer from diabetes, and in 2020, 38% of all US adults had pre-diabetes. Its prevalence is approximately 11.6%, which is higher than the 9% rate observed in Europe. The United States lacks a central agency to negotiate prices with hospitals and drug manufacturers. As a result, prices are primarily set by pharmaceutical companies and negotiated with insurers, leading to high costs for consumers. This creates significant price disparities, especially compared to Europe, where government intervention is more prevalent. These disparities can be also linked to healthier dietary patterns in some European countries, public health policies focused on chronic disease prevention, and better access to universal healthcare. In contrast, the fragmented healthcare system and higher obesity rates in the U.S. contribute to the increased incidence of diabetes. Additionally, prevalence rates can vary from state to state due to socioeconomic, cultural, and political differences (Danaei et al., 2009). Numerous cohort studies conducted in recent years have identified that an increased risk of diabetes is primarily linked to factors such as age, health education, income, ethnicity, family history of diabetes, smoking, obesity, and physical inactivity (Deshpande et al., 2008). Therefore, the primary objective of this study is to investigate the influence of key demographic and socioeconomic variables, such as sex, age, education, and income, on the risk of developing diabetes, according to the supplied database. According to the literature reviewed and the importance of other variables in this study, an additional analysis was conducted to examine the association between diabetes, BMI, and physical activity. Methodology Study Design and Sample Selection A Random digit dialing sampling strategy provides residential telephone numbers an equal probability of selection (Groves, 1998). Sample sizes requirements were projected to ensure statistical precision adequate for state and sub-state analyses (Levin, 2006). Stratification within the context Data System Sampling to provide valid estimates for subgroups reflecting the practices in the Behavioral Risk Factor Surveillance System (CDC, 2021). A continuous structured telephone survey interview was structured with a core standardized health behavior and demographic questions component and optional modules that cover emerging health issues or specific state needs. Optional modules are included which hold emerging health issue questions or local health priority, and State-Added Questions (Ezzati-Rice. et al, 1998). The proposed study recruited a diverse group of people at one point in time or across the shortest duration, providing an overall picture on health behaviors for that given population, thence a cross-sectional study by definition and also an observational study, since subjects were just observed without any variables under control as they merely described the health-related behavior and demographics in the market at present. (Godis, 2014).The dependent variables were the Diabetes 2 status-self-reported diagnosis and the independent variables the Demographics (Sex and Age) and Socioeconomic Measures (Education level and Income level)(Godis, 2014). Data management and Analysis Recoding and Variable Selection was conducted by converting the key variables of interest (age, education, and income) to factors using the function factor(), and recoding the diabetes and gender variables to be categorical factors(Wickham, 2016). The Descriptive Analysis performed was a stratified analysis by diabetes status, using tbl_summary() from the “gtsummary” package and adding an overall summary by using the add_overall() to visualize the frequencies in each category. Following, the Association Testing was conducted by Chi-square tests using chisq.test(), which should, in case of low frequencies, involve the Monte Carlo simulation approach, keeping the significance at p