Over half of pneumonia patients face diagnostic uncertainty, study finds

Reviewed by Danielle Ellis, B.Sc.Aug 5 2024

A national cohort study representing more than 2 million patient encounters found that more than half of all patients hospitalized and treated for pneumonia experienced discordant diagnoses from initial presentation to discharge.

The research highlights a significant prevalence of treatment for other diagnoses and frequent expressions of diagnostic uncertainty, underscoring the critical need for recognizing and addressing diagnostic uncertainty in pneumonia-related care practices.

The authors suggest that these findings call for enhanced attention to the diagnostic process and research that acknowledges uncertainty to improve patient outcomes in pneumonia treatment. This is important because pneumonia is the leading cause of death from infection. The study is published in Annals of Internal Medicine.

Researchers from the University of Utah and Salt Lake City VA Healthcare System used administrative data and natural language processing to study the accuracy of pneumonia diagnoses among over 2 million admissions to Veterans Administration hospitals between 2015 and 2022.

To explore the change in a pneumonia diagnosis across a hospitalization, the researchers classified each hospitalization according to concordance or discordance in pneumonia diagnoses between three states: initial diagnosis made in the ED, initial chest image reports, and discharge. They then measured the concordance of the admission diagnosis of pneumonia, a radiographic finding consistent with pneumonia, and a discharge diagnosis of pneumonia.

They found that 36% of patients had an admitting diagnosis of pneumonia but no discharge diagnosis, and 33% had a discharge diagnosis of pneumonia but not an admission diagnosis. Discordance between a diagnosis of pneumonia and chest radiograph findings was also common.

An accompanying editorial from the University of Connecticut School of Medicine says that clinicians accept diagnostic uncertainty in pneumonia because treating all possibilities has not been shown to be inferior to any available alternative. While CT for all patients with suspected pneumonia may improve accuracy, it still would not likely change the treatment, as pneumonia symptoms and treatment overlap with several other conditions, including heart failure and COPD.

The author agrees that more research is needed and suggests that studies should quantify the level of harm related to an incorrect diagnosis of pneumonia, as doing so would better define the level of importance of interventions to address the issue.