Risk factors for community-acquired pneumonia among adults in Kenya: a case-control study
Muthumbi E, Lowe BS, Muyodi C, Getambu E, Gleeson F, Scott JAG
Pneumonia (Nathan). 2017;9
BACKGROUND: Pneumonia is a leading cause of morbidity and mortality among adults worldwide; however, the risk factors for community-acquired pneumonia in Africa are not well characterized. METHODS: The authors recruited 281 cases of community-acquired pneumonia and 1202 hospital controls among patients aged >/=15 years who attended Kilifi District Hospital/Coast Provincial General Hospital in Kenya between 1994 and 6. Cases were admissions with an acute illness with >/=2 respiratory signs and evidence of consolidation on a chest radiograph. Controls were patients without signs of pneumonia, frequency matched by age, sex and hospital. Risk factors related to socio-demographic factors, drug use, clinical history, contact patterns and exposures to indoor air pollution were investigated by questionnaire, anthropometric measurements and laboratory assays. Associations were evaluated using a hierarchical logistic regression model. RESULTS: Pneumonia was associated with human immunodeficiency virus (HIV) infection (Odds Ratio [OR] 2.06, 95% CI 1.44-3.08), anemia (OR 1.91, 1.31-2.74), splenomegaly (OR 2.04, 95% CI 1.14-3.41), recent history of pneumonia (OR 4.65, 95% CI 1.66-12.5), history of pneumonia >2 years previously (OR 17.13, 95% CI 5.01-60.26), coryza in the 2 weeks preceding hospitalization (OR 2.09, 95% CI 1.44-3.03), current smoking (2.19, 95% CI 1.39-3.70), use of khat (OR 3.44, 95% CI 1.72-7.15), use of snuff (OR 2.67, 95% CI 1.35-5.49) and contact with several animal species. Presence of a Bacillus Calmette-Guerin (BCG) scar was associated with protection (OR 0.51, 95% CI 0.32-0.82). The risk factors varied significantly by sex. CONCLUSION: Pneumonia in Kenyan adults was associated with global risk factors, such as HIV and smoking, but also with specific local factors like drug use and contact with animals. Intervention strategies should account for sex-specific differences in risk factors.