Socioeconomic disparities and chronic respiratory diseases in Thailand: The National Socioeconomics Survey

Multi-class classification algorithms for the diagnosis of anemia in an outpatient clinical setting

Anemia is one of the most pressing public health issues in the world with iron deficiency a major public health issue worldwide. The highest prevalence of anemia is in developing countries. The complete blood count is a blood test used to diagnose the prevalence of anemia. While earlier studies have framed the problem of diagnosis as a binary classification problem, this paper frames it as a multi class (three classes) classification problem with mild, moderate and severe classes. The three classes for the anemia classification (mild, moderate, severe) are so chosen as the world health organization (WHO) guidelines formalize this categorization based on the Haemoglobin (HGB) values of the chosen sample of patients in the Complete Blood Count (CBC) patient data set. Complete blood count test data was collected in an outpatient clinical setting in India. We used Feature selection with Majority voting to identify the key attributes in the input patient data set. In addition, since the original data set was imbalanced we used Synthetic Minority Oversampling Technique (SMOTE) to balance the data set. Four data sets including the original data set were used to perform the data experiments. Six standard machine learning algorithms were utilised to test our four data sets, performing multi class classification. Benchmarking these algorithms was performed and tabulated using both10 fold cross validation and hold out methods. The experimental results indicated that multilayer perceptron network was predominantly giving good recall values across mild and moderate class which are early and middle stages of the disease. With a good prediction model at early stages, medical intervention can provide preventive measure from further deterioration into severe stage or recommend the use of supplements to overcome this problem.

Associations between indoor environment in residential buildings in wet and dry seasons and health of students in upper northern Thailand

We performed a repeated questionnaire study on home environment and health (six medical symptoms) in 1159 junior high school students (age 12.8 ± 0.7 years) in upper northern Thailand in wet and dry seasons. Data on outdoor temperature, relative humidity (RH), and air pollution were collected from nearest monitoring station. Odds ratios (OR) were calculated by multi-level logistic regression. Most common symptoms were rhinitis (62.5%), headache (49.8%), throat (42.8%), and ocular symptoms (42.5%). Ocular symptoms were more common at lower RH and rhinitis more common in dry season. Water leakage (28.2%), indoor mold (7.1%), mold odor (4.1%), and windowpane condensation (13.6%) were associated with all six symptoms (ORs: 1.3-3.5). Other risk factors included cat keeping, environmental tobacco smoke (ETS), other odor than mold odor, gas cooking, and cooking with biomass fire. Biomass burning inside and outside the home for other reasons than cooking was associated with all six symptoms (ORs: 1.5-2.6). Associations between home environment exposure and rhinitis were stronger in wet season. In conclusion, dampness-related exposure, windowpane condensation, cat keeping, ETS, gas cooking, and biomass burning can impair adolescent health in upper northern Thailand. In subtropical areas, environmental health effects should be investigated in wet and dry seasons.

The Impact of Socioeconomic Factors on Knowledge, Attitudes, and Practices of Dog Owners on Dog Rabies Control in Thailand

Rabies is a deadly zoonotic disease responsible for almost 60,000 deaths each year, especially in Africa and Asia including Thailand. Dogs are the major reservoirs for rabies virus in these settings. This study thus used the concept of knowledge, attitudes, and practices (KAP) to identify socioeconomic factors that contribute to the differences in the canine rabies occurrences in high and low-risk areas which were classified by a Generalized Additive Model (GAM). Multistage sampling was then applied to designate the study locations and a KAP-based questionnaire was used to retrieve data and relevant perspectives from the respondents. Based on the responses from 476 participants living across four regions of Thailand, we found that the knowledge of the participants was positively correlated with their behaviors but negatively associated with the attitudes. Participants who are male, younger, educated at the level of middle to high school, or raising more dogs are likely to have negative attitudes but good knowledge on rabies prevention and control whereas farmers with lower income had better attitudes regardless of their knowledge. We found that people in a lower socioeconomic status with a lack of knowledge are not willing to pay at a higher vaccine price. Public education is a key to change dog owners' behaviors. Related authorities should constantly educate people on how to prevent and control rabies in their communities. Our findings should be applicable to other countries with similar socioeconomic statuses.

Assessing the relative importance of social determinants of health in malaria and anemia classification based on machine learning techniques

Disparate types of data including biological and environmental have been used in supervised learning to predict a specific disease outcome. However, social determinants of health, which have been explored very little, promise to be significant predictors of public health problems such as malaria and anemia among children. We considered studying their contribution power in malaria and anemia predictions based on Variable Importance in Projection (VIP). This innovative method has potential advantages as it analyzes the impact of independent variables on disease prediction. In addition, we applied five machine learning algorithms to classify both diseases, using social determinants of health data, and compared their results. Of them all, artificial neural networks gave the best results of 94.74% and 84.17% accuracy for malaria and anemia prediction, respectively. These results are consistent and reflect the significance of non-medical factors in disease prediction.