A Randomized Trial of Flattery

April 1, 2014

By Michael Tanner, MD

A small body of low-quality evidence suggests that flattery will get you nowhere. We, rigorously applying the tools of evidence-based medicine, set out to test the opposite hypothesis: that when it comes to forming warm, healing relationships with patients, you catch more flies with honey than vinegar.

To determine the effect of compliments on patient self-esteem and satisfaction with their doctors in an inner-city hospital clinic.

Randomized controlled trial of outpatients aged 21-98 at the Bellevue Adult Primary Care Center seeing a physician for the first time.

Patients were stratified by age, sex, body-mass index, and country of origin.

Using a 5-point Likert scale, patients were separated into quintiles by likeability, with quintile 5 being reserved exclusively for real sweetie-pies. At the end of the visit, patients in quintiles 2, 3, 4 & 5 were told: “You are my favorite patient.” Patients in quintile 1 were told: “You are one of my favorite patients.” [See endnote 1.]

Female patients were told by male physicians: “There must be some mistake with the date of birth on your clinic card. You really look much younger than the birth year I’m seeing here. You look to me like you are [actual age in years minus 12] years old.”

Female patients were told by female physicians: “You have really pretty eyes.”

Male patients having their height measured were told: “You are [actual height in inches plus 1.75] inches tall.”

Recent immigrants were asked: “Cuanto tiempo ha estado in los Estados Unidos? El Ingles es magnifico!” (“How long have you been in the US? Your English is astonishing.”)

The control group was randomized to usual care, consisting of a hectic 35-minute office visit during which they were told after being weighed: “You have class [1, 2, or 3] obesity. You really need to eat more fruits, vegetables, whole grains, non-fat dairy, and legumes, and exercise for 30 minutes 5 times per week. You need to lose at least [25-200] pounds. Can I interest you in a biliopancreatic diversion?”

After the visit, patient self-esteem was evaluated using the Rosenberg self-esteem scale. Patients also rated their physicians, using a 5-point Likert scale, with 5 being “I am very satisfied with my doctor” and 1 being “I am never coming back here again.”

Statistical Analysis
All analyses were performed according to the intention-to-lay-it-on-pretty-thick principle. Blushing was quantified using Pantone colors 732 (tickled pink) through 740 (cayenne).

Flattered patients were significantly more likely than control-group patients to agree with the statements: “I feel that I am a person of worth” and “I feel that, compared to other people, I am quite awesome.” (p=<0.0001)

Control group patients were significantly more likely than flattered patients to agree with the statements: “All in all, I am inclined to feel that I am a failure,” “At times I feel I am no good at all,” and “Sometimes I feel so low I could go out and do chin-ups on the curb.” (p=<0.0001)

On February 14, 2014, the steering committee decided to terminate the trial early because it was pretty clear how the thing was playing out.

Humans love praise. Who among us doesn’t like to be told “Hey! You look like you’ve been working out”? Or “That dress looks great on you”? Or “You look exactly like Jennifer Lawrence”?

The antidepressants Celexa®, Zoloft® & Prozac® would vanish if flattery could be put in a capsule (Aplaudid? Accolaid? Ovatia? Kvell?), administered per aurem (by ear).

Conversely, in their heart of hearts, people hate criticism of any kind, constructive or not. As with tickling, a little bit of criticism goes a long way. The sonata form of the “feedback sandwich” goes like this: start with praise, transition to constructive criticism, then recapitulate with more praise. But for most of us, a feedback sandwich is a lot like a tongue sandwich. People really feel: “Oh, feedback sandwich deli man, could I please just have the crisp, airy ciabatta? Hold the tongue.”

That’s why I had Mr. Collins from Bio-Medical Engineering hang the height stick in my office 1.75 inches low. What’s the harm, really, if I tell my 5’6” patient that he’s 5’8”? Even if he gets challenged when he goes back to his neighborhood and starts telling people he’s 5’8”? His buddies may well say, “That’s a lie! You’re 5’6”–tops!” But my patient can counter, “Oh yeah? My doctor measured me and he says I’m 5’8”!”

I have tremendous admiration for many of my patients: the sisters who trade 12-hour shifts taking care of their 95-year-old mother, intensive care unit survivors with healed-over tracheotomy and feeding tube scars, people who have survived cancer, the loving parents of children with Down syndrome… To such patients I always say the following words: “I have tremendous admiration for you.” You can sense the pride they feel when they hear it: “A doctor has just told me that he admires me.” We should always use the power differential for our patients’ benefit. Every word that comes out of our mouths should be directed toward the patient’s recovery. The timely blandishment can only strengthen our therapeutic bond with the patient, and that can only lead to better health. Any favorable thing about the patient should be acknowledged loud and clear.

Saying nice things to another person is an expression of love. When I tell my wife that she is incredibly beautiful, true, I am flattering her, but I am just stating a fact. And it has never gone over badly.

In conclusion, our data clearly demonstrate that, in the medical school curriculum, too much time is devoted to complement and not enough to the compliment. Too much to meconium and not enough to the encomium.

The author wishes to thank every person who has ever paid him a compliment. He has not forgotten a single one.

1. Groves JE. Taking care of the hateful patient. N Engl J Med. 1978;298(16):883-887.

Dr. Michael Tanner is the Executive Editor of Clinical Correlations


8 comments on “A Randomized Trial of Flattery

  • Avatar of Michael Lawrence
    Michael Lawrence on

    Dr. Tanner:

    Your writing es magnifico! This mustbe the work of a gracious and tall physician.This is my favorite website, by far.

  • Avatar of Michael Brabeck
  • Avatar of Joe Shin
    Joe Shin on

    Well done. You are by far my most favorite contributor to clinical correlations, and definitely much much taller than Dr. Shapiro (he does have nice eyes though)

  • Avatar of Gary Starkman
  • Avatar of Sapana shah
    Sapana shah on

    In an era where half of everything we are currently doing to help our patients will be proven to be ineffective in the next 10 years, and the other half of what we do has been sponsored by the pharmaceutical industry, I can’t agree more with your study findings. The compliment will never hurt your patient the way complements can.

  • Avatar of Michelle McMacken
    Michelle McMacken on

    Usually I am wary of studies stopped early, but this is an exception. Thanks for this outstanding piece of research. I plan to fill my next patient care session with compliments!

  • Avatar of Jaime Gonzalez

Comments are closed.