When the presence of health care-associated pneumonia
criteria are considered as a diagnostic test, the positive
predictive value for resistant infections in the Shorr et al12
study was only 58.2%. Sensitivity, specificity, and negative
predictive value were 86.9%, 48.6%, and 81.7%, respectively.12
A 2010 study by Schreiber et al31 encountered
similar results, showing that health care-associated pneumonia
as a tool for predicting resistant infections had a
sensitivity and specificity of 78.3% and 56.2%, and positive
and negative predictive values of 45.2% and 84.9%.31 These
results emphasize the lackluster ability of health care-associated
pneumonia criteria to predict patients who will have
pneumonias caused by drug-resistant pathogens and indicate
that a greater understanding of risk factors for specific
pathogens may be more clinically relevant than the health
care-associated pneumonia classification system.