Corey Diamond, PharmD
Over the past decade, reported morbidity and mortality rates from community acquired pneumonia (CAP) have stagnated. While treatments have improved, tightening the gap between guideline-directed CAP treatment and general medical practice using electronic decision support systems remains unexplored.
A recent study published by Dean and colleagues in the American Journal of Respiratory and Critical Care Medicine from March of 2022, investigated the deployment of an electronic pneumonia clinical decision support tool (ePNa) to improve physician adherence to the 2007 and 2019 American Thoracic Society/Infectious Diseases Society of America pneumonia treatment guidelines. Researchers found that ePNa deployment was associated with improved processes of care and lower mortality in patients diagnosed with CAP.
Unique design with interesting results
Dean and colleagues conducted a cluster-controlled trial, which looked at the outcomes of patients with CAP before and after deployment of an ePNa program. ePNa was used across 6 clusters of 16 hospital emergency departments (EDs) at 2-month intervals between 2017 and 2018. The intention-to-treat analysis included 6,848 patients, of whom 4,536 were seen before, and 2,312 were seen after ePNa deployment.
Confounders were controlled for using a step-wedge cluster design but were not randomized due to ethical concerns that arose from prior research the same authors conducted, which showed ePNa implementation may present a mortality benefit. The sequential roll out of the ePNa in clusters was required to facilitate the intensive education, monitoring, and feedback required of the program.
Patients were included in the analysis if they were older than 18 years and had radiographic pneumonia on ED chest imaging, plus a discharge diagnosis of pneumonia. The study’s primary analysis used a mixed-effects model to evaluate the relationship between ePNa deployment and severity-adjusted 30-day mortality.
Secondary statistics included observed trends in antibiotic use, patient disposition (level of care needed throughout treatment), and physician ePNa use.
The regression analysis revealed that 30-day all-cause mortality was 8.6% in clusters before ePNa deployment versus 4.8% after ePNa deployment.
ePNa was used in 67% of eligible patients with CAP, and more so in larger hospitals. Its’ deployment was associated with a statistically significant increase in guideline-concordant prescribing from 79.5% to 87.9%. Additionally, there was a significant reduction in anti-MRSA agent use, time to first antibiotic use, and inpatient disposition.
What is ePNa?
Dean and colleagues’ study used an ePNa that integrated pneumonia detection with a management tool, which presented needed information to ED clinicians assessing patients with suspected pneumonia. The ePNa software first automatically identified patients with possible CAP, based on documented presentation in the electronic health record (EHR), using symptomatic, radiographic, and laboratory evidence. The ePNa software then calculated the percent probability of pneumonia and alerted ED clinicians if the probability of pneumonia was greater than 40%.
The ED clinicians could then opt to launch ePNa through the hospital’s EHR. ePNa would then calculate the patient’s pneumonia illness severity using validated severity factors—such as eCURB score, PaO2/FiO2 ratio, sCAP score, and pleural effusion size. If the combined severity factors warranted a higher level of care, the ePNa algorithm triggered a recommendation for either a hospital admission or an intensive care admission.
Finally, ePNa calculated the patient’s risk of antibiotic-resistant pathogens using risk factor logic derived from the American Thoracic Society/Infectious Diseases Society of America pneumonia treatment guidelines. The patient’s risk score prompted ePNa to provide recommendations for antibiotic coverage, with coverage recommendations becoming broader as the patient’s drug resistance risk score increased.
Considerations
While Dean and colleagues’ ePNa technology is encouraging, the study itself has notable limitations. For instance, there was a significance decrease in median age of patients after ePNa use. Similarly, the pre-ePNa patient clusters had significantly more comorbidities, including higher rates of chronic renal disease, chronic heart disease, COPD, and diabetes.
Considering the baseline demographic imbalances in the comparison groups, the study’s results are much more susceptible to bias. Despite the use of adjusted regression model, the decline in pre-existing morbidity in the patient population after ePNa implementation may have confounded the reduction in mortality observed.
Overall, however, the mortality results of Dean and colleagues’ study are encouraging and have the potential to smooth transitions of care for patients with CAP. ePNa technology may provide the support that general medicine needs to compartmentalize patient risk and allow future studies to discover which interventions drive which benefits. ■