Find MODELLING COMMUNITY EFFECTS ON ANEMIA IN SCHOOL-AGED CHILDREN IN MALAWI USING A MULTILEVEL REGRESSION MODEL Publication: MODELLING COMMUNITY EFFECTS ON ANEMIA IN SCHOOL-AGED CHILDREN IN MALAWI USING A MULTILEVEL REGRESSION MODEL
| datacite.subject.fos | Natural sciences | |
| dc.contributor.author | Mshali, Glory Gondwe | |
| dc.date.accessioned | 2025-01-21T13:16:48Z | |
| dc.date.available | 2025-01-21T13:16:48Z | |
| dc.date.issued | 2023-08-01 | |
| dc.description | Thesis submitted to the Department of Mathematical Sciences, Faculty ofScience, in partial fulfilment of the requirement for degree of Master of Science(Biostatistics) | |
| dc.description.abstract | Anemia is a serious health problem in Malawi that usually results from poor nutrition,infection, or chronic diseases. Anemia in school-aged children 5-14 years has been associated with poor cognitive performance, impaired immunity and decrease working capacity. Therefore the present study focused in modeling community effects on anemia in school-aged children 5-14 years old in Malawi using multilevel logistic regression analysis. The study used Cross-sectional data from the Malawi Demographic Health Survey (MDHS) 2015-16 and the Malawi Micronutrient Survey(MNS) (2015-16). The statistical models that suited the hierarchical data such as variance components model, random intercept model and random coefficients model were used in the analysis. The Log Likelihood approach was used to estimate the fixed effect and random effects in the multilevel analysis. The results of the descriptive statistics showed that effect on anemia in school-aged children 5-14 years was 19%. Performing the logistic regression analysis showed that: place of residence, age of child, education of child, source of drinking water, inflammation, wealth index and head of household sex had a significant effect on anemia in school-aged children. Multilevel logistic regression model better suited the hierarchical clustered data with higher values of log likelihood estimates of -348.65 verses -355.63 for logistic regression. The random intercept model, AIC of 730.53 and random coefficient model of 733.50 did not differ much in variations and were both treated as the better fit model as compared to variance component model with AIC of 769.69. Therefore, anemia in school-aged children 5-14 years still remains a challenge in Malawi and that Multilevel modeling identified considerable community variations in its distribution which requires stakeholders, policy makers and the public health to pay attention to these significant effects on anemia. | |
| dc.identifier.uri | https://dspace.unima.ac.mw/handle/123456789/626 | |
| dc.language.iso | en | |
| dc.schoolscentersoptions | b0e8094e-c230-4f28-8ae9-b0fc60729932 | |
| dc.subject | Community | |
| dc.subject | Multilevel Regression Model | |
| dc.subject | Anaemia | |
| dc.subject | Public health | |
| dc.subject | Chronic diseases | |
| dc.subject | Infection | |
| dc.subject | Inflammation | |
| dc.subject | Source of drinking water | |
| dc.subject | Education of child | |
| dc.subject | Impaired immunity | |
| dc.supervisor | 2b902e2b-261b-49c2-8ceb-2bc466a6b85a | |
| dc.title | MODELLING COMMUNITY EFFECTS ON ANEMIA IN SCHOOL-AGED CHILDREN IN MALAWI USING A MULTILEVEL REGRESSION MODEL | |
| dc.type | text::thesis::master thesis | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | University of Malawi |