Diagnostic tools

New Tool Created for Distinguishing Between Types of Community-Acquired Pneumonia

A mobile app featuring a 2-step diagnostic tool can distinguish bacterial from viral community-acquired pneumonia (CAP) and typical from atypical CAP in children, according to the results of a recent study.

To improve the ability to determine the cause of CAP in children and enable more appropriate use of antibiotics, the researchers built and validated a diagnostic tool based on clinical, analytical, and radiographic characteristics that distinguish viral from bacterial and typical from atypical CAP.

The tool resulted from an observational, multicenter, prospective cohort study that enrolled 495 children (aged 1 month-16 years) with CAP and consecutively admitted to 1 of 24 Spanish hospitals. The study’s main outcome measure was whether CAP was viral or bacterial. The researchers used 70% of the patients as the training set and 30% as the validation set to develop a score in 2 sequential steps.

Step 1 distinguished viral from bacterial CAP, and step 2 distinguished typical from atypical CAP. The researchers then selected optimal cutoff points to maximize the specificity of the score while aiming for a high sensitivity. They used multivariable logistic regression to calculate the weights of each variable.

A diagnosis of viral or bacterial CAP was made in 262 of 495 (53%) participants (median age, 2 years). About half of the participants were boys. The step 1 score classified cases of bacterial CAP with a sensitivity of 97% and a specificity of 48%.

Patients with bacterial CAP were then assessed in step 2 to determine whether the bacterium was typical or atypical. The step 2 score classified cases of typical bacterial CAP with a sensitivity of 100% and a specificity of 64%.

These findings led the researchers to conclude that, “This 2-step tool can facilitate the physician’s decision to prescribe antibiotics without compromising patient safety.”

The tool has been incorporated into a mobile app, the Pneumonia Etiology Predictor, that determines the probability that a child has one of the 4 types of CAP included in the study.


—Ellen Kurek



Tagarro A, Moraleda C, Domínguez-Rodríguez S, et al. A tool to distinguish viral from bacterial pneumonia. Pediatr Infect Dis J. 2022;41(1):31-36. doi:10.1097/INF/0000000000003340