How and Who Create the Health Grades

Medicare Hospital Compare overall hospital rating

The Hospital Compare overall hospital rating summarizes up to 57 quality measures across 7 areas of quality into a single star rating for each hospital. Once reporting thresholds are met, a hospital’s overall hospital rating is calculated using only those measures for which data are available. This may include as few as 9 or as many as 57 measures. The average is about 39 measures. Hospitals report data to the Centers for Medicare & Medicaid Services- Opens in a new window through the Hospital Inpatient Quality Reporting (IQR) Program- Opens in a new windowExternal Link icon and the Hospital Outpatient Quality Reporting (OQR) Program- Opens in a new window External Link icon. Star ratings are not calculated for Veterans Health Administration (VHA) hospitals.

The methodology uses a statistical model known as a latent variable model. Seven different latent variable models are used to calculate scores for 7 groups of measures.

  1. Mortality
  2. Safety of Care
  3. Readmission
  4. Patient Experience
  5. Effectiveness of Care
  6. Timeliness of Care
  7. Efficient Use of Medical Imaging

A hospital summary score is then calculated by taking the weighted average of these group scores. If a hospital is missing a measure category or group, the weights are redistributed amongst the qualifying measure categories or groups.

Finally, the overall hospital rating is calculated using the hospital summary score.

Only hospitals that have at least 3 measures within at least 3 measure groups or categories, including one outcome group (mortality, safety, or readmission), are eligible for an overall hospital rating. Not all hospitals report all measures. Therefore, some hospitals may not be eligible for an overall rating.

The comprehensive methodology report- Opens in a new window External Link icon provides additional detail on the methodology used to calculate the Hospital Compare overall hospital rating.

National distribution of overall hospital ratings

The following table shows the national distribution of the overall rating based on December 2017 results.

Overall Rating Number of Hospitals (N=4,579, %)
5 stars 337 (7.36%)
4 stars 1155 (25.22%)
3 stars 1187 (25.92%)
2 stars 753 (16.44%)
1 star 260 (5.68%)




Leapfrog grades 2,650 hospitals from ‘A’ to ‘F.’ How did yours fare?

The Leapfrog Group released its Fall 2017 Leapfrog Hospital Safety Grades, giving 855 an “A,” but giving more than 1,000 a “C” or below.

The updated ratings for each hospital are available on Leapfrog’s website.


For the latest rankings, Leapfrog assigned “A” to “F” letter grades to general acute-care hospitals in the United States based on 13 process and structural measures and 15 outcome measures that represent medical errors, accidents, injuries, and infections.

Leapfrog for the latest ratings expanded its methodology to include an assessment of a hospital’s use of bar code medication administration, which can reduce the risk of a patient receiving the wrong medication.

How hospitals performed

In the latest report:

  • 855 hospitals earned an “A” rating;
  • 638 hospitals earned a “B” rating;
  • 976 hospitals earned a “C” rating;
  • 164 hospitals earned a “D” rating; and
  • 17 hospitals earned an “F” rating.

According to the Leapfrog Group, 42 hospitals nationwide have received an “A” rating in every update since the rankings began in spring 2012.



Healthgrades is an online database of doctors, dentists, and hospitals that has over 100 million users and has amassed data on more than three million U.S. healthcare providers. When a patient searches Google for your practice, Healthgrades will likely show up on the first page.

On Healthgrades, potential patients can search by doctor name, office name, specialty, condition or procedure. While Healthgrades creates physicians’ profiles based on information from government and commercially available services, physicians can claim and edit these profiles.

Once a profile is claimed, you will have the ability to add photos, age, gender, information about specialties, a physician’s background such as when they graduated and information on scheduling appointments. You can claim your practice’s profile, but the main focus is on individuals.

On Healthgrades, patients can complete patient satisfaction surveys about a practice or physician. The surveys evaluate the office and staff, wait time, experience with the doctor and likelihood of recommending the doctor to family and friends.

Pros: Healthgrades is very user-friendly for patients. It doesn’t ask patients to answer questions. Instead, they simply complete a survey. If you have a large number of physicians in your office, you can contact the support center and submit a spreadsheet with your physicians’ information. The support center will then work to update all of the data for you without a fee. Healthgrades will also send physicians personalized, printed postcards with a URL to their profile that can be given to patients.

Cons: Although physicians can post a general response to patient surveys completed, they cannot reply individually to lowly-graded surveys.




What followed after determining initial eligibility?

Hospitals had to show that they treated a given number of patients who had specifically defined conditions or procedures. The threshold number, which varied by specialty, was based on traditional fee-for-service Medicare inpatients who were discharged during the three years from 2014 to 2016. Because the rankings focus on challenging care, only patients who had particular procedures or conditions at a defined level of severity and complexity were included. The minimum in cardiology & heart surgery, for example, was 1,391 patients, 500 of whom had to be surgical. In orthopedics the minimum was 302 patients, 275 of them surgical.

A hospital that fell short was still evaluated if it was nominated by at least 1 percent of the physicians in a specialty who responded to the 2016, 2017 and 2018 reputational surveys.

A total of 1,897 hospitals met these standards and qualified for further consideration in at least one specialty.

How many hospitals were ranked?

Across all 16 specialties, only 158 U.S. hospitals performed well enough to be nationally ranked in one or more specialties.

What determined whether a hospital would be ranked?

We evaluated each hospital’s performance using a variety of measures. Some data came from the federal Centers for Medicare & Medicaid Services’ Standard Analytical File database. Other information came from the American Hospital Association and from professional organizations. We put the heaviest reliance on outcomes because of the self-evident connection between a patient’s survival chances and the quality and safety of their care. Other data, such as the number of patients and the ratio of nurses and patients, are less obviously related to quality and safety, but ample research supports the connection. The physician survey also played a role, although it accounted for only slightly more than one-fourth of each hospital’s score (more on that below).

In the four reputational specialties, most care is delivered on an outpatient basis. The number of outpatients who die in these specialties is so low that risk-adjusted mortality rates, heavily weighted in the 12 other specialties, mean little. Hospitals in these specialties are therefore ranked solely on reputation.

How were the different factors combined?

Each hospital analyzed in the 12 data-driven rankings received an overall score from 0 to 100 based on four elements: survival, patient safety, care-related factors such as the intensity of nurse staffing and the breadth of patient services, and expert opinion obtained through the physician survey. The hospitals with the 50 highest scores in each specialty were ranked. Scores and data for all eligible hospitals in each specialty are also posted. The four elements and their weightings, in brief:

Survival (37.5 percent). A hospital’s success at keeping patients alive was judged by comparing the number of Medicare inpatients with certain conditions who died within 30 days of admission in 2014, 2015 and 2016 with the number who would be expected to die given the severity of illness. Hospitals were scored from 1 to 10, with 10 indicating the highest survival rate relative to other hospitals and 1 the lowest rate. Widely used research software (3M Health Information Systems Medicare Severity Grouper) took each patient’s condition into account.

Patient safety (5 percent). This score reflects a hospital’s success in preventing four harmful types of medical errors or oversights, such as injuries caused during surgery. Two of these types of “patient safety indicators,” defined by the federal Agency for Healthcare Research and Quality, were dropped this year because of their rarity. Hospital performance in the four individual PSIs is displayed online from 1 to 3 on a three-point scale, and the overall Patient Safety Score is displayed from 1 to 9.

Other care-related indicators (30 percent). These include nurse staffing, patient volume, certain clinically proven technologies and other measures related to quality of care. The 2016 American Hospital Association Annual Survey was the primary source.

Expert opinion (27.5 percent). Each year, board-certified physicians in the 16 Best Hospitals specialties are invited to list up to five hospitals, ignoring location and cost, that they consider to be the best in their area of expertise for complex or difficult cases. The figures displayed in the rankings represent the average percentages of responding specialists in 2016, 2017 and 2018 who named the hospitals. More than 21,000 physicians responded in 2018. In one specialty, cardiology & heart surgery, expert opinion receives a weight of 24.5 percent, with 3 percent weight determined by the extent to which a hospital is publicly transparent about certain heart-care-related performance metrics.

Doesn’t reputation still determine which hospitals are ranked?

No. Many ranked hospitals have very low, even zero, reputational scores but are strong clinical performers. In the 2018-19 orthopedics rankings, to take this specialty as one of many examples, 28 ranked hospitals had reputational scores below 2 percent.

How are hospitals ranked in the four specialties without objective data?

In the four reputationally determined specialties, ranked hospitals had to be cited by an average of at least 5 percent of the physicians who responded to the latest three years of U.S. News surveys of specialists. That resulted in lists of 11 hospitals in psychiatry, 12 in ophthalmology and 13 each in rehabilitation and rheumatology.

Sign up for our Newsletter

Get the latest news about the health care industry
straight to your inbox!