Published online before print August 25, 2008, doi: 10.1161/CIRCULATIONAHA.108.804872
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2008;118:e498-e500.)
© 2008 American Heart Association, Inc.
Cardiology Patient Page |
How Are Our Hospitals Measuring Up?
"Hospital Compare": A Resource for Hospital Quality of Care
From the Section of Cardiovascular Medicine (M.A.J., H.M.K.) and the Robert Wood Johnson Clinical Scholars Program (H.M.K.), Department of Internal Medicine, and Section of Health Policy and Administration (H.M.K.), School of Public Health, Yale University School of Medicine; Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, Conn (H.M.K.); and Department of Health Care Policy, Harvard Medical School and the Department of Biostatistics, Harvard School of Public Health (S.-L.T.N.), Boston, Mass.
Correspondence to Dr H.M. Krumholz, Yale University School of Medicine, 333 Cedar St, PO Box 208088, New Haven, CT 06520–8088. E-mail harlan.krumholz{at}yale.edu
 |
How well are hospitals in the United States taking care of their patients, and how can you learn more about the quality of care at hospitals in your area compared with care provided elsewhere? |
|---|
 Top How well are hospitals... Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
The Centers for Medicare & Medicaid Services, the federal agency that administers the Medicare program, publishes information on the Internet about the quality of hospital care for many conditions, including heart attacks and heart failure. The quality measures for cardiac disease include information about how often eligible patients are treated with certain medications, how often smokers are counseled to quit, how often hospitals provide rapid treatment to restore blood flow in the heart arteries for patients who need that therapy quickly, and how often an indicated test to measure heart function is done for patients with heart failure. All of these measures assess care that is strongly recommended by experts. The measures include only those patients who should receive this care and, ideally, performance on these measures should be very high. The results are provided in terms of percentages, ie, how many people out of a hundred are treated according to what is recommended. For example, aspirin is recommended for all patients admitted with a heart attack, and one measure determines the percent of patients in the selected hospital who were treated with aspirin. With the attention given to these measures, the rates have improved for all of them over the last several years.
You can also find information about how patients fare in the 30 days after being admitted to the hospital for a heart attack and heart failure. This information includes a number called the risk-standardized mortality rate, numbers that are called the interval estimate, and a number that is called hospital volume. In this article, we explain how you can find these numbers, how you can interpret them, and how they can best be used (and how they should not be used).
|
Where Do I Find This Information? |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
The information collected from hospitals is posted at the agency’s publicly available "Hospital Compare" website, http://www.hospitalcompare.hhs.gov.
|
...And What Do I Do When I Get There? |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
The "Hospital Compare" home page features an overview section with definitions and background reading about topics, including how patient outcomes are measured, and a section with information about surveys completed by hospital patients regarding the care they received. By clicking on the "learn more" link below each topic, viewers can access a range of material that will help them make the most of the wealth of information at the site. The home page also features a general information section with a link to a hospital checklist, a set of questions that may be helpful when discussing hospital care with a healthcare provider, and a Medicare document that details the rights of every patient.
To access a comparison of hospitals through the home page, the "find and compare hospitals" tab provides 2 ways to search (the Figure): search by specific conditions or procedures or general search (search for hospital information locally or nationally).
|
Either option will reveal a succession of screens, navigated with the "continue" tab, in which terms can be selected to narrow results.
|
How to Search by Specific Conditions or Procedures |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
Step 1: Click on the "begin search" tab to start your search for information about hospital process-of-care measures, hospital outcome-of-care measures (when available), survey of patients’ hospital experiences, and Medicare payment and volume (the number of patients that each hospital treats for the condition or procedure) for selected medical conditions or surgical procedures.
Step 2: Select one of the following options that will determine how "Hospital Compare" will search for your information: hospital name, city, ZIP code, state/territory, and county. If you choose city or ZIP code, you will be given drop-down fields in which you can indicate how wide a range (in miles) from that city or ZIP code you would like to extend your search. For example, if you choose city, enter "Boston," select as your distance in the drop-down menu 25 miles, and enter "Massachusetts" in the state field, your search will return the names of hospitals within a 25-mile radius of Boston, Mass.
Step 3: Select a medical condition or surgical procedure. Your choices for medical conditions are heart attack, heart failure, chronic lung disease, pneumonia, diabetes in adults, and chest pain. Your choices for surgical procedures are heart and blood vessels; abdominal; neck, back, and extremities (arms and legs); bladder, kidney, and prostate; and female reproductive. This results in the "select hospitals to compare" screen, a list of the hospitals that match the criteria you have specified.
If you click on a maximum of 3 check boxes next to the names of hospitals that you are interested in comparing and then select "compare" at the bottom or top of your screen, the final comparison is presented. By clicking either "view graphs" or "view tables," you can see the information in the format that is clearest for you. By clicking the "reset checkboxes" tab, you can compare a different combination of hospitals.
|
How to Perform a General Search |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
The "begin search" tab in this option initiates the same steps as above, except that the specification of medical condition or surgical procedure is optional.
|
What Is 30-Day Mortality and Why Is It Important? |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
A mortality rate is a death rate. It conveys the percentage of patients (how many out of a hundred) who die from any cause within a certain period of time. The 30-day mortality rate for a hospital for a given condition provides information about the number of patients who die divided by the total number of patients and then multiplied by 100. The rate is essentially the number of people who die for every 100 patients treated. For example, a rate of 10% indicates that for every 100 patients treated, 10 die. This rate would be equivalent to a survival rate of 90%.
The problem with comparing the simple calculation of mortality rates among hospitals is that different hospitals treat different numbers and types of patients. We need advanced statistical techniques to take these differences into account. The result of such calculations produces a number that is called the risk-standardized mortality rate. This rate is important because it conveys how successful hospitals are at helping their patients to survive. Certainly, not all deaths in patients with heart attacks are preventable, even with the very best care. However, there are times when higher quality of care can prevent deaths, and hospitals that provide the best care would be expected to have the lowest mortality rates.
This rate also has 2 numbers associated with it, called the interval estimate. The interval estimate conveys the uncertainty about what the true rate is. This reporting approach contrasts with that for medication and counseling for which no uncertainty is reported. The interval estimate provides information similar to what is reported in political polls about the accuracy of the results. For example, we may hear that a candidate’s approval rate is 46% and the accuracy of the poll is 3%. This implies that the candidate’s true approval rate is anywhere between 43%=46%–3% and 49%=46%+3%, giving an interval estimate of 43% to 49%. How does this work for risk-standardized mortality rates? Suppose Hospital X has a reported mortality rate of 14% and the interval estimate is 11% to 17%. We interpret this to mean that the true risk-standardized mortality rate for Hospital X is anywhere between 11% and 17%, with our best guess at 14%.
Lastly, there is a number for hospital volume, the number of patients whom the hospital sees for the condition being reported. Hospital volume is important because it is difficult to know the quality of hospitals that treat only a small number of patients with a given condition. With little information, our best guess tends to be closer to the average for all hospitals because we know so little about the performance of that particular hospital.
|
How Do I Interpret the Risk-Standardized Mortality Rate? |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
Consider Hospital X again with an estimated risk-standardized mortality rate of 14% and an interval estimate ranging from 11% to 17%. The easiest way to interpret these numbers is to compare the interval estimate of Hospital X with the national average mortality rate. If the national rate is 16%, then Hospital X is providing care as expected given the patients it treats because our best estimate of care at Hospital X is anywhere between 11% and 17%, which includes 16%. If the national rate were 10%, then we would say that Hospital X provided worse-than-expected care because the lower end of the interval estimate, 11%, is higher than the national rate of 10%. Conversely, if the national rate were 18%, then we would conclude that Hospital X provided better-than-expected care because the upper end of the interval estimate, 17%, is entirely below the national rate of 18%.
|
How Is the Information on "Hospital Compare" Best Used? |
|---|
|
TopHow well are hospitals...Where Do I Find......And What Do I...How to Search by...How to Perform a...What Is 30-Day Mortality...How Do I Interpret...How Is the Information... |
|---|
"Hospital Compare" is best explored when you have time to read and evaluate the information; it is not designed to be used as a tool for hospital "shopping" online or for selecting hospitals during an emergency. Rather, "Hospital Compare" should be considered a valuable resource through which patients can share in the latest knowledge that is emerging about the quality of care and outcomes in our hospitals and healthcare systems, knowledge that can be used to engage in informed discussions with physicians and others surrounding this important topic.
|
Acknowledgments |
|---|
Disclosures
Dr Krumholz has contracts with the Colorado Foundation for Medical Care to develop outcomes and surveillance measures for public reporting. Dr Normand is funded by the Massachusetts Department of Public Health to monitor the quality of care after cardiac surgery or percutaneous coronary intervention. M.A. Johnson reports no conflicts.
|
Footnotes |
|---|
The information contained in this Circulation Cardiology Patient Page is not a substitute for medical advice, and the American Heart Association recommends consultation with your doctor or healthcare professional.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||