75% of discharge summaries don't mention pending test results
■ Primary care physicians often are left in the dark about what tests to follow up on.
By Kevin B. O’Reilly — Posted Aug. 31, 2009
Patients are especially vulnerable to harmful medical mix-ups when they leave the hospital. Doctors and hospitals have devoted much attention in recent years to improving the process of reconciling the medications patients were taking before they arrived with what they should take after discharge.
New research highlights another continuity-of-care challenge: following up on tests ordered in the hospital. Four in 10 patients are discharged with test results pending, and about 9% of those tests should lead to a change in the patient's care.
But how can primary care doctors check up on these tests if they do not know about them?
Office-based physicians face this situation at an alarmingly high rate, according to a new study. Researchers reviewed medical charts for nearly 700 hospital patients at two academic medical centers and found about 3,000 tests with pending results. Only 16% of those tests were documented in the patients' discharge summaries, said the study, published in the September Journal of General Internal Medicine.
Three-quarters of the discharge summaries did not mention any of the pending test results. Only 13% documented all pending tests.
The failure rate means that primary care doctors nationwide could be unaware of pending tests that could change the direction of treatment for as many as 300,000 patients hospitalized each year, said Martin C. Were, MD, the study's lead author.
"Physicians are doing a terrible job during the transition of care of informing the follow-up [practitioners] about what they should know about a patient's pending tests," said Dr. Were, assistant professor of medicine at Indiana University School of Medicine and a research scientist at the IU-affiliated Regenstrief Institute Inc. "We have bad transitional care during the distance from the inpatient to the outpatient setting."
It did not matter much who prepared discharge summaries, the study found. Attending physicians, medical residents and nurse practitioners all did poorly.
"It's very easy to start blaming doctors, but really the doctors aren't the problem -- the system is the problem," Dr. Were said.
The IT fix?
Often, physicians and others preparing discharge summaries do not know about all the tests ordered during a patient's stay, he said. Dr. Were and his team are developing a system to link discharge summaries with an electronic notification of any pending test results. They hope to have the system up and running by the end of 2010 at Wishard Memorial Hospital in Indiana.
Discharge summaries often are unreliable, with more than a third containing errors or inaccuracies, according to previous research. Still, experts were taken aback that pending test results were documented in just 25% of cases.
"I'm not surprised that it's low, but I am a bit surprised at how low it is," said Chaim Bell, MD, PhD, assistant professor of medicine and health policy management and evaluation at the University of Toronto and staff physician at St. Michael's Hospital.
Medication reconciliation, pending test results and follow-up appointments are the three areas hospitals and physicians need to include in discharge summaries to help primary care doctors give hospitalized patients the right care.
"If you don't highlight them, they can be easily lost in everything else that happened in the summary of the admission," Dr. Bell said.
Dr. Bell co-authored a study, also in the September Journal of General Internal Medicine, examining how physicians fared with a new electronic discharge summary system. Housestaff at St. Michael's found it easy to use the Web-based system, which allows them to note pending test results, medication changes and other care during the patient's stay. But primary care doctors did not see much difference in the quality, completeness, organization or timeliness of the electronic summaries when compared with dictated discharge notes.
The study was too small to yield meaningful information about any difference in patient outcomes. The key to improvement, Dr. Bell said, is to simplify the process for hospital physicians to include the drug, test and appointment information that primary care doctors need.
"We need to make it easy for people to fill in those components," Dr. Bell said. "Is it going to be 100% because of that? No, but it will improve from something like 25%."