Choosing the right hospital can be overwhelming. Patients do not always go to the closest facility, as they may need one with specialized services to treat their condition. For planned care, researching hospitals can help patients make a more informed choice.
In 2005, the Centers for Medicare & Medicaid Services (CMS) launched Hospital Compare, a public website created through the efforts of Medicare and the Hospital Quality Alliance to make hospital-quality information more accessible to consumers. The reporting system, which started as a small set of measures, has since expanded to include data on patient experience, mortality, readmissions, and healthcare-associated infections (HAIs) under its current name, Care Compare. All hospitals included in the data are Medicare-certified; however, they do not treat only Medicare patients.
As a medical malpractice firm with a doctor on staff, the attorneys at Coates & Johnson truly understand how important medical decisions are, and how distressing it is when hospital treatment doesn’t go as planned. That’s why we analyzed the CMS hospital data: to make hospital metrics easier to understand for patients and their families.
We partnered with 1Point21 Interactive to examine these hospital quality-of-care measures across the country. The analysis draws on two companion datasets: national benchmarks for 20 quality-of-care measures, and state-level data showing the number of hospitals that performed below, met, or exceeded those benchmarks.
The following article translates this data into simple visualizations to help consumers make informed health care decisions.
Key Takeaways
- Most hospitals meet national benchmarks. When looking at hospital quality of care measures, most hospitals were rated about the same as national benchmarks, with shares ranging between 85.5% for pneumonia mortality to 99.7% for in-hospital fall-associated fractures.
- Mississippi stands out as the weakest state for death rates. Mississippi was the only state with a negative net-performance balance across all mortality measures analyzed.
- Massachusetts stands as the strongest state for death rates. Massachusetts had a positive balance across all major condition-specific and hospital-wide mortality measures.
- Pneumonia had the widest gap between strong and weak hospitals. It had the highest share of hospitals with worse-than-national death rates (5.5%, or 209 hospitals) among 30-day mortality measures, and one of the highest shares doing better than average (8.9%).
- Pneumonia also had the highest national death rate. An estimated 15.2% of pneumonia patients died (of any cause) within 30 days of admission. Heart attack and stroke tied for second with a mortality rate of 11.9%.
- Pressure ulcers had the highest share of underperforming hospitals across all quality-of-care measures. 181 hospitals (5.7%, or about 1 in 18) were rated worse than the national rate.
National Hospital Complication Rates
The following section covers 20 hospital safety and mortality measures. They focus on serious events patients experience, rather than solely on whether a hospital followed a process of care. The number of hospitals considered per measure varied from 890 to 4,084 hospitals. Measurement periods vary from 24 to 39 months, and the measures are generally refreshed annually.
The measure groups found in the table and charts below are included and defined by this article to help guide readers; these categorizations do not exist in the official reporting system.
- 30-Day mortality measures: Estimates deaths within 30 days of admission for selected conditions, procedures, or hospitalizations overall.
- Deaths after serious surgical complications measure: Tracks in-hospital deaths among eligible surgical patients who developed specified serious, treatable complications.
- Surgical complication measures: Tracks serious complications occurring during or after surgery.
- Inpatient patient-safety measures: Tracks selected safety events identified during a hospital stay.
- Composite patient-safety measure: Combines multiple patient-safety indicators into one overall measure.
30-Day mortality measures
CMS uses deaths within 30 days rather than deaths during hospitalization because hospital stays vary in length. This provides a more consistent time period for measuring outcomes. MORT_30_AMI through MORT_30_STK measure death rates as a risk-standardized percentage of eligible patients admitted for the named condition or procedure who died from any cause within 30 days of admission.
For example, for MORT_30_AMI, 11.9% of heart attack patients died within 30 days of admission. The Hybrid_HWM measure is different in that it reviews deaths within 30 days of admission across all patients, procedures, and conditions; however, it is also expressed as a risk-standardized percentage.
| Measure ID | Measure Name | Definition | National Rate |
| MORT_30_AMI | Death rate for heart attack patients | Deaths within 30 days of admission for a heart attack. | 11.9 |
| MORT_30_CABG | Death rate for CABG surgery patients | Deaths within 30 days of coronary bypass surgery. | 2.4 |
| MORT_30_COPD | Death rate for COPD patients | Deaths within 30 days of admission for chronic obstructive pulmonary disease. | 8.6 |
| MORT_30_HF | Death rate for heart failure patients | Deaths within 30 days of admission for heart failure. | 11.1 |
| MORT_30_PN | Death rate for pneumonia patients | Deaths within 30 days of admission for pneumonia. | 15.2 |
| MORT_30_STK | Death rate for stroke patients | Deaths within 30 days of admission for stroke. | 11.9 |
| Hybrid_HWM | Hybrid Hospital-Wide All-Cause Risk Standardized Mortality Rate | Deaths within 30 days of admission across a hospital’s patients, conditions, and procedures. | 3.9 |
Deaths after serious surgical complications measure
PSI_04 begins with the narrow group of surgical patients who developed serious but potentially treatable complications. It measures how hospitals recognized and responded to complications, rather than a broad measure of death risk.
In other words, it measures “failure to rescue”. The rate for PSI_04 is measured as the number of deaths per 1,000 eligible surgical discharges with qualifying complications for patients ages 18 years and older.
| Measure ID | Measure Name | Definition | National Rate |
| PSI_04 | Death rate among surgical inpatients with serious treatable complications | In-hospital deaths among surgical patients who developed serious but treatable complications. | 173.3 |
Surgical complication measures
CMS Patient Safety Indicators (PSIs) measure the quality of care for hospitalized adults and focus on complications and adverse events occurring after or during surgery that may be preventable. While COMP_HIP_KNEE is not considered a PSI, it is combined in this category since it is a surgical complication measure. It is measured as the risk-standardized percentage of eligible elective hip or knee replacement patients with at least one qualifying complication during the measure’s applicable follow-up window.
The remaining PSIs listed in the table below are measured as the number of events per 1,000 eligible surgical discharges for patients 18 years and older. The national benchmark for PSI-09, postoperative hemorrhage or hematoma, was 2.34 events per 1,000 eligible surgical discharges among patients age 18 and older.
| Measure ID | Measure Name | Definition | National Rate |
| COMP_HIP_KNEE | Rate of complications for hip/knee replacement patients | Patients with one or more serious complications after a planned hip or knee replacement. | 4.1 |
| PSI_09 | Postoperative hemorrhage or hematoma rate | Serious bleeding or pooled blood after surgery. | 2.34 |
| PSI_10 | Postoperative acute kidney injury requiring dialysis rate | Acute kidney failure after surgery that requires dialysis. | 1.67 |
| PSI_11 | Postoperative respiratory failure rate | Serious breathing failure after surgery. | 9.42 |
| PSI_12 | Perioperative pulmonary embolism or deep vein thrombosis rate | Blood clots in the lungs or deep veins occurring around surgery. | 3.52 |
| PSI_13 | Postoperative sepsis rate | A life-threatening response to infection after surgery. | 5.27 |
| PSI_14 | Postoperative wound dehiscence rate | A surgical incision that separates or reopens after surgery. | 1.77 |
| PSI_15 | Abdominopelvic accidental puncture or laceration rate | An unintended puncture or cut to an abdominal or pelvic organ during surgery. | 1.06 |
Inpatient patient-safety measures
The inpatient patient-safety measures cover harmful events that can occur during a hospitalization and are not limited to the surgical procedure itself. The PSIs listed below are measured as the number of events per 1,000 eligible surgical discharges for patients ages 18 years and older.
For example, the national benchmark for PSI-03, the pressure ulcer rate, was 0.63 events per 1,000 eligible medical and surgical discharges among patients age 18 and older.
| Measure ID | Measure Name | Definition | National Rate |
| PSI_03 | Pressure ulcer rate | Serious pressure ulcers, also called bed sores, identified during a hospital stay. | 0.63 |
| PSI_06 | Iatrogenic pneumothorax rate | A collapsed lung caused by medical treatment or a procedure. | 0.21 |
| PSI_08 | In-hospital fall-associated fracture rate | A fall during a hospital stay that results in a fracture. | 0.27 |
Composite Patient-Safety Measure
PSI-90 is a weighted composite measure built from the following 10 patient-safety indicators:
- PSI_03 (pressure ulcer rate)
- PSI_06 (iatrogenic pneumothorax rate)
- PSI_08 (in-hospital fall-associated fracture rate)
- PSI_09 (postoperative hemorrhage or hematoma rate)
- PSI_10 (postoperative acute kidney injury requiring dialysis rate)
- PSI_11 (postoperative respiratory failure rate)
- PSI_12 (perioperative pulmonary embolism or deep vein thrombosis rate)
- PSI_13 (postoperative sepsis rate)
- PSI_14 (postoperative wound dehiscence rate)
- PSI_15 (abdominopelvic accidental puncture or laceration rate)
CMS combines the indicators using weights that reflect the frequency and harm associated with each event. The national rate of 1.00 is used as a reference value. For example, the chart below demonstrates how 92.51% of hospitals met the national rate, while 5.45% of reportable hospitals were rated worse than the national rate, and 2.04% were rated better than the national rate.
| Measure ID | Measure Name | Definition | National Rate |
| PSI_90 | Composite patient-safety measure | An overall score combining several measures of hospital safety events and adverse outcomes. | 1 |
How Many Hospitals Are Worse Than Average?
The interactive visualization below shows the share of hospitals across the nation rated better than, the same as, or worse than the national benchmark for each measure listed in the tables above. Each measure had hospitals with too few cases to compare their rate to the national average. Those were excluded from the overall share.
Most hospitals were rated about the same as the national benchmark on every measure. The “about the same” share ranged from 85.5% for pneumonia mortality (MORT_30_PN) to 99.7% for in-hospital fall-associated fractures (PSI_08).
Pressure-ulcer rates (PSI_03) had the largest share of hospitals rated worse than the national benchmark: 5.69%, or 181 hospitals. Another 1.54% were rated better, while 92.77% were rated about the same.
The death rate for pneumonia (MORT_30_PN) also had a relatively high share of hospitals rated worse than the national benchmark: 5.55%, or 209 hospitals. Nearly 9% of hospitals were rated better than the national average, and 85.52% were rated about the same.
Geographic Breakdown
The interactive map below presents the difference between the share of reportable hospital results rated better than the national rate and the share rated worse, defined as “net performance (percentage points).”
Positive values, shown in lighter shades, indicate more better-than-benchmark results than worse-than-benchmark results. Negative values, shown in darker shades, indicate more worse-than-benchmark results than better-than-benchmark results. The dropdown menu allows readers to view these results for each measure listed in the tables above.
Among all states, Massachusetts stands out for a recurring positive pattern across mortality outcomes: heart attack (+4.65 points), CABG (+7.69), COPD (+9.43), heart failure (+29.09), pneumonia (+35.71), stroke (+4.00), and hospital-wide mortality (+26.32). However, this trend did not necessarily carry over to other safety measures. California also had favorable results across several mortality measures.
In contrast, Mississippi was the only state with a negative net performance balance across every condition-specific and hospital-wide mortality measure shown: heart attack (-4.35), CABG (-7.14), COPD (-6.98), heart failure (-9.26), pneumonia (-24.24), stroke (-3.23), hospital-wide mortality (-15.49), and PSI_04 (-5.56). It also had negative results on several safety measures, including perioperative blood clots (-9.52), pressure ulcers (-5.17), and PSI-90 (-9.30).
Heart attack
The heart-attack measure, MORT_30_AMI, shows fewer dramatic state differences than several other mortality measures. The most pronounced favorable balances appeared in the District of Columbia (+20.00 points), South Dakota (+16.67), and Utah (+8.33), while Nebraska (-5.56), Tennessee (-4.65), and Mississippi (-4.35) had more reportable hospitals rated worse than the national benchmark than rated better than the national benchmark. Because several of these jurisdictions have relatively few reporting hospitals, the percentage-point differences should be interpreted cautiously.
Stroke
Stroke mortality (MORT_30_STK) showed the highest positive balance in New York (+12.50), which was especially notable since the state was positive across all seven of the primary condition-specific and hospital-wide mortality measures. South Dakota (+11.11), Massachusetts (+4.00), Ohio (+3.33), Florida (+4.20), and Minnesota (+3.03) also had a positive net performance.
The largest negative stroke balances were in Arkansas (-12.00), West Virginia (-11.11), Hawaii (-10.00), Montana (-9.09), and New Mexico (-9.09). Arkansas also showed negative values on heart failure, pneumonia, or hospital-wide mortality, which makes the broader pattern more noteworthy than an isolated single-measure result.
Pneumonia
Pneumonia, MORT_30_PN, produces some of the clearest differences. Massachusetts (+35.71), the District of Columbia (+28.57), Connecticut (+23.08), California (+16.30), Delaware (+14.29), New York (+13.73), and Florida (+12.78) had the highest net performance balances.
The strongest negative pneumonia patterns were in Mississippi (-24.24), Alabama (-14.29), Arkansas (-12.70), and West Virginia (-12.50). Mississippi had the largest negative balance, and it was also negative across the other major mortality measures, as mentioned earlier.
Composite Patient-Safety Measure
The composite patient-safety measure, PSI_90, combines several measures into one weighted score and gives prospective patients a summarized indicator of a state or hospital’s performance. By state, net performance balances were largely neutral to negative; 30 states had negative net performance, six had positive values, and the remaining states were at zero or had unavailable results. South Dakota (+6.25) had the highest positive balance, followed by Nebraska (+4.00), and Connecticut (+3.85).
The most negative PSI-90 result was in Alaska (-28.57), followed by Vermont (-28.57), New Mexico (-18.52), Montana (-18.18), Iowa (-16.13), New York (-11.94), Maryland (-11.63), West Virginia (-11.54), Washington (-10.42), Arizona (-10.17), and Georgia (-10.11).
However, states with smaller populations, like Alaska and Vermont may have fewer reportable hospitals, so their large percentage-point differences should be treated cautiously.
Researching Hospital-Level Data
While these national and state figures provide context about geographic patterns in hospital performance, readers who are choosing a hospital may want to review results for a specific facility. To research hospitals in their area, or a hospital recommended by a physician, visit Compare Care.
The site allows users to search for hospitals by name, location, or medical service and compare facility-level quality information, including CMS measures for mortality, complications, and patient safety. Readers should compare hospitals using the measures most relevant to their care and consider practical factors such as the hospital’s specialty services, insurance coverage, distance, and guidance from their health care provider.
Hospital comparisons should be treated as one source of information rather than a definitive assessment of care. CMS measure results are risk-adjusted and based on eligible patients treated during specific reporting periods, and some hospitals may not have enough cases to receive a result for every measure.