MARKOV CHAIN MODELING OF PNEUMONIA CASES IN THE BUILSA NORTH MUNICIPALITY OF THE UPPER EAST REGION OF GHANA

Abstract

Pneumonia is an infection of the lungs. It causes inflammation of the alveoli making it fill with fluid, causing difficulty in breathing and reduces oxygen intake. It is the leading cause of deaths in children worldwide. In this study, secondary data on monthly pneumonia cases collected from the disease control unit from 2015 to 2020 of the Builsa North Municipality was modeled using discrete-time Markov chain modeling. The data was grouped into low, moderate and high states of the pneumonia cases. Chi square test of independence indicated that the data followed the first-order Markov chain assumption. Transition probabilities and relevant disease metrics were estimated for the disease. The model revealed that there is higher chances of moderate pneumonia cases to be recorded in the Builsa North Municipality in the long run than low and high pneumonia cases with a probability of approximately 51%. The expected length of time the pneumonia cases are expected to stay in the low and moderate states were estimated to be approximately 3 months each and 2 months for the high state. It was known from the estimation that it will take the states twenty-eight (28) months to be in equilibrium. The model also revealed that in the long-run the municipality will record on average approximately 116 pneumonia cases.