Skip to main content

Advanced Search

Advanced Search

Current Filters

Filter your query

Publication Types



Press Release


Oct 29, 2008

Patients Say Hospitals Need to Do A Lot More to Improve Quality: NEJM Study

First-of-its-kind Analysis Shows Hospitals With More Nurses per Patient Score Higher on Basic Measures of Patient Experience

Embargoed for release:
5:00 P.M., ET, Wednesday,
October 29, 2008

New York, NY, October 29, 2008— How do patients feel about the care they receive in U.S. hospitals? A new Commonwealth Fund—supported study by Boston researchers shows that while patients are generally satisfied with their care, satisfaction levels are not as high as they could be and rise significantly when hospitals have more nurses at the bedside. The study, in the October 30 New England Journal of Medicine, reveals that patients frequently feel hospitals fall short in addressing basic quality issues--controlling pain, communicating about medications, and coordinating discharge planning.

In the first analysis of its kind, Harvard School of Public Health (HSPH) researchers examined patient experience information collected by the federal government through its Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey. They examined patient views on communication with doctors, communication with nurses; nursing services; communication about medications; pain control; room conditions; and discharge information. The HCAHPS analysis is based on data from 2,400 hospitals, and illustrates how patients perceive care at U.S. hospitals.

While two-thirds of patients gave their hospital care a high rating, very few hospitals received the highest rating from 90 percent or more of their patients. The performance of for-profit hospitals was worse than that of private and nonprofit hospitals in all areas, but differences between teaching and non-teaching hospitals were small and insignificant.

What Makes a Difference?
On average, patients rate their experiences as better at hospitals with a higher number of nurses at the bedside. In fact, the ratio of nurses-to-patient days was a critical predictor of how patients rated their experience overall. Hospitals with higher nurse-staffing levels per patient were more likely to get high ratings on pain control, communication about medications, and handling of discharge instructions than hospitals with fewer nurses.

But there are still a lot of areas in need of improvement:

  • Nearly one-third of patients did not give high ratings for how well institutions controlled pain, a significant gap since this has been a target of quality improvement initiatives for years.
  • Despite Medicare's efforts to reduce unnecessary readmissions to hospitals, 21 percent of patients did not rate their discharge instructions highly.
  • Another 42 percent of patients gave hospitals low ratings for communicating about medications, a significant gap since this can lead to medical errors.

"These data represent a fundamental change in the way we do business in healthcare. We now have information that is valid and reliable and reflects how patients perceive the care they receive. Our analysis shows that hospitals can do a lot better at providing patient-centered care," says study lead author Ashish Jha, M.D., who co-authored the study with HSPH colleagues Jie Zheng, Ph.D., and Arnold Epstein, M.D., and E. John Orav, Ph.D., of Brigham and Women's Hospital. "The good news is that publicizing hospital performance is likely to stimulate improvements in the future, since hospitals will now have benchmark data to compare how they are doing," he says.

Satisfaction Linked to Clinical Quality, and Both Depend on Where You Live
The study also found that patient satisfaction with care was highly associated with the quality of clinical care in the hospital. This was true for all four conditions measured: acute myocardial infarction, congestive heart failure, pneumonia, and prevention of surgical complications. Care was consistently better in hospitals that received high ratings across all conditions.

The study also showed that patient care experiences vary substantially between geographic regions. For example, patients in Birmingham, Ala., Knoxville, Tenn., Charlotte, N.C., Milwaukee, Wis., and Indianapolis, were more likely to rate their hospital experience as high. In contrast, patients in Orlando, Fla., Chicago, New York City, Fort Lauderdale, Fla., and East Long Island, N.Y. were more likely to give their hospital a low rating. The authors say this may reflect regional differences in care related to practice style and other factors. But they did find a substantial range of performance across the 40 largest regions. In fact, they found differences of 15 to 25 percentage points between the best and worst regions in performance on different basic care components.

"Transparency and public reporting are critically important, so this kind of information is a significant step forward in our ability to assess quality in hospitals," says Commonwealth Fund Vice President for Quality Improvement and Efficiency Anne-Marie J. Audet, M.D. "It is clear that more work needs to be done to make hospital care more patient-centered. We are not talking about high-tech things here. These are basic measures of good care that all patients should be getting."

Publication Details


Oct 29, 2008