Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia
Dengue viral infection is a global health problem that has spread exponentially across the tropical and sub-tropical regions of the world. Malaysia is one of the affected tropical regions that has experienced a significant rate of morbidity and mortality in the last few years. There was an increase...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2019
|
Subjects: | |
Online Access: | http://psasir.upm.edu.my/id/eprint/83191/1/FS%202019%2051%20ir.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-upm-ir.83191 |
---|---|
record_format |
uketd_dc |
institution |
Universiti Putra Malaysia |
collection |
PSAS Institutional Repository |
language |
English |
advisor |
Shohaimi, Shamarina |
topic |
Bayesian statistical decision theory - Case studies Dengue viruses - Malaysia Dengue - Malaysia |
spellingShingle |
Bayesian statistical decision theory - Case studies Dengue viruses - Malaysia Dengue - Malaysia Ajibola, Lamidi-Sarumoh Alaba Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
description |
Dengue viral infection is a global health problem that has spread exponentially across the tropical and sub-tropical regions of the world. Malaysia is one of the affected tropical regions that has experienced a significant rate of morbidity and mortality in the last few years. There was an increase of 63.9% morbidity rate and 71.4% mortality rate of dengue incidence in Malaysia between 2013 and 2015 and more than half of these incidences occurred in the state of Selangor. There are many complex factors that contribute to the incidence of dengue, for example, climatic condition, environmental factors, socioeconomic status, socio-demographic variables, human behaviour, and health belief pattern among others. Among the various factors aforementioned, human behaviour tends to contribute immensely to the incidence of dengue because the vectors spreading dengue virus depends on the human environment and human behaviour for their survival and sustenance. Therefore, quantifying the awareness about the dengue vectors, dengue infection and preventive practices can be used to curtail the spread of the infection and reduce the vectors. The main objective of this research is the novel application of Partial least square path analysis (PLS-PA) to explore the latent sub-constructs of knowledge, attitude, and practices (KAP) on dengue via R programming language. Also, the application of Bayesian network (BN) to assess the influence of socio-demographic variables on dengue KAP in some selected areas of Selangor, Malaysia. Data of socio-demographic variables, medical history about dengue, knowledge about dengue fever and its vectors, attitude towards dengue incidence, elimination of dengue vectors and practices to eradicate the dengue infection and vectors was collected using a structured validated bilingual questionnaire. The data collected was used to learn the structure of BN via some known algorithms using R programming language. Based on related literature on KAP regarding dengue studies done in Malaysia from 1986 to 2017, a framework of handcraft BN was also formulated with dengue KAP expert opinion. The actual KAP on dengue that was inadequate among the respondents were verified and revealed with PLS-PA. The knowledge that was inadequate among the respondents were knowledge on primary and secondary transmission of dengue fever, knowledge on possible breeding sites of dengue vectors and knowledge on the severity of dengue fever and vectors. The attitude that was deficient among the respondents were the attitude towards the elimination of dengue vectors and the attitude towards severity and prevention of dengue fever. The preventive practice that was inadequate among the respondents was the elimination of adult mosquitoes. Knowledge, attitude and practice Bayesian network model (KAPBNM) was used to visualize and quantify the pattern of KAP on dengue infection and vectors given the socio-demographic variables of the respondents. Quantifying the KAP dengue based socio-demographic variables from KAPBNM, individuals with primary school education training had low knowledge on dengue. The Chinese and Indians working in the private sector had poor attitude and individuals living in flats and apartment buildings had poor preventive practices. Using a BN to model dengue KAP provides the best effective approach for approximate reasoning and classification because of its capability to encode dependencies among socio-demographic variables and the latent construct of dengue KAP. It also narrows down the search for individuals who need to improve the KAP regarding dengue. There was a clear confirmation that the novel application of the PLS-PA and BN to KAP on dengue study has shown improved results and reduces vagueness of intensifying KAP regarding dengue programs in some selected areas in Selangor, Malaysia. In conclusion, the PLS-PA presented the exact dengue KAP that is insufficient among the respondents and the approximate reasoning with BN showed highest probability of individuals who has low KAP on dengue based on socio-demographic variables attributed each latent constructs of dengue KAP concurrently. |
format |
Thesis |
qualification_level |
Doctorate |
author |
Ajibola, Lamidi-Sarumoh Alaba |
author_facet |
Ajibola, Lamidi-Sarumoh Alaba |
author_sort |
Ajibola, Lamidi-Sarumoh Alaba |
title |
Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
title_short |
Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
title_full |
Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
title_fullStr |
Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
title_full_unstemmed |
Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia |
title_sort |
bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in selangor, malaysia |
granting_institution |
Universiti Putra Malaysia |
publishDate |
2019 |
url |
http://psasir.upm.edu.my/id/eprint/83191/1/FS%202019%2051%20ir.pdf |
_version_ |
1747813357344260096 |
spelling |
my-upm-ir.831912022-01-10T03:15:59Z Bayesian network of influence of sociodemographic variables on dengue related knowledge, attitude, and practices in selected areas in Selangor, Malaysia 2019-08 Ajibola, Lamidi-Sarumoh Alaba Dengue viral infection is a global health problem that has spread exponentially across the tropical and sub-tropical regions of the world. Malaysia is one of the affected tropical regions that has experienced a significant rate of morbidity and mortality in the last few years. There was an increase of 63.9% morbidity rate and 71.4% mortality rate of dengue incidence in Malaysia between 2013 and 2015 and more than half of these incidences occurred in the state of Selangor. There are many complex factors that contribute to the incidence of dengue, for example, climatic condition, environmental factors, socioeconomic status, socio-demographic variables, human behaviour, and health belief pattern among others. Among the various factors aforementioned, human behaviour tends to contribute immensely to the incidence of dengue because the vectors spreading dengue virus depends on the human environment and human behaviour for their survival and sustenance. Therefore, quantifying the awareness about the dengue vectors, dengue infection and preventive practices can be used to curtail the spread of the infection and reduce the vectors. The main objective of this research is the novel application of Partial least square path analysis (PLS-PA) to explore the latent sub-constructs of knowledge, attitude, and practices (KAP) on dengue via R programming language. Also, the application of Bayesian network (BN) to assess the influence of socio-demographic variables on dengue KAP in some selected areas of Selangor, Malaysia. Data of socio-demographic variables, medical history about dengue, knowledge about dengue fever and its vectors, attitude towards dengue incidence, elimination of dengue vectors and practices to eradicate the dengue infection and vectors was collected using a structured validated bilingual questionnaire. The data collected was used to learn the structure of BN via some known algorithms using R programming language. Based on related literature on KAP regarding dengue studies done in Malaysia from 1986 to 2017, a framework of handcraft BN was also formulated with dengue KAP expert opinion. The actual KAP on dengue that was inadequate among the respondents were verified and revealed with PLS-PA. The knowledge that was inadequate among the respondents were knowledge on primary and secondary transmission of dengue fever, knowledge on possible breeding sites of dengue vectors and knowledge on the severity of dengue fever and vectors. The attitude that was deficient among the respondents were the attitude towards the elimination of dengue vectors and the attitude towards severity and prevention of dengue fever. The preventive practice that was inadequate among the respondents was the elimination of adult mosquitoes. Knowledge, attitude and practice Bayesian network model (KAPBNM) was used to visualize and quantify the pattern of KAP on dengue infection and vectors given the socio-demographic variables of the respondents. Quantifying the KAP dengue based socio-demographic variables from KAPBNM, individuals with primary school education training had low knowledge on dengue. The Chinese and Indians working in the private sector had poor attitude and individuals living in flats and apartment buildings had poor preventive practices. Using a BN to model dengue KAP provides the best effective approach for approximate reasoning and classification because of its capability to encode dependencies among socio-demographic variables and the latent construct of dengue KAP. It also narrows down the search for individuals who need to improve the KAP regarding dengue. There was a clear confirmation that the novel application of the PLS-PA and BN to KAP on dengue study has shown improved results and reduces vagueness of intensifying KAP regarding dengue programs in some selected areas in Selangor, Malaysia. In conclusion, the PLS-PA presented the exact dengue KAP that is insufficient among the respondents and the approximate reasoning with BN showed highest probability of individuals who has low KAP on dengue based on socio-demographic variables attributed each latent constructs of dengue KAP concurrently. Bayesian statistical decision theory - Case studies Dengue viruses - Malaysia Dengue - Malaysia 2019-08 Thesis http://psasir.upm.edu.my/id/eprint/83191/ http://psasir.upm.edu.my/id/eprint/83191/1/FS%202019%2051%20ir.pdf text en public doctoral Universiti Putra Malaysia Bayesian statistical decision theory - Case studies Dengue viruses - Malaysia Dengue - Malaysia Shohaimi, Shamarina |