Spatial analysis of tuberculosis occurrence and its socio-economic determinants in Si Sa Ket Province, Thailand | |
| Author | Siriwan Hassarangsee |
| Call Number | AIT Diss. no.RS-18-08 |
| Subject(s) | Tuberculosis--Thailand--Si Sa Ket Economics--Sociological aspects--Thailand--Si Sa Ket Spatial analysis (Statistics) |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Remote Sensing and Geographic Information Systems |
| Publisher | Asian Institute of Technology |
| Abstract | Tuberculosis (TB) is an airborne-infectious disease which has caused global concerns for years. More than 98 percent of deaths from TB are in developing countries. Remained untreated, about 70 percent of TB patients will die within 10 years. In Thailand, about one third of the people are found to be TB infected. TB is a social disease which commonly affects people under deprivation. This is also the case in Thailand. High TB morbidity rates were commonly detected in the provinces with widespread poverty, especially in the northeastern area. Although there is a great deal of research to support the Thai government's efforts on TB prevention and control, GIS and spatial-based studies have rarely been conducted. Therefore, this study is proposed to investigate distribution pattern of TB, related socio- economic condition and relationship between the disease and the socio-economic factors in order to obtain supportive information for public health authorities to issue effective and targeted TB prevention and control strategies. This research was a cross-sectional retrospective population-based study conducting at local-administrative-organization (LAO) level in Si Sa Ket province. The analyses in this study were based on secondary data during the years 2004-2008. Numbers of registered TB cases were obtained from district TB clinics in Si Sa Ket hospitals. Population data were gathered from the Provincial Administration Department (DOPA) database, while the majority of information regarding socio-economic status of the study area was obtained from the Department of Community Development, Ministry of Interior. Additional economic data was retrieved from the National Economic and Social Development Board (NESDB) and the National Statistical Office (NSO) data sets. Microsoft Access 2010, Microsoft Excel 2010, SPSS 16.0, SmartPLS3, GeoDa 1.12.1.129, GWR4, and ArcGIS 10.1 software were used for the data analyses. Descriptive statistics was first applied to the collected data sets in order to portray an overview of TB incidence and socio-economic status of the study area. There were 12006 registered cases during 2004-2008. It was found that, more men were diagnosed with positive TB disease comparing to women. Additionally, in relation to adolescence, up to 27 times of the cases were elderlies aged 65 years or older. Geographically, uneven distribution of the cases was detected among districts. Khukhan had the highest amount of TB cases followed by Mueang Si Sa Ket, Kantharalak, Kanthararom and Uthumphon Phisai, while the lowest number of reported cases was in Bueng Bun district. Among the seven registration types, the "New" category was highest followed by "Transferred in", "Other", "Relapse", "Treatment after defaulted", "Treatment after failure" and "Not specified" cases. Abiding by the WHO's guideline, only "New" and "Relapse" cases were considered as the incidence in this research. Based on related literatures and availability of the data, eleven observed socio-economic variables were chosen as independent variables, while only age-gender-standardized morbidity ratio (SMR) of the TB incidence was considered the dependent variable for the following steps of the analyses. The observed socio-economic variables were "Alcoholics", "BasicCare", "Food", "Housing", "Sanitation", "AnnCare", "TapWater", "PopDen", "ElevMean", "NotFarmer" and "Road". Through bivariate correlation analysis, all eleven observed variables were found to be significantly correlated with TB SMR. The negative values of correlation indicate that TB SMR is likely to be high in the area with low socio- economic condition. Prior to relationship assessment, exploratory factor analysis and was performed for the socio-economic variables in order to eliminate multi-collinearity among the variables. Five latent factors including "Demography", "Facility", "Infrastructure", "Living Condition", and "Occupation" factors were generated. Additionally, partial least squares path modeling was utilized to assess the relationship between TB SMR and the latent factors. It was found that the latent factors could explain l3.6% of the total variance of the TB burden. To identify spatial pattern of TB burden and socio-economic condition, global Moran's I statistic, Anselin's local Moran's I (LISA) and Getis-Ord Oi" statistics were used. Spatial pattern of TB SMR was found to be clustered. Hot spots were mainly detected in the northeast, while cold spots were found in the southwestern districts. As for latent socio- economic factors, only "Facility", "Infrastructure" and "Occupation" were clustered. Spatial pattern of "Demography" and "Living Condition" were found to be random. To evaluate the relationships between TB incidence and the socio-economic factors, non- spatial regression (classic statistics), global spatial regression (autoregression), and local spatial regression analyses (geographically weighted regression) were applied. For each type of the relationship assessment, three sub-modeling approach including linear, logistic, and Poisson models were developed. It was found that the results of the spatial models, especially the local ones, were better fit according to the lower AICc value and higher R2. It could be concluded that TB incidence and its socio-economic determinants were unevenly distributed within the study area. Additionally, associations were found between the selected socio-economic factors and the TB incidence. These results could elucidate the causes of spatial affect in TB occurrence in Si Sa Ket province. Therefore, it is essential to establish localized policies and measures for TB prevention and control in the province based on the local effects of specific factors. |
| Year | 2018 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing (RS) |
| Chairperson(s) | Tripathi, Nitin Kumar |
| Examination Committee(s) | Souris, Marc;Sarawut Ninsawat;Anal, Anil Kumar;Mansor, Shattri Bin |
| Scholarship Donor(s) | Royal Thai Government Fellowship |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2018 |