A score built from ordinary bedside vital signs may help emergency doctors solve one of the hardest timing problems in medicine: deciding which patients with a suspected infection need antibiotics within the hour and which can safely be evaluated further first.
Researchers analyzed electronic health records from nearly 100,000 adults treated for suspected infection at nine US emergency departments between 2015 and 2024. Among patients with a National Early Warning Score (NEWS2) of 7 or higher, each additional hour before antibiotics was associated with increased mortality. Among patients with lower scores, earlier antibiotics were not associated with better survival.
The study was led by the Center for Sepsis Epidemiology and Prevention Studies at the Harvard Pilgrim Health Care Institute and published in The Lancet Respiratory Medicine on August 5.
The Trade Off Runs in Both Directions
This is not a story about doctors being slow. It is about a genuine clinical bind that has no cost-free answer.
Sepsis is the body's dysregulated response to infection, in which the immune reaction itself begins damaging organs. It kills quickly, which is why guidelines have long emphasized speed.
But sepsis is also hard to identify early, because its symptoms overlap with many non-infectious emergencies and there is no gold-standard diagnostic test. Many patients treated for possible sepsis turn out to have something else entirely. Giving every such patient immediate broad-spectrum antibiotics exposes people to drug side effects, secondary infections, and contributes to antibiotic resistance that makes future infections harder to treat for everyone.
International guidance has been shifting from immediate antibiotics for everyone with suspected sepsis toward prioritizing patients in septic shock, where delay clearly matters most. The open question has been whether septic shock alone captures everyone who benefits from urgency.
The study's answer is that it does not. Many patients with NEWS2 scores of 7 or higher did not have septic shock, yet still showed an association between antibiotic delays and worse outcomes. To the researchers' knowledge, this is the first large study to directly compare the two competing international strategies: the United Kingdom's National Institute for Health and Care Excellence recommendation to use NEWS2 to guide urgency, and the Surviving Sepsis Campaign's emphasis on septic shock.
NEWS2 Is Built from Things Already Being Measured
Part of what makes this finding practical is that the score requires no new test, no laboratory turnaround and no additional cost.
NEWS2 assigns points based on routinely collected bedside parameters including respiratory rate, oxygen saturation, whether supplemental oxygen is being given, systolic blood pressure, pulse rate, level of consciousness and temperature. Higher totals indicate greater physiological derangement and higher risk of deterioration.
The score has been adopted by many health systems internationally, and the NICE recommendation to use it for antibiotic urgency already existed. What had been missing, the researchers note, was empirical evidence supporting that approach.
Lead author Chanu Rhee, a Harvard Medical School associate professor at the Harvard Pilgrim Health Care Institute, said the score can help identify the patients most likely to benefit from immediate treatment, including many beyond those with septic shock, while giving clinicians greater confidence to spend additional time evaluating lower-risk patients.
Senior author Michael Klompas, professor of population medicine at Harvard Medical School, said the findings suggest NEWS2 offers an objective and practical way to identify those patients using a bedside score already in routine use. Klompas is also a coauthor on a recent Lancet seminar on sepsis reviewing the state of the field.
The Study Cannot Prove Cause and Effect
The researchers are explicit about the limitation, and it should not be buried.
This was a retrospective, observational analysis of records, not a randomized trial. It cannot prove that earlier antibiotics directly caused the survival differences observed. Patients who receive antibiotics faster may differ systematically from those who do not, in ways that statistical adjustment cannot fully capture. Sicker-looking patients get treated faster, which cuts both ways in the analysis.
The authors note that additional studies in other health care settings will be important to confirm the findings. Nothing here changes what a patient should do. Current guidelines remain in force, and the score is a decision aid rather than a rule.
What Patients and Families Should Take from It
Sepsis remains the thing to recognize early, and that responsibility falls partly on households.
Warning signs that warrant emergency evaluation include a fever or a very low temperature combined with confusion or disorientation, extreme pain or discomfort, clammy or mottled skin, shortness of breath, a very high heart rate, and a feeling that something is seriously wrong. Sepsis frequently follows a known infection such as pneumonia, a urinary tract infection, a skin infection, or a wound.
If you or a relative develops these signs, say the word sepsis at triage and mention any recent infection, surgery or catheter. Clinicians take that history seriously, and it speeds recognition.
The people at highest risk are adults over 65, infants, people with weakened immune systems, those with chronic conditions including diabetes, lung disease, kidney disease and cancer, and anyone who has recently had a serious infection or hospitalization. Recovery does not end at discharge, as MedicalDaily has reported in covering readmission risk among sepsis survivors.
Do not request antibiotics for viral illnesses. The resistance problem this study is partly about is driven by exactly that kind of use, and it is the reason clinicians are trying to be more selective rather than less.
MedicalDaily will report confirmatory studies and any changes to sepsis guidance as they appear.
Frequently Asked Questions
What is NEWS2? A score based on routinely measured vital signs including respiratory rate, oxygen saturation, blood pressure, pulse, consciousness level and temperature.
What did the study find? Among patients scoring 7 or higher, each hour of delay in antibiotics was associated with higher mortality. At lower scores, earlier antibiotics showed no association with survival.
How large was it? Nearly 100,000 adults treated for suspected infection at nine US emergency departments between 2015 and 2024.
Does it prove delay causes death? No. It is retrospective and observational and cannot establish cause and effect. The authors say confirmation in other settings is needed.
Why not just give everyone antibiotics immediately? Many patients treated for suspected sepsis do not have a bacterial infection. Unnecessary antibiotics carry side effects and drive resistance.
Does this change treatment now? No. Current guidelines remain in place. The score is a decision aid, not a rule.
What are the warning signs of sepsis? Confusion, extreme discomfort, clammy or mottled skin, shortness of breath, rapid heart rate, and fever or abnormally low temperature.
