By Cheree Cleghorn, Editor

The Mayo Clinic may have just published one of the most important studies of 2007—- one that affects every single patient in America and most doctors.

At the end of this story, you will find tips on what you can do about this from us.

The study offers a picture of how fast intricate computer-generated biostatistics have overtaken even medical students, residents and faculty.

“….clinicians struggling to explain new studies to patients are on their own and often make big mistakes.” These are the words of Abigail Zuger, M.D., in Comment for Journal Watch, August 29, 2007. She also says, “It seems only the rare physician is truly fluent in statistics and in the language of medical research.”

Evidence-based medicine—a term used by experts to refer to proven treatments tested in rigorously designed studies—has been all the talk since the 1990s. Advocates consider the use of evidence-based medicine an important part of the way quality must be measured, in groups and for individual clinical performance.

This the future.

Right now, we are stuck with everyone talking about it and very, very few being able to do assess the evidence, if, in fact, the Mayo research findings also apply to all doctors in America.

You will see a huge gap between what doctors think they should be doing with biostatistics and how prepared or confident they are when it comes to their actual work.

The researchers studied 301 Mayo medical students, internal medical residents, and internal faculty members who completed a questionnaire. Sixty-four percent (301 of the 468 eligible participants) responded.

The answers give a picture of how fast intricate computer-generated biostatistics have overtaken even medical students, residents and faculty. They also show how this discipline—-which combines statistics and biology—-hasn’t gotten its place in the medical school curriculum yet, either. Every hour already is packed.

If this kind of analysis is hard for Mayo doctors, pity our local doctors who must take care of us anyway. They never were trained in use of these kinds of statistics to get good at it. Even academics doctors use statisticians for their studies. Our doctors don’t have one they can call.

There are no bad guys here. There are only scientific and technological advances which are coming so quickly the doctor at the community and regional levels have to struggle to try to catch up as soon as they can. This is why physicians have to sit for relicensing examinations, to show that they are current — as defined by licensing boards — with what’s most important in their fields.

From the Medline Free Abstract:

Ed. Note: Findings are reformatted and re-ordered for your ease in reading. No change in wording. The % sign is used although the authors spelled it out.

- “A total of 92.7% of respondents believed biostatistics is an important part of evidence based medicine (EBM).”

- “A total of 88% believed evidence-based medicine is important to clinical practice.

- “However, biostatistics was not evaluated as being as important as many other study within medicine.” (Given a choice among important subjects, where would respondents place biostatistics in relationship to any other subject? This study didn’t ask that. Why? Because that would presume respondents would say biostatistics was important. Their goal was to find out what doctors responding thought about its importance first.)
- “A total of 87.3% of respondents believed it would benefit their career to better understand biostatistics.”

- “Only 17.6% believed their training in biostatistics was adequate.”

- “A total of 23.3% of respondents agreed they could identify when correct statistical methods had been applied in a study.”

- “Twenty-eight percent agreed they could design their own research projects with confidence.”

- “14.6% agreed they could conduct their own statistical analyses with confidence.”

Private and government health plans, the prestigious Institute of Medicine, employers and academic researchers want doctors to do what’s been proven to work. To restate: This is what’s called “evidence-based medicine.”

How can doctors do this if they aren’t sure about how to read the data?

How can any doctor provide evidence-based medical care if interpreting those statistics is something the physician can’t do? Or, at least, not well?

Let’s also note that this is far from a black and white issue. Anyone who follows medical news knows that there is a lot of back-and-forth, with one study contradicting the last one or adding a new question or angle. Example: For a while, estrogen was thought to offer protection against heart attacks for women. That now has been disproved but for years, women took the medication to help prevent heart disease. There was a lot of what anyone would call evidence in favor of hormone therapy as protection against this killer disease. Then, there were other studies which up-ended the original ones. That is how science works.

Even those who are fluent in the language of biostatistics are the first to say one study is one study. Unless there is a study about which a specialty agrees — a new disease model — medical evidence adds up one paper at a time. Doctors are not quick to grab the latest and go for this very reason.

Many new studies are controversial even among experts who know the numbers well. Interpretation disagreements often occur between well-known researchers whose credentials are comparable. There are nuances tucked behind these numbers that only these calculators can appreciate. There are glass-half-full and glass-half empty debates about one number which you read or hear about in the news. And these are people who know what they’re doing with the stats.

Comparing what your doctor suggests to what is called the “standard of care” (evidence-based medicine is the foundation for standards of care for treatment) recommended for your problem is the best way to know find out if you are getting care that is current. You can do that by going to trusted medical information resources and comparing their information about your diagnosis. Some good ones are listed below.

Each one explains a topic a little differently so it helps to look at them all.

“What Are The Odds This Will Work For Me?” or What Can You Do Today?

First, ask your doctor about his or her comfort level with studies on your problem. No physician can keep up on all problems. It should not make you change your opinion if the doctor is honest and says, “I’m not.” In fact, a doctor’s honesty should raise your opinion, not lower it.

Second, ask your doctor how you can work together to figure out what a study means to you, taking into account your personal medical history. These studies usually consider a single condition or the inter-relationship among risk factors. An example would be a patient who has several chronic conditions and the medicines for each don’t work right or, in fact, cause a new problem.

You may have other health care needs which require adjusting the state-of-the-art recommendation. In those cases, caregivers have to choose among what is most important to your health and life. The other conditions will be managed around your main problem.

Third, one take-away from reading the Mayo study is how important getting a serious second opinion is to you if the outcome of the diagnosis and/or care will affect your life significantly.

By serious, we mean those that come from doctors whose expertise is sufficient to offer you another, meaningful point of view.

You and your doctor can work on this together. If your doctor shows discomfort with the idea of your consulting others, do it on your own. That is not easy, but it is achievable and well worth the effort.

Currently, volume is a placeholder for quality and experience in most settings. Quality is harder to measure that makes any sense to laypeople. Experience you can find out about. When a doctor has a high volume of patients with your problem, that serves as an informal indicator of both quality and experience.

One exception to the volume rule-of-thumb: A doctor or hospital which treats a large number of patients because there are no other providers around. Otherwise, doctors who care for the highest numbers of patients with your same diagnosis likely are providing recommended care.

Patient and Family Resources

• MedlinePlus (Free)
• Mayo Clinic (Free)
• PubMed (Also known as Medline, which is for scientists. This is where you find original studies. Abstracts usually are free. For full articles, there may be a small fee.)

Depending on what concerns you, there are some excellent special topic newsletters which are up-to-the-minute. Johns Hopkins Health After 50 covers a wide range of topics, often with illustrations. Harvard publishes the Harvard Heart Letter.

You likely can find these at your local library if you don’t already subscribe. The Johns Hopkins website’s publications section describes what’s available.

Johns Hopkins Health After 50. The White Papers Series also offers current information, in-depth, about a single topic.

Consult the websites of associations or societies dedicated to research and information about your issue. The American Cancer Association, American Heart Association and American Diabetes Association — the top three killer diseases in the U.S. — are examples of patient advocate groups which have signficant resources to offer you. Most diseases have advocate groups such as these. They, too, are concerned with what is state-of-the-art treatment.

As it is clear the knowledge is challenging to understand for nearly all of us, that means checking what the expert’s experts think is a good way to go. Read widely. Ask questions. Make notes. Work with your doctor.

With your doctor’s help, you will arrive at a conclusion about what likely is best for you.

This is not contradicting the findings of the Mayo study. You need to agree on a plan with the person who knows you best or who is caring for you in a serious situation, however you arrive at your answers.

Sources

Primary: PubMed (in process): Publication ID is 17673062 By West CP and Ficalora RD. Clinician attitudes toward biostatistics. Mayo Clin Proc 2007 Aug; 82:939-43.

Secondary: Journal Watch