Introduction. Diagnosing community-acquired pneumonia (CAP) in the elderly remains a clinical challenge for various reasons. The clinical manifestation in the elderly is not frank and atypical manifestations, e.g. falls, decrease of functional status and food intake or urinary incontinence, may be present. These reasons may be associated with under or over diagnosis, which consequently contribute to the higher observed mortality rate in the elderly population with CAP. Study about factors related to diagnosis of CAP in the elderly was ra rely performed. Methods. From January to October 2010, 158 elderly patients suspected of having pneumonia at RSCM were registered. Relationship between clinical, laboratory and radiologic factors which consist of classic manifestations (cough, productive cough, dyspnea, fever, rales, leucocytosis, infiltrates) and atypical manifestations (decrease of intake and functional status, falls, urinary incontinence) with diagnosis community acquired pneumonia were analyzed. Receiver operating characteristics analysis of C-reactive protein was performed to find its association with diagnosis of pneumonia. Results. Of 158 subject, 106 were confirmed of having pneumonia. Final model of multiple logistics regression analysis revealed three factors: cough (p<0,0001), rales (p=0,02) and infiltrate (p<0,0001) related to diagnosis of pneumonia. All four atypical manifestations were proved unrelated with diagnosis of pneumonia. The area under the ROC curve for c-reactive protein was 0,57 (95% CI 0,47- 0,66). Conclusions. Factors related with diagnosis of community-acquired penumoni in the elderly are cough, rhales and infiltrates. All four atypical manifestations are proven unrelated with diagnosis of pneumonia. C-reactive protein does not predict diagnosis of CAP in the eldery.