A retrospective cross-sectional hospital-based survey was conducted in Northern-Tanzania using a sample size of 384 CT images of adults who underwent brain CT scans in three referral hospitals. CT images were evaluated using diagonal brain fraction (DBF) method to determine the presence of brain atrophy. Data for other covariates were also collected. A prevalence of 60.67% for brain atrophy and 35% for the associated with confusion state was reported. Association between confusion state and brain atrophy was statistically significant (2 = 21.954, p<0.001). Brain atrophy was prognosticated by : age (adjusted OR: 1.11; 95% CI [1.05, 1.20], p < 0.001), smoking (adjusted OR: 6.97; 95% CI [2.12, 26.19], p < 0.001), alcohol-consumption (adjusted OR: 11.87; 95% CI [3.44, 40.81], p < 0.001), hypertension (adjusted OR: 61.21; 95 CI [15.20, 349.43],  p < 0.001), type-2 diabetes mellitus (adjusted OR: 15.67; 95% CI [5.32, 52.77], p < 0.001) and white matter demyelination (adjusted OR: 13.45; 95% CI [4.66, 44.25], p < 0.001). There is high prevalence of brain atrophy and associated confusion state among hospitalized adults in Northern Tanzania. Reported prognostic factors for brain atrophy such as age, smoking, alcohol consumption, hypertension, type-2 diabetes mellitus, and white matter demyelination could help focus interventions in this area.

Leticia Kalumbilo is a Masters student from NM-AIST, Tanzania. She has participated in different national scientific conferences (Theme; eradication of non-communicable disease). Also, she is a secretary of Science, Technology, Engineering, Mathematics, and Innovation (STEMi) project at NM-AIST. Currently she is working as co-researcher in imaging brain technology research at NM-AIST.