Nurses perform countless life-saving duties throughout the day. This question is just one more. (Credit: PeopleImages on Shutterstock)
In A Nutshell
- A simple 0–10 “how hard is it to breathe” check can flag patients at higher risk of dying in the hospital and after discharge.
- Risk is highest when shortness of breath starts after admission, not just at arrival.
- The higher the breathing score, the higher the risk; pain scores did not track death risk.
- Normal oxygen readings do not rule out danger; if breathing feels harder, speak up.
Twice a day during hospital stays, nurses ask patients a question that takes less than a minute to answer but can reveal who’s at serious risk of dying. The question isn’t about pain levels or how they’re feeling overall. It’s simpler than that: On a scale of zero to 10, how hard is it to breathe right now?
A study published in ERJ Open Research found that patients who report breathing discomfort during their hospital stay face dramatically higher risks of death, both during hospitalization and for up to two years after discharge. Researchers at Beth Israel Deaconess Medical Center in Boston tracked nearly 10,000 hospital patients and discovered that this quick breathing assessment strongly predicted who was more likely to die in the hospital and after discharge.
Patients who developed shortness of breath after being admitted to the hospital had six times the risk of dying during their stay compared to patients who never reported breathing difficulty. Even patients who arrived at the hospital already short of breath had only three times the risk, making the later-developing breathing problems an especially ominous sign.
Breathing Problems Beat Pain as a Predictor
The research team, led by Dr. Jennifer Stevens and Dr. Robert Banzett, didn’t just track breathing discomfort. They also monitored pain levels using the same zero-to-10 scale. Pain turned out to be far more common, with 72% of patients reporting some level of pain during their hospital stay. Only about 27% reported breathing discomfort at any point.
But pain levels showed no meaningful connection to whether patients lived or died. Breathing difficulty, on the other hand, was a powerful predictor of mortality.
The researchers suggest that breathing discomfort integrates information from multiple systems in the body that monitor how well the heart and lungs are exchanging oxygen and carbon dioxide. When someone feels short of breath, it often signals that something fundamental isn’t working right. Pain, while certainly important to treat, more often indicates localized problems with skin, muscle, or bone rather than life-threatening system failures.
Death Rates Rose With Breathing Difficulty
Among the 9,785 patients in the study, conducted between 2014 and 2016, about 73% never reported breathing difficulty during their entire hospital stay. These patients had just a 1% chance of dying in the hospital.
About 18% arrived at the hospital already experiencing breathing discomfort. Their in-hospital death rate jumped to 3.3%.
But the 10% who developed breathing difficulty after admission faced the highest risk: a 5.9% death rate during hospitalization.
The severity of breathing discomfort mattered too. Patients who rated their worst breathing difficulty between 1 and 3 on the 10-point scale had three times the normal death rate. Those who rated it between 4 and 10 were six times more likely to die.
Risk Continued After Discharge
The breathing assessments didn’t just predict what would happen during the hospital stay. Patients who reported any breathing difficulty had 50% higher mortality over the following two years compared to those who never struggled to breathe.
The timing pattern continued to matter. Patients still experiencing breathing difficulty on the day they left the hospital faced more than double the risk of dying within two years, with a death rate of 23.4% compared with 1.4% among patients who had no breathing difficulty on the day of discharge.
Patients discharged while still short of breath also had higher rates of other medical conditions. Their average comorbidity scores were more than double those of patients without breathing issues, and they were much more likely to have respiratory diagnoses like COPD or pneumonia.
Patients who developed new breathing difficulty during their stay were also twice as likely to need transfer to intensive care, twice as likely to require a rapid response team activation, and three times more likely to stay in the hospital seven days or longer. They were also twice as likely to need discharge to a skilled nursing facility or long-term care rather than going home.
The researchers found no relationship between breathing difficulty and oxygen saturation levels measured by the standard fingertip monitors that hospitals use routinely. This means breathing discomfort picks up on problems that don’t show up in those routine measurements.
What Hospitals Can Do Differently
Nurses at Beth Israel Deaconess Medical Center have been documenting these breathing assessments since before the study began, and they report that asking the question adds less than a minute to their routine patient checks. Nurses document dyspnea at least once per shift, typically every 12 hours.
The study tracked patients across 14 different hospital units, including general medicine, cardiology, oncology, surgery, and orthopedics. About 77% of the patients had come through the emergency department before being admitted to a hospital unit.
Nurses asked patients to rate their breathing discomfort while sitting or lying down, not during physical activity. The scale ranged from zero (no discomfort) to 10 (unbearable), with markers at mild, moderate, and severe breathing difficulty.
Patients should speak up if they notice increasing difficulty breathing during a hospital stay, even if oxygen monitors show normal readings. The feeling itself carries important information about how well the body’s systems are working together.
Disclaimer: This article is for general information. It is not medical advice. If you or a loved one has trouble breathing, seek care from a qualified clinician right away.
Paper Summary
Methodology
Researchers conducted a retrospective study examining 9,785 consecutive adult patients admitted to Beth Israel Deaconess Medical Center between March 2014 and September 2016. The study focused on non-ICU patients across 14 different hospital units, including general medicine, cardiology, oncology, surgery, and various surgical specialties. Nurses documented patient-reported breathing discomfort and pain levels at least once per 12-hour shift using zero-to-10 scales. Researchers manually transcribed these ratings from hand-written nursing notes that had been scanned into electronic records, achieving 93% accuracy in transcription. The team categorized patients into groups based on when they first reported breathing difficulty: never, at admission, after admission, or at discharge. Statistical analysis used generalized linear models and Kaplan-Meier survival analysis to test associations between breathing discomfort ratings and various outcomes including in-hospital mortality, two-year mortality, ICU transfers, rapid response activation, length of stay, and discharge destination.
Results
About 73% of patients reported no breathing difficulty throughout their hospitalization. Among the remaining 27%, about 18% reported breathing discomfort at admission and 10% developed it after admission. Only 3.8% still had breathing difficulty at discharge. Patients with no breathing discomfort had 1% in-hospital mortality. Those with breathing difficulty at admission had 3.3% mortality (three-fold increase), while those who developed breathing difficulty after admission had 5.9% mortality (six-fold increase). The relationship was dose-dependent: mild breathing difficulty (ratings 1-3) tripled death risk, while moderate to severe difficulty (ratings 4-10) increased it six-fold. Over two years following discharge, patients with any breathing difficulty during hospitalization had 50% higher mortality than those without. Patients with breathing difficulty at discharge faced 2.6 times the mortality risk in the following two years, with 23.4% dying compared to 1.4% of those discharged without breathing problems. Pain was reported by 72% of patients but showed no association with mortality. Breathing difficulty predicted other poor outcomes including doubled ICU transfer rates, doubled rapid response team activation, tripled likelihood of hospital stays exceeding seven days, and doubled rates of discharge to extended care facilities.
Limitations
The study was conducted at a single tertiary academic medical center, limiting generalizability to smaller hospitals or community settings. Breathing difficulty and pain assessments relied on how nurses asked the questions, which could vary between nurses and shifts. The researchers acknowledge that more standardized electronic collection might strengthen the associations found. Some outcome measures like rapid response team activation and ICU transfer could be confounded because breathing difficulty might influence clinician decisions to summon help. The mortality outcome isn’t subject to this limitation. Breathing difficulty ratings weren’t documented in the emergency department, so the researchers couldn’t assess whether earlier documentation might provide even more predictive information. The hand-written records required manual transcription, which had a 7% error rate despite quality checks. The study excluded intensive care unit patients, so the results apply specifically to general hospital ward patients.
Funding and Disclosures
This research was supported by grant NR010006 from the National Institutes of Health. The authors declared no conflicts of interest. All six authors contributed to study design, data analysis, or manuscript preparation, with all authors reviewing and approving the final version.
Publication Details
Stevens JP, Schwartzstein RM, Sheridan AR, O’Donnell CR, Baker KM, Banzett RB. Patient-reported dyspnoea predicts 6-fold hospital mortality,” was published November 9, 2025 in ERJ Open Research. DOI:10.1183/23120541.00804-2025. Published by the European Respiratory Society under Creative Commons Attribution Non-Commercial Licence 4.0.







