Keeping track of aging patients’ medications

From the Worcester Telegram & Gazette, January 28, 2008

By Elizabeth Cooney TELEGRAM & GAZETTE STAFF

WORCESTER— Older people take a lot of medications.

That’s a good thing, allowing them to live with chronic conditions such as high blood pressure or diabetes.

Older people also tend to be hospitalized more than younger people, going home with almost half of their medications changed from the ones they went in with.

That can be a bad thing if the transition from hospital to home and from hospital specialist to primary care physician isn’t seamless. Given the greater vulnerability to side effects from drugs that comes with age — kidneys don’t clear substances as easily as they once did, for example — older people can be at particular risk for adverse drug events at this crucial point.

Researchers at the Meyers Primary Care Institute, which is a collaboration of University of Massachusetts Medical School, the Fallon Clinic and Fallon Community Health Plan, are embarking on a project to test health-information technology tools as solutions to the dangers of handoffs from hospital to primary care when it comes to prescribing and monitoring medications in people over 65. Institute executive director Dr. Jerry H. Gurwitz will lead the study, funded by a three-year, $1.2 million grant from the federal Agency for Healthcare Research and Quality. The goal is to see if systems based on patients’ electronic health records prevent adverse drug events, emergency room visits and repeat hospitalizations. AHRQ is making similar grants across the country focusing on transitions in medical care.

Dr. Gurwitz is a geriatrician who has spent much of his career studying drugs that older people take and the harm that can befall them when preventable errors occur, whether they live in nursing homes, are patients in hospitals or get their care in doctors’ offices. For years researchers focused their attention on identifying particular drugs that posed hazards for older people, advising doctors to avoid prescribing them.

But the problem isn’t so simple, he said last week, citing a study published last month in the Journal of the American Medical Association. Researchers from the Centers for Disease Control and Prevention discovered that those few drugs accounted for only about 3 percent of problems related to older people who arrived at emergency rooms with adverse drug events, from allergic reactions to overdoses to falls.

“We’ve come to the conclusion that it’s really all the drugs that doctors and other health providers prescribe,” Dr. Gurwitz said. “It’s not just a certain finite list of medications that lead to problems. Mainly it’s how drugs are used, how drugs are prescribed, how drugs are monitored and how patients take them that can lead to problems.”

A lot of hope is being placed in health information technology to solve these problems, he said, but until electronic interventions are tested, the jury is still out.

“If we are going to convince the nation and the health-care systems across the country that it’s worth the investment, we have to make a compelling argument through science and evidence that these things work,” he said.

That’s where the conversion to electronic health records completed last year at Fallon Clinic comes in. The Meyers study will test what is known as a medication-reconciliation system on the nearly 30,000 members of insurer FCHP’s senior plan who get their medical care from Fallon Clinic. When members of the plan go to hospital emergency rooms or become hospitalized, their primary care physicians in some cases will receive e-mail alerts about changes in their medications and be notified when high-risk medications require additional monitoring. People who schedule appointments will also receive alerts and additional messages will be sent if the appointments or lab tests don’t happen.

These encounters will be assessed anonymously in order to count how many patients land back in the ER or hospital and see how many of their doctors got the special alerts.

“Right now, a piece of paper is faxed to (primary care physicians), but we’re going to use technology to highlight the changes to make sure it’s very clear what’s happened to the medications, what’s been stopped, what’s been started and what kind of monitoring needs to take place,” said Dr. Lawrence Garber, an internal medicine doctor and architect of Fallon Clinic’s transition from paper to electronic medical records. Dr Garber, Dr. Jennifer Tjia and Dr. Terry Field are co-investigators on the Meyers project.

Beth Israel Deaconess Medical Center in Boston uses a medication-reconciliation system that automatically updates medication lists for its patients when they are seen at its affiliated primary-care clinics or in the hospital. The system also includes the decision-support piece of the package, flagging possibly harmful drug interactions and advising lab tests to monitor side effects if warranted.

Dr. John D. Halamka, chief information officer at Beth Israel Deaconess and Harvard Medical School as well as an emergency physician, said the benefits are compelling.

“You typically see in the literature and in our experience that by putting in place e-prescribing and all the electronic medical record controls, you can reduce medication errors by about 50 percent,” he said. “About 2 million preventable medical errors are committed every year that cause harm. That’s equivalent to a 747 crashing every day and killing everybody on board.” Among the elderly, about 14 percent of people leaving the hospital had medication discrepancies and 14 percent of that group needed to be rehospitalized, three times more likely than patients returning for other reasons, researchers at the University of Colorado Health Sciences Center reported in the Archives of Internal Medicine in 2005. Half of those discrepancies were patient factors, which could mean that the patients couldn’t afford to fill the new prescriptions they were given or they misunderstood their instructions.

Talking with patients soon after they are discharged can help solve that type of problem, too, according to Dr. Stefan Gravenstein of Quality Partners of Rhode Island and Warren Alpert Medical School of Brown University.

“When the instructions say two pills two times a day, do they mean four pills?” he said. “Physicians and clinicians are comfortable with how to interpret that, but others might get confused. If somebody sits down with them and goes over the medications with them, they can pick up discrepancies.”

The Meyers project concentrates on that vulnerable period after hospital discharge. It seems like a no-brainer that alerts and prompt doctors’ visits will help, Dr. Gurwitz said, but then there’s also the danger of “alert fatigue,” when so many things are flagged that an overloaded clinician no longer pays attention.

“There are people who believe in this with almost religious fervor, but there’s always the risk of unintended things happening,” Dr. Gurwitz said. We have to figure out what works and what doesn’t.”

Dr. Garber thinks the timing is right for such a trial.

“With the advent of electronic health records, we have the tools to allow us to do the interventions so that we can eliminate preventable adverse events,” he said. “We now have the toolbox that allows us to make any fix we need to do. We just need to know which fixes.”