Survey: Physicians recommend different treatments than those they choose for themselves

Ubel PA. Arch Intern Med. 2011;171:630-634.

Issue: May 10, 2011

The act of making a recommendation influences how physicians think about medical choices, according to a survey.

Peter A. Ubel, MD, of the Sanford School of Public Policy and the Fuqua School of Business at Duke University, and colleagues surveyed two random samples of US primary care physicians (general internists and family medicine physicians) listed in the American Medical Association Physician Masterfile. Covariates included age, sex, race and ethnicity, type of medical degree, practice setting, patient volume, and how they allocated their professional time (eg, seeing patients, teaching, etc.).

The first group (n=500) of PCPs were sent a colon cancer scenario in which they were asked to imagine that they had been diagnosed with the disease and could choose between two types of surgeries or that a patient of theirs had been diagnosed and it was up to the physician to recommend one surgery over another.

Surgery 1 was presented as having a complication-free success rate in 80% of patients, but a postoperative mortality rate within 2 years for 16% of patients. The surgery was also presented with additional complications, including 1% of patients needing a colostomy, experiencing a wound infection, chronic diarrhea or, conversely, intermittent bowel obstruction. Surgery 2 was presented as curing the cancer in 80% of patients while failing in 20%, all of whom died of colon cancer within 2 years.

Of the 242 physicians who completed and returned the survey, 37.8% chose the surgery with a higher death rate for themselves; 24.5% recommended the same treatment to their hypothetical patients. Multivariate logistic regression analysis confirmed that the physician’s perspective significantly influenced treatment choice (OR=0.55; 95% CI, 0.31-0.99).

The second group (n=1,600) of PCPs were randomly assigned to imagine that they had contracted a strain of avian influenza new to the US or that a patient of theirs had been. Each scenario stated that people who contracted the virus had a 10% death rate and a 30% hospitalization rate.

Both scenarios also included a hypothetical experimental immunoglobulin treatment that had been shown in Asia and Europe to cut in half the chance of adverse events from the virus, meaning that with the treatment, there was a 5% death rate and a 15% hospitalization rate, with an average 1-week length of stay. However, the treatment was presented as having a 1% death rate in patients and a 4% rate of neurological paralysis, especially in the lower extremities.

Of the 698 physicians who completed and returned the second survey, 62.9% of respondents chose not to use the experimental treatment, whereas 48.5% chose it on behalf of patients. Multivariate logistic regression analysis showed that the physician’s perspective greatly influenced treatment choice (OR=0.55; 95% CI, 0.41-0.75), but choice did not correspond with the physician’s age, sex or volume of patient care.

“The very act of making a recommendation changes the way physicians weigh medical alternatives,” the researchers concluded. “Better understanding of this thought process will help determine when or whether recommendations improve decision-making.”

“Physicians may be more able to imagine their patients’ abilities to adapt to significant disability than the patients themselves could,” wrote Eric Shaban, MD, and colleagues of the Palliative Care Program, department of medicine, University of Rochester Medical Center. “But only when considering the consequences for themselves can they imagine the level of suffering that might be a by-product of an unfavorable outcome.”

Disclosure: The researchers of the study and the accompanying editorial report no relevant financial disclosures.

Follow HemOncToday.com on Twitter.