Logistic regression analysis of environmental and other variables and incidences of tuberculosis in respiratory patients

Abstract

The objective of this study was to examine the association of 14 variables with TB in respiratory patients. The variables included: urban/rural, persons in 1200 sqft area, TB in family, crowding, smoking (family member), gender, age, education, smoking, workplace, kitchen location, cooking fuel, ventilation, and kerosene uses. Eight hundred respiratory patients were tested for sputum positive pulmonary TB; 500 had TB and 300 did not. An analysis of the unadjusted odds ratio (UOR) and adjusted OR (AOR) was undertaken using logistic regression to link the probability of TB incidences with the variables. There was an inconsistency in the significance of variables using UOR and AOR. A subset model of 4 variables (kerosene uses, ventilation, workplace, and gender) based on significant AOR was adjudged acceptable for estimating the probability of TB incidences. Uses of kerosene (AOR 2.62 (1.95, 3.54)) consistently related to incidences of TB. It was estimated that 50% reduction in kerosene uses could reduce the probability of TB by 13.29% in respiratory patients. The major recommendation was to replace kerosene uses from households with a supply of clean fuel like liquid petroleum or natural gas and rural electrification.

Heart Failure and PAHs, OHPAHs, and Trace Elements Levels in Human Serum: Results from a Preliminary Pilot Study in Greek Population and the Possible Impact of Air Pollution

“…In logistic regression, the relationship between a binary dependent variable, for example, the occurrence of a phenomenon or not, with independent variables, which affect that phenomenon, is assessed, as generally used in medical and epidemiological studies. Although logistic regression has many similarities with linear regression, the estimation of variables’ coefficients is performed by the maximum likelihood technique [ 148 ]. Under this prism, combination of logistic regression with PCA could reveal the probability of each factor to be associated with the occurrence of heart failure.…”

Eksplorasi Dan Analisis Regresi Logistik Terhadap Kondisi Sungai Tercemar Limbah Di Desa/Kelurahan Provinsi Dki Jakarta Indonesia

“…Selanjutnya, analisis regresi logistik diterapkan dalam upaya pengidentifikasian faktor-faktor yang diduga memengaruhi sungai-sungai di desa/kelurahan di DKI Jakarta tercemar limbah. Analisis ini dipilih dikarenakan sangat berguna dalam pemodelan data lingkungan, seperti pada penelitian Rahardiantoro et al, (2019) tentang pemodelan untuk prediksi data gangguan spektrum autis (autistic spectrum disorder), Pathak et al, (2020) tentang identifikasi peubah berpengaruh pada kasus tuberkulosis, dan Adiat et al, (2020) tentang pemodelan untuk prediksi kerentanan air tanah di pertambangan emas. Praktiknya, analisis regresi logistik diterapkan dengan perangkat lunak R (Hilbe, 2015), dengan peubah tak bebas yang digunakan adalah kondisi sungai tercemar atau tidak tercemar.…”