Research advances: Reducing risk to public health and patient safety by improving antibiotic prescribing practices in cellulitis

Michael Pulia, MD, PhD

Dr. Michael Pulia leads the Emergency Care for Infectious Diseases (EC-ID) health services research program at the BerbeeWalsh Department of Emergency Medicine. Much of Dr. Pulia’s research is focused on diagnostic uncertainty and its role in driving inappropriate antibiotic prescribing for infectious diseases, such as COVID-19, urinary tract infections, cellulitis, and pneumonia.

In emergency and acute care settings, accurately diagnosing the cause of infectious diseases is paramount for clinicians to provide timely and appropriate medical care, including antibiotics. “Infectious diseases are really the final frontier of diagnosis,” says Dr. Pulia. “This is the last remaining set of very common conditions where diagnosis remains highly clinical given there are no specific tests that can rapidly and reliable confirm the presence of a bacterial infection.”

Advancements in infectious disease diagnostics are necessary to improve health outcomes for patients, reduce healthcare costs, and to deter the misuse of antibiotics. Prescribing inappropriate antibiotics poses a significant threat to patient safety due to the risk of serious adverse drug events and damage to the microbiome. Additionally, antibiotic overuse is directly linked to antimicrobial resistance (AMR), which the World Health Organization (WHO) has identified as a top global public health threat.

Latest Research

For conditions like cellulitis and pneumonia, existing diagnostic modalities are insufficient for health care providers to distinguish between bacterial and viral infections, or noninfectious etiologies. In broad terms, this leads to “just in case” use of antibiotics to “catch” possible bacterial infections, which often are not even present.

Dr. Pulia and the EC-ID research team are working to understand and improve the diagnosis of these conditions, as highlighted by two recent publications in the Journal of the American Medical Association (JAMA) Network of medical journals.

Cellulitis: improving diagnostic accuracy through thermal imaging

One method of improving antibiotic stewardship is to develop tools to improve diagnostic accuracy. The EC-ID team is moving the science of infectious disease diagnosis forward as demonstrated by their new findings in JAMA Dermatology (“Validation of Thermal Imaging and the ALT-70 Prediction Model to Differentiate Cellulitis from Pseudocellulitis”). This study highlights how thermal imaging of the skin’s surface has the potential to improve diagnosis of cellulitis, a common bacterial skin infection that causes redness, swelling, and pain in the infected area of the skin. If untreated, it can spread and cause serious health problems. Cellulitis is one of the most common dermatologic reasons for hospitalization.

Like many infectious diseases, cellulitis does not have a “gold standard” test to distinguish between inflammation of the skin that is bacterial (cellulitis) versus caused by noninfectious mimics (pseudocellulitis). Many common clinical characteristics like expanding redness of the skin, warmth, tenderness and edema are nonspecific and commonly found in other dermatologic conditions. A 2017 study in JAMA Dermatology by Weng, Raff, Cohen et al found that diagnostic uncertainty in suspected cellulitis contributes to a misdiagnosis rate of 30% and concluded that the resulting unnecessary hospital admissions and overuse of antibiotics is a threat to patient safety and public health.

The Pulia team’s study was the largest to date validating that patients with cellulitis have a significantly higher skin surface temperature than patients with pseudocellulitis. The research team suggests that surface thermal imaging, alone or in combination with commonly assessed clinical features such as age or white blood cell count, could significantly decrease overdiagnosis of cellulitis.

Their work builds upon foundational research in this area and provides further evidence that skin surface temperatures can provide actionable, reliable, and objective data to inform cellulitis risk.

“I feel incredibly fortunate to have the opportunity to conduct research that has the potential to improve antibiotic prescribing for millions of patients seen in the emergency department,” says Dr. Pulia. “It really is full circle to have this recent JAMA Dermatology publication. A pilot study I conducted over 10 years ago examining MRSA swabs in patients with skin infections was what initially sparked my interest in pursuing a research career.”

Example of mild cellulitis on the lower leg.

Dr. Pulia’s future plans for his cellulitis research include expanding on this single-site study by replicating the team’s findings in larger, diverse patient populations and developing a user-friendly clinical scoring system that can help clinicians more accurately identify those patients with true cellulitis. The researchers will also focus on designing a decision-making tool to help patients understand the risk of antimicrobial resistance so they can make informed choices about using antibiotics.

Dr. Pulia and his team are not alone in their enthusiasm for this line of research. In an editorial response by Li and Mostaghimi, both Boston-based dermatologists, the authors note that “[the Pulia team’s] contribution reflects the latest step in a quest for improved diagnosis and treatment of cellulitis and reflects both the opportunity and the challenge for clinicians moving forward.”

Pneumonia: quantifying the prevalence of inappropriate antibiotic prescribing

Dr. Pulia also contributed to a recent study published in JAMA Internal Medicine that describes the prevalence of inappropriate antibiotic prescribing for community-acquired pneumonia  (“Inappropriate Diagnosis of Pneumonia Among Hospitalized Adults”).

Led by researchers at the University of Michigan and University of Utah, this cohort study found that of all patients hospitalized for pneumonia across a large network of Michigan hospitals, 12% were inappropriately prescribed antibiotics. Older adults, those with dementia, and those presenting with altered mental status were at the highest risk for misdiagnosis and subsequent antibiotic treatment.

With the advantage of the large, 48-hospital network involved in this study, the authors were able to hypothesize that one factor likely contributing to inappropriate antibiotic prescribing for pneumonia is diagnosis uncertainty because pneumonia symptoms are nonspecific and often overlap with other cardiopulmonary diseases.

While the number of factors contributing to inaccurate diagnoses of pneumonia is likely multifaceted and was beyond the scope of this study, Dr. Pulia states that “these findings reinforce the need and urgency for more research on diagnostic tools and tests to minimize diagnostic uncertainty for pneumonia.”

Next Steps

These two studies highlight the importance of research that seeks to understand the prevalence of inappropriate antibiotic prescribing in specific infectious diseases as a result of misdiagnosis and to develop tools to improve diagnostic accuracy. This research has the potential to improve antibiotic prescribing for millions of patients seen in emergency departments, including patients with serious viral infections like COVID-19 and influenza.

“It is always rewarding to see the culmination of our work published broadly,” says Becky Schwei, MPH, a researcher with the EC-ID lab and PhD candidate with the UW Institute for Clinical and Translational Research. While the diagnostic process is complex, the team shared their enthusiasm for the potential implications of their work.

“We hope to build on these promising findings,” says Schwei, “so that we can better understand the factors driving diagnostic uncertainty and develop solutions to promote appropriate antibiotic prescribing for all patients.”


Michael Pulia, MD, PhD, is an Associate Professor with Tenure at the BerbeeWalsh Department of Emergency Medicine, University of Wisconsin School of Medicine and Public Health, as well as Director of Emergency Medicine Antimicrobial Stewardship. He is a Fellow of the Infectious Disease Society of America (FIDSA), the nation’s leading infectious diseases professional society, and was among the first emergency medicine physicians in the U.S. to be named a National Academy of Medicine Scholar in Diagnostic Excellence. He is engaged in collaborative, interdisciplinary antimicrobial stewardship initiatives in national forums and is a recognized leader in the field of emergency care for infectious diseases.

This edition of Research Advances is based on the following publications:

Pulia MS, Schwei RJ, Alexandridis R, et al. Validation of Thermal Imaging and the ALT-70 Prediction Model to Differentiate Cellulitis From Pseudocellulitis. JAMA Dermatol. Published online March 27, 2024. doi:10.1001/jamadermatol.2024.0091

Li DG, Mostaghimi A. The Quest for a More Accurate Diagnosis of Cellulitis: In Dreams Begin Responsibilities. JAMA Dermatol. Published online March 27, 2024. doi:10.1001/jamadermatol.2024.0089

Weng QY, Raff AB, Cohen JM, et al. Costs and Consequences Associated With Misdiagnosed Lower Extremity Cellulitis. JAMA Dermatol. 2017;153(2):141–146. doi:10.1001/jamadermatol.2016.3816

Gupta AB, Flanders SA, Petty LA, et al. Inappropriate Diagnosis of Pneumonia Among Hospitalized Adults. JAMA Intern Med. Published online March 25, 2024. doi:10.1001/jamainternmed.2024.0077


The research mentioned in this article was funded by the following:

Cellulitis study: supported by grant KO8HS024342 (PI: Pulia) from the Agency for Healthcare Research and Quality and by grant UL1TR002373 (PI: Pulia) from the National Center for Advancing Translational Sciences.

Pneumonia study: funded in part by the Gordon and Betty Moore Foundation through grants GBMF8839 and 8839.01 to Drs. Gupta and Vaughn. This project was also supported by grant funding from the Claude D. Pepper Older Americans Independence Center and the Blue Cross Blue Shield of Michigan and Blue Care Network as part of the Blue Cross Blue Shield of Michigan Value Partnerships program. Dr Vaughn is supported by grant number K08HS026530 from the Agency for Healthcare Research and Quality.