Where Should My Mother Have Her Knee Replaced?

Where should my mother have her knee replaced?
Elizabeth Mort, MD, MPH, is senior vice president of Quality and Safety at Massachusetts General Hospital and the Massachusetts General Physicians Organization

By Elizabeth Mort, MD, MPH

Elizabeth Mort, MD, MPH, is senior vice president of Quality and Safety at Massachusetts General Hospital and the Massachusetts General Physicians Organization. A practicing general internist, Dr. Mort leads high-stakes quality and safety measurement and improvement work across a broad range of initiatives at Mass General. She also serves as senior medical director at Partners HealthCare, Inc., and is an assistant professor of both Medicine and Health Care Policy at Harvard Medical School.

At least once a week, someone asks me to recommend a doctor or a hospital for themselves or a loved one. These inquiries are commonplace for anyone who works in health care. Finding information about the quality of health care is much more challenging than tracking down similar details about cars or restaurants. A seemingly simple question like “where should my mother have her knee replaced?” can require detective work — even for those on the inside. At Massachusetts General Hospital, scores of people are working on making this quest attainable. We’re making progress, but we have a ways to go.

Despite nearly 200 years of trying, our ability to define, measure and communicate information about healthcare quality is sketchy. In areas like organ transplantation, plenty of information on basic outcomes is available on national websites. For other procedures, like knee replacement, it remains elusive. Why the difficulty?

Experts have not always agreed on how to measure healthcare quality. Take mortality. Everyone agrees mortality is an important outcome, but it isn’t really a great measure of hospital quality. One challenge is that the chance of dying is related to the severity of a patient’s illness and we haven’t been able to make statistical adjustments robust enough to compare hospitals on an apples-to-apples basis. Mortality rates for specific procedures or diseases are more useful. For many low-risk procedures though, mortality rates are vanishingly low, so information about mortality rates isn’t always helpful for patients making important healthcare choices.

Defining Healthcare Quality

Before we look at some more promising measures, let’s define healthcare quality. In 1999, the Institute of Medicine (IOM) defined it as “the extent to which health services provided to individuals and patient populations improve desired health outcomes.” The IOM also described six pillars of quality. Health care should be safe, effective, patient-centered, timely, efficient and equitable. Another helpful way of thinking about quality was introduced by Avedis Donabedian, MD, at the University of Michigan in the 1960s. He suggested we look at structure, process and outcomes. Structure means the setting where health care is delivered. Processes refer to the practices engaged in by the physicians and nurses. The outcomes are, of course, what happens to the patient’s health as a result. The IOM’s six pillars and Donabedian’s framework, provide us with concrete characteristics that can be measured.

So what information would I like to have available for a patient facing a knee replacement? For structural attributes, I’d like to know if the hospital is equipped with state-of-the-art operating rooms and technology. Is the orthopedic surgeon board certified? How long has he or she been in practice? These facts are usually available on state licensing websites. In Massachusetts, you can find basic credential information on the Board of Registration in Medicine’s website. I would like to know what types of knee procedures the surgeon has performed as well as their outcomes.

In areas like organ transplantation, plenty of information on basic outcomes is available on national websites. For other procedures, like knee replacement, it remains elusive. Why the difficulty?

I’d like to know about a hospital’s safety record. Does it report infection rates and compliance with guidelines designed to reduce the risk of adverse events like blood clots? Public websites and some hospitals report basic safety information. Massachusetts has websites that report hospital complication rates and similar measures. I’d also want to know about preoperative wait times.

Patients’ views are of great interest. Did patients surveyed indicate that they received adequate pain relief, compassionate nursing care and helpful explanations about discharge plans and medications? Would they go back to that hospital? This information is easy to find on the Centers for Medicare and Medicaid’s (CMS) Hospital Compare website. Results are reported for the hospital as a whole, and not specifically for joint replacement patients. Equity of care, the sixth IOM pillar, is very important to many patients. Do hospitals provide the same quality of care for patients of different ages, genders and racial and ethnic backgrounds? Information on equity is scarce. Some hospitals, like MGH, have reported some measures by race, but this is not yet widespread.

Surgeon Experience With Knee Replacement

The IOM’s “effectiveness” pillar refers to whether the procedure did what it was designed to do. Outcomes can include clinical outcomes like a stable joint with good range of motion. The quality of life and functioning achieved after a full recovery are also particularly important. Some hospitals keep registries of their results but very few make that information public. Outcomes can also include the undesired results or complications. Bleeding, infection and blood clots are specific concerns in joint replacement. I’d like to be sure all measures to reduce risk of complications are implemented reliably for every patient. Some hospitals report their compliance with guidelines, Centers for Medicare and Medicaid reports some of this information as well.

Another very important consideration in recommending a hospital is the patient’s baseline health. Is she a healthy 50-year-old with no medical conditions other than her knee arthritis? Or is she a fragile 83-year-old with severe heart disease and diabetes? Patients with complicated medical histories should look into whether the hospital has a well-staffed intensive care unit. The Leapfrog Group’s report includes information on hospitals’ ICU capabilities. The quality of nursing care should be considered and hospitals that have achieved recognition as a “Magnet hospital” would instill confidence. Depending if the knee replacement is the patient’s first or a repeat, I’d like to know whether the surgeon has experience with both.

There are a few other attributes worth mentioning. Reputation among peers is important. Studies shown doctors are pretty good judges of clinical quality (no wonder people ask our opinions). Reputation gets at some of the hard to measure characteristics such as advocacy for patients, excellent diagnostic and problem-solving skills and consistently excellent outcomes. I’d also prefer a hospital that invests in continuous quality improvement. Wouldn’t you? It’s possible to know whether hospitals are participating in professional society programs to improve quality or are publishing results of their research or improvement work. Our industry is developing measures of safety culture that are likely to have more visibility in the coming years.

Just about every aspect of quality that I’ve described can be measured. But only a fraction is measured and available for patients or physicians. Finding what’s out there requires a thorough knowledge of the public reporting programs and time and effort to pull out and package the relevant information. You can find rankings and ratings, some which have stood the test of time like U.S. News & World Report, and others that are newcomers like whynotthebest.org, a site published by the Commonwealth Fund. No one tool has all the information patients need and even with some detective work, it’s hard to get all the information patients may want.

In the last decade we have seen steady growth in the amount of healthcare data available to patients. Payers are requiring public reporting of basic quality and safety measures and hospitals are developing individual report cards. But we’re still only scratching the surface. Clinical outcomes, physical functioning and quality of life after common procedures such as knee replacement procedures should be measured and made available to physicians and patients alike. It will take time and money and it will require expertise, sophisticated statistics and careful presentation, but it is possible.

At Mass General, we are in a position to lead by example. Our physicians and nurses participate in important registries and professional society programs that measure care and work collaboratively on improving it. Our experts are populating major committees to influence national policies on quality measurement. We have an award-winning quality and safety website that displays a broad array of metrics. But we are not yet positioned to provide patients with the kind of information I have outlined. We will fix that. Until then, if you need advice on where your mother should have her knee replaced, feel free to give me a call.

Elizabeth A. Mort, MD

Massachusetts General Hospital
Associate chief medical officer, MGH
Vice president quality and safety, MGH/MGPO
Dr. Mort is a primary care physician with clinical interests in women’s heath, prevention, and chronic disease management. Dr. Mort completed her residency in primary care internal medicine at MGH in 1986 followed by two years at the Department of Health Care Policy at Harvard Medical School where she completed a research fellowship in health care policy. She also completed a masters in public health at the University of Michigan in 1982, which focused on health planning and administration and population planning. Dr. Mort currently holds several administrative positions including vice president of quality and safety at the MGH and MGPO, and associate chief medical officer at MGH.