The Consumers Reports Health Ratings Center has been created to provide unbiased analyses and ratings to help consumers make informed health–care decisions. Our objective is to provide up–to–date comparisons of health services, drugs, devices and consumer experiences. These analyses and ratings will help consumers navigate an increasingly expensive and complex health–care system.

While Consumer Reports has been testing consumer products for over 70 years, we are now expanding our efforts to meaningfully evaluate health services so consumers can make their own informed comparisons and decisions. Through the Health Ratings Center, we have developed rigorous procedures of analysis. We hold high standards for those we work with, how we deal with conflicts of interest, how we use data objectively and how we hold ourselves accountable for decisions we make.

Health information includes more than data. It also includes a variety of commercial influences that shape how we think about health and make decisions. We take these commercial factors into account when we evaluate the data and do our best to present a balanced picture to consumers. Our experts are entirely independent of industry, so we can uniquely analyze information while keeping our focus only on you, the consumer.

The Health Ratings Center works with information sources and experts who are willing to adhere to Consumers Union's principles and standards. These include:

• Information that provides good quality evidence, transparent methodologies, and current information.
• Rigorous internal review by our doctors, health researchers and statisticians Continuous evaluation of our ratings efforts through surveys, focus groups and other feedback methods.

We'd like to hear from you. So send any input or ideas to the Director of the Health Ratings Center, John Santa, M.D. M.P.H.

How does Consumer Reports Health rate treatments?

The Treatment Ratings tables in ConsumerReportsHealth.org are based on information provided by our expert partner (BMJ Group) and are exclusive to Consumer Reports. Like Consumers Union, BMJ Group seeks to provide credible, comprehensive information free of industry influence.

Treatments are ranked on a relative scale according to the evidence for the benefits and harms in most people. They are sorted into the following five categories:

Benefits very likely outweigh harms
Benefits likely outweigh harms
Benefits may not outweigh harms
Unlikely to provide benefit or may be harmful
No benefit or likely to be harmful

Two additional categories are unrated. They are:

• Needs further study
• Other treatments

ConsumerReportsHealth.org uses the familiar Consumer Reports "blobs" in our Treatment Ratings, but unlike product ratings we do not give a specific "score" to any of the treatments or therapies. Instead we hope this helps you to "score" treatments with your doctor. The information in our Treatment Ratings tables is designed to help you understand the evidence for different treatment options so that you can work with your doctor to make better health-care decisions. You and your doctor will need to consider your personal medical history and circumstances, preferences, and resources (time, effort and cost) before deciding on the best course of action for you to follow.

Our Review Process

A team of BMJ Group doctors and medical specialists, expert at finding the best information from all over the world, conducts an exhaustive search for studies that measure the effectiveness and harms of each treatment. Published studies are evaluated for quality based on the scope of data and the size and validity of clinical trials. If the specialists can't find good reviews or trials, they will include less rigorous studies and explain their limitations.

Once the research has been collected, medical research specialists use validated tools to weigh the evidence and filter out studies that use poor research methodology. This thorough process often results in a clear explanation of why certain treatments work and which treatments currently work best. Our sources are included on each page of the site. Evidence for the effectiveness of each treatment is then analyzed, summarized and categorized by practicing physicians. Health communication specialists at BMJ Group then translate this evidence for patients and write supporting information that explains each condition. Their goal is to make the information as understandable and easy to use as possible.

How Current are the Treatment Ratings?

We aim to provide detailed, up-to-date information about each of the health conditions and treatments described. Most reports are updated at least once a year and more frequently if new, relevant research findings or scientific developments emerge.

Updates may include:

• New evidence or analysis of previously published evidence that changes what we know about how well a treatment works.
• Treatments that have not been included before.
• New health conditions. These are chosen based on how common they are, and how much the research evidence is needed.

To make sure that we don't miss important new research, we examine the top medical journals weekly, regularly check whether the U.S. Food and Drug Administration (FDA) has authorized new drugs, approved new uses for those drugs, or altered its advice on any medicines, and track whether leading medical groups and government agencies have changed their guidelines on treatments.

Learn about BMJ Publishing Group

The Natural Medicine Ratings on ConsumeReportsHealth.org are provided by the Natural Medicines Comprehensive Database from the Therapeutic Research Center. The Database is supported by the best available scientific evidence. This evidence is analyzed and evaluated using the same high standards the Database editors use to evaluate conventional pharmaceuticals in their other professional publications, Prescriber's Letter and Pharmacist's Letter. When evidence does not exist or is deficient for a particular product (which is often the case), this deficiency is clearly acknowledged with a statement indicating a lack of data.

To gather the scientific data, editors, researchers, and contributors systematically review medical journals from around the world. Hundreds of articles are reviewed and analyzed for reliable data. This does not mean that only flawless studies are analyzed. But it does mean that the Database is based on the best available data at the current time.

This approach is highlighted by the Effectiveness Ratings. Each natural product is assigned an Effectiveness rating based on the quality of the evidence for a given indication.

Learn about Therapeutic Research Center

The Consumer Reports National Testing and Research Center, in Yonkers, N.Y., is the largest nonprofit educational and consumer product testing center in the world. Research and testing are pivotal components of our work.

Before a product even enters one of the dozens of labs at our Yonkers headquarters, it has been subjected to considerable research. We gather data about products and services, about consumer demand in the marketplace, and about what our subscribers plan to purchase. Editorial, technical, and research staff then scrutinize that material, along with suggestions from our subscribers, to develop our testing schedule.

Staff shoppers--assisted by a network of more than 150 anonymous shoppers distributed throughout 65 U.S. cities--buy the products we use as test samples.

To supplement laboratory testing, the survey research department gathers the experiences that hundreds of thousands of our subscribers have had with products and services through an annual questionnaire. Those results are the basis of our well-known auto Frequency-of-Repair index and other product-reliability reports.

More than 125 testing and research experts test and analyze the products we rate. State-of-the-art testing equipment is always used and is sometimes complemented by equipment designed by our engineers. The actual tests are based not only on government and industry standards but also on standards our specialists think should apply. (Tour our labs or pay a virtual visit to our auto-test track.)

Consumer Reports has been testing products since 1936. Visit our online gallery to see a selection of vintage testing photographs from our archives.

Consumer Reports National Research Center

The Consumer Reports National Research Center expands the reach of the world-renowned CR labs into the consumer marketplace. Using scientifically valid research methods, highly trained social scientists survey subscribers about the very practical, consumer-focused health-care topics that matter to them – from what it takes to be a successful dieter, to which eyeglass retailers provide the best service and price.

Like the rest of Consumers Union, the Consumer Reports National Research Center is free of corporate influence and advertising. Its surveys are not commissioned or funded by industry, government, academia, or big media.

With over five million Consumer Reports online and print subscribers available to be polled, our surveys often contain large sample sizes, and detailed analyses. For each health-related survey we conduct, tens of thousands of subscribers typically respond. When appropriate, we also survey nationally representative samples and health-care professionals.

Selection of survey topics is guided by consumer interest and need. We conduct monthly readership surveys of Consumer Reports and Consumer Reports on Health customers, as well as broader market research studies, to determine the types of information consumers will find most useful. Recent survey topics include: hospitals, insurance plans, pharmacies, mental health, and alternative medicine.

What is the basis for this hospital information?
The information is based on the 2008 Dartmouth Atlas of Health Care, a study conducted by John Wennberg, M.D., Elliott Fisher, M.D., and colleagues at the Dartmouth Institute for Health Policy and Clinical Practice in Lebanon, New Hampshire. We have collaborated with the Dartmouth Atlas group to present this data for consumers.

What type of patients were included in the study?
The Atlas study included 4.7 million recipients of traditional Medicare (but not Medicare managed care).The Atlas researchers looked at all medical care received by those patients, whether in or out of the hospital, during the last two years of their lives. The groups studied were those who:

• suffered from of one of nine common, life-threatening chronic diseases in the years 2001 through 2005
• were hospitalized at least once for something other than surgery
• were at least 67 years old when they died

What diseases did the Atlas select for study, and why?
The nine chronic illnesses were selected because about 90 percent of Medicare patients' deaths are associated with these diseases. They are also among the most challenging diseases to manage, and many of the patients studied had more than one of these conditions.

From most to least prevalent, the nine diseases are:

• congestive heart failure
• chronic lung disease
• cancer
• dementia
• coronary artery disease
• chronic kidney failure
• peripheral vascular disease (poor circulation in the legs and feet)
• diabetes with organ damage
• severe chronic liver disease

Why study only patients who died? How does their experience apply to me or my family member?
To fairly compare hospitals' style of treatment, the Dartmouth researchers looked at patients who had the same outcome – they died. But by looking back a full two years before death, the Atlas is able to examine care given not only in the last weeks and months of life, when patients were in the final, most severe stages of their long illnesses, but also during earlier stages of chronic illness when they were not so ill. For that reason, the Dartmouth experts believe they are measuring aggressive vs. conservative care throughout the course of chronic chronic serious illnesses, information clearly useful to patients or family members suffering from those diseases.

How does this information apply to people who don't have one of the nine studied conditions?
Other studies from Dartmouth have shown that hospitals with an aggressive approach to treating patients at the end of their lives predicts the style of care a hospital gives to all patients, regardless of age and prognosis.

What do the percentile numbers mean?
The Dartmouth Atlas ranks 2,857 U.S. hospitals on measures of aggressive or conservative treatment of people with one of the nine common chronic illnesses. For example, a hospital ranked in the 97th percentile on the scale, is more aggressive in its treatment than 97% of the hospitals studied. The percentiles are based on an index derived from two statistics: the number of days the average patient spent in the hospital in the last two years of life, and the number of times they saw a doctor in the hospital during those two years.

What are the differences between aggressive and conservative care?
Conservative care typically means:

• Fewer days spent in the hospital
• Fewer days spent in the intensive care unit
• Fewer doctor visits overall
• More focus on primary care doctors, less focus on specialists

Aggressive care typically means:

• More days in the hospital
• More days in intensive care units
• More doctor visits overall
• Less focus on primary care doctors, more focus on specialists

If only people who died were studied, how do you know that the treatment in more aggressive hospitals didn't lead to longer life?
Other research shows that more aggressive treatment does not necessarily extend life. In a study of Medicare beneficiaries with colon cancer, heart attacks, and hip fractures, the Dartmouth team analyzed all tests and treatments they received for up to five years after their diagnosis. Patients who received the most aggressive treatment were slightly more likely to die and survivors were no better off in their ability to function day to day. Also, patients who were treated most aggressively were no more satisfied with their care and spent more on co-pays. This study was reported in the Annals of Internal Medicine in the issues of February 18, 2003 and October 7, 2004. The study was later repeated looking specifically at teaching hospitals associated with medical schools, with similar results.

How can more treatment lead to the same or worse results?
More frequent hospitalization increases the risk of infections and other mishaps. If care is not carefully coordinated, seeing more specialists increases the risk of medication errors and other types of miscommunication. Sometimes more care isn't harmful, but doesn't help, either, because it duplicates what's already been done (for example, repeating tests) or doesn't affect the patient's health one way or another.

Why can’t I find my hospital?
Some small hospitals had so few deaths during the study period that the Atlas statisticians could not calculate reliable measures. The Atlas also excluded Veterans Administration hospitals and hospitals in other countries where Medicare recipients died. In addition, Medicare patients who are enrolled in an HMO are not included in the analysis, therefore a hospital that has only HMO patients (such as hospitals run by Kaiser) will not be listed. Medicare pays a fixed annual amount per enrollee and the HMOs do not submit individual claims to Medicare. If your hospital is neither small, foreign, or a VA hospital or an HMO, it is possible it's listed under a different "hospital referral region" in your state from the one you originally checked.

What's included in "patient cost for doctor visits"?
This shows the average costs that patients (or their supplemental insurance carriers) paid for doctor visits, tests, outpatient procedures, and durable medical equipment such as oxygen tanks. Individuals may have much higher or much lower costs, depending on their illness and the treatment received.

What if patients at a particular hospital were sicker than average? Wouldn't that affect how often they saw a doctor or went to the hospital?
Yes. For that reason, Atlas researchers statistically adjusted their results to account not only for how sick patients were, but also for differences in age, race, and gender.

What about patient privacy?
The data used to create the Atlas comes from a national Medicare claims database that has long been available to researchers approved by the federal government's Center for Medicare and Medicaid Services. Analyses using this database do not use or include patients' names, addresses, Social Security numbers, or any other information that could be used to identify them individually.

Patients don't always go to the same hospital. How does the Atlas handle that situation?
Atlas researchers assigned patients to the hospital where they received most of their hospitalizations during that two-year period. They already knew, from earlier research, that seriously ill patients are very loyal to their hospitals and doctors. In the few cases where patients received exactly half their care at one hospital and half at another, they were assigned to the hospital where they were last admitted before they died.

Is aggressiveness of care the only measure available for choosing a hospital?
No, it should only be part of the equation. Medicare outcomes data, how hospitals give treatment and preventive care, and patient satisfaction information are available at the government's Hospital Compare Website.