Who is to blame when patients are injured by missed or delayed primary care diagnoses? The finger of blame points in many directions, a new study shows.

The study is one of the first to look at medical errors that happen not in hospitals, but in doctors' offices across the nation. It comes from medical-error researchers Tejal K. Gandhi, MD, MPH; David M. Studdert, ScD; and colleagues at Brigham and Women's Hospital in Boston.

Gandhi and colleagues looked at the medical and legal records for 181 malpractice claims for a missed diagnosis settled against primary care doctors. The harm was serious in 59 percent of the cases; 30 percent of the cases resulted in death.

The main findings:

--A missed cancer diagnosis -- particularly breast cancer -- was the most common error.

--Failure to order the appropriate test, seen in 55 percent of the cases, was the most common diagnostic breakdown.

--Failure to create a proper follow-up plan was the breakdown in 45 percent of the cases.

--Failure to get a proper history from a patient or to perform a proper physical exam was the breakdown in 42 percent of the cases.

--Incorrect interpretation of test results was the breakdown in 37 percent of the cases.

--The typical case had three different breakdowns.

In all of these cases, doctors made mistakes. They had failures of judgment (79 percent of cases); failures of vigilance or memory (59 percent of cases); too little knowledge (48 percent of cases); and failed properly to hand the case off to another doctor (20 percent of cases).

At least two doctors erred in 43 percent of the cases -- and at least three doctors were involved in 16 percent of the cases.

The study doesn't let patients off the hook. Nearly half the time, patients failed to provide their doctors with crucial information or missed appointments.

"The prospects for 'silver bullets' in this area seem remote," Gandhi and colleagues conclude. "Our findings are humbling, and they underscore the need for continuing efforts to develop the 'basic science' of error prevention in medicine, which remains in its infancy."

The study appears in the Oct. 3 issue of Annals of Internal Medicine.

By Daniel J. DeNoon, reviewed by Louise Chang, MD

SOURCES: Gandhi, T.K. Annals of Internal Medicine, Oct. 3, 2006; vol 145: pp 488-496. Wachter, R.M. Annals of Internal Medicine, Oct. 3, 2006; vol 145: pp 547-549.