Objective: The aim of study is to determine the correlation between different pneumonai severity classifications, such as pneumonia severity index (PSI), CURB65, Turkish Thoracic Society Pneumonia Severity Classification (TTSPSC) (2002 and 2009), according to the indications for hospitalization. Material and Methods: Hundred and forty four pneumonia cases hospitalized and treated in our clinic between January 2008 and December 2009 were included in this study. Demographic properties, clinic, radiologic and laboratory findings and treatments of cases were recorded from the files. All cases were grouped according to PSI, CURB65, TTSPSC (2002 and 2009) respectively. The correlation of these groups with each other according to the indications for hospitalization was examined. The efficiency of different pneumonia scoring systems in determination of the possibility of intensive care indication or death in community-acquired pneumonia was assessed. Results: Sixty one (42.4%) female, 83 (57.6%) male 144 pneumonia cases aged 63.4±19.2 (min: 15; max: 95) years were included in the study. Thirty three cases who need hospitalization according to PSI do not require e hospitalization according to CURB65. On the other hand, 6 cases who need hospitalization according to CURB65 do not require hospitalization according to PSI. CURB65 and PSI correlated in terms of hospitalization in 105 cases. CURB65 and PSI were statistically correlated with each other ( $kappa$ =0.48, p<0.001). TTSPSC 2002 was not correlated with CURB65 and PSI ( $kappa$ =0.12, p=0.54; $kappa$ =-0.5, p=0.53) but TTSPSC 2009 was correlated with CURB65 and especially strongly correlated with PSI ( $kappa$ =0.43, p<0.001;$kappa$ =0.78, p<0.001). CURB 65, PSI ve TTSPSC 2009 were statistically significant in determination of death probability of CAP cases (p=0.016, p=0.021, p=0.05 respectively). CURB 65 and TTSPSC 2009 were also statistically significant in determinating the probability of intensive care unit indication at CAP (p=0.023, p=0.015 respectively). However, PSI was not statistically significant in determinating the probability of intensive care unit indication at CAP (p=0.076). Conclusion: It is seen that currently used three pneumonia severity classifications (CURB65, PSI, TTSPSC 2009) were correlated with each other according to the indications for hospitalization. Since CURB 65 is efficient at determining the probability of intensive care unit indication and of death and also the scoring criteria are fewer and easy to application even at the first line medical centers, of these pneumonia scoring systems it can be the first choice in the assessment of hopitalization indication in CAP. (Turk Toraks Derg 2012; 13: 158-62)