Of course you want the best car or job or school or doctor, but how do you know which is best? Word-of-mouth? Name or logo recognition? Most expensive? An objective measure is clearly better: resale value of the car; how much the job pays; test scores at the school; and for a doctor...what?
To determine the best doctor, you might look someone up: online; at the Better Business Bureau; in magazines touting the Best Doctors in Chicago or Seattle or New York; or in books such as Who's Who in Medicine or the Best Doctors in America.
Let me assure you, these have nothing to do with quality. I am listed in most and never, absolutely never, has anyone ever asked for my results with patients. You get on these "Best Doctors" lists by having an impressive title, being at a prestigious institution, or writing lots of papers. These provide a Q-factor but does notoriety equal quality?
Increasingly, you can get quantitative data but beware! There are four BIG problems interpreting the available data on physician performance and related patient outcomes.
(1) The only medical data routinely tracked are outcomes patients do NOT want: deaths, complications, adverse impacts and lawsuits. They do not collect information on the outcomes we DO want like restored function; productivity; longevity; and good health.
(2) The medical data is very user-unfriendly. There is too much of it. The organization is confusing. The labels are incomprehensible: blepharoptosis, laparascopic cholecystectomy, gastrorraphy, and Fontan operation.
(3) The public believes that doctors who get sued are like people who are arrested: probably guilty of something. A lawsuit is thereby used as an indicator of a Bad Apple. However, there is abundant scientific proof that there is NO relationship between lawsuits and quality of medicine practiced.
(4) Most important, medical data are not risk-adjusted. Comparing some one who operates on morbidly obese patients (over 400 pounds) with someone who operates on low-risk patients is unfair. Same idea applies if comparing doctors who work with a drug-addicted east LA population with those who practice in Beverly Hills (well okay, movie people may be into drugs too but they are more likely to do what their high-priced doctors recommend).
So how do you determine the best doctor? The answer is to demand information that is readily comprehensible to you; that compares apples to apples (not apples to hand grenades); is long-term; and most important, shows outcomes you want not the ones you want to avoid. If your orthopedic surgeon recommends a knee replacement, you want to know success rates - restoration of pain-free mobility; for how long; in how many; how long in hospital; compared to other facilities; etc. - not just hearing the doctor say that 99% survived and no one has ever complained.
In the absence of good hard useful data, I recommend asking a healthcare professional that you trust to find out about a doctor you are thinking of using. It isn't scientific, but word gets around about whose patients have good outcomes and whom you should avoid.
Finally, talking about the "best" doctor implies that this is a competition. It isn't or shouldn't be. You don't want the best doctor - you want the doctor who will do the best for you. There can be many of those.
To determine the best doctor for you, you need understandable (by you) data on long-term, positive, comparable risk-adjusted outcomes.