Adverse Drug Events in Nursing Homes: Common and Preventable

August 9, 2000

WORCESTER, Mass.-Medication-related injuries in nursing homes are common and often preventable, according to authors of the largest study to date evaluating adverse drug events due to medication errors in U.S. nursing homes.

More than half of the 546 medication-related injuries that were identified in the 18 community nursing homes were preventable, according to Jerry H. Gurwitz, MD, the Dr. John Meyers Professor of Primary Care Medicine at the University of Massachusetts Medical School and executive director of the Meyers Primary Care Institute, a joint initiative of the Fallon Health Care System and UMMS. The study, "Incidence and Preventability of Adverse Drug Events in the Nursing Home Setting," was published in the August 2000 issue of the American Journal of Medicine.

Of the total number of adverse drug events, one was fatal, 31 (6 percent) were life-threatening, 206 (38 percent) were serious and 308 (56 percent) were significant, according to the report. The study also identified 188 potential adverse drug events or "near misses", 80 percent of which were associated with the use of the blood thinner, warfarin. Interestingly, more severe adverse drug events were more likely to be preventable than less serious ones, but the reason is unclear. "Clearly this offers an incredible opportunity to impact meaningfully on the health of a very vulnerable patient population," said Gurwitz. Currently there are over 1.5 million residents of U.S. nursing homes.

Psychoactive drugs (antipsychotics, antidepressants, sedatives and hypnotics) and anticoagulants were the most common medications associated with preventable adverse drug events (oversedation, confusion, hallucinations, delirium, and falls). More than 36 percent of the nursing home residents used antidepressants, 24 percent used sedatives/hypnotics and 17 percent used antipsychotic medications. Flawed prescribing of psychoactive medications in nursing homes has long been a concern, the authors of the study noted.

Preventable adverse drugs events occurred most often when health care providers ordered the drugs (47 percent) and when caregivers failed to adequately monitor residents (49 percent). Among the ordering errors, wrong doses were the most common problem followed by ordering drugs with harmful interactions, and wrong choices of drugs. Most common among monitoring errors was inadequate lab monitoring of drug therapies and failure or delay in responding to symptoms or signs of drug toxicity.

"This study points out that it's not just a small list of so-called 'bad drugs' in the elderly that we need to be concerned about, it's the whole range of drugs," Gurwitz said. "However, we have identified some drug categories that appear to cause more problems than others. We're not saying people shouldn't be on these drugs. We're saying that sometimes providers are not monitoring residents appropriately and are missing signals that patients are experiencing adverse effects from drugs."

If the study's findings can be generalized, an average-sized U.S. nursing home (106 beds) will have at least 24 adverse drug events and eight "near misses" per year, according to the report. If the findings are extrapolated to the 1.5 million U.S. nursing home residents, then at least 350,000 adverse drug events occur every year and more than half are preventable. The authors consider these estimates to be conservative.

The report recommended:

  • Enhanced surveillance and reporting systems for adverse drug events and medication errors occurring in the nursing home setting;
  • Continued educational efforts relating to optimal use of drug therapies in the frail elderly patient population; and
  • Systems-based prevention strategies to prevent ordering and monitoring errors in drug therapy.

A report from the Institute of Medicine published earlier this year ("To Err is Human") made similar recommendations.

The study, funded by a grant from the National Institute on Aging (one of the National Institutes of Health), is part of an ongoing series of studies being performed by the investigators. The NIA, in conjunction with the Agency for Healthcare Research and Quality, is also currently funding a study to identify strategies to improve the safety of medication use in the ambulatory geriatric patient population.

In July 2000, Gurwitz and colleagues began translating their research findings into practice. They are examining the impact of computerized physician ordering systems for reducing medication errors in two large long-term care facilities: Baycrest Centre for Geriatric Care in Toronto and Masonicare in Wallingford, Conn. The three-year study is funded by the Agency for Healthcare Research and Quality.

Physician order entry programs can alert prescribers of medications to potential problems such as excessive doses based on a patient's poor kidney function, allergies, or a dangerous interaction with a drug the patient is already receiving. These programs can also cue health care providers to monitor certain patients more closely for early signs of adverse drug events.

"We're interested in intervening and educating health care providers at the time when a drug is ordered," Gurwitz said. "If we prove physician order entry works in these model facilities, it might lead to health care policy changes to encourage widespread use of this technology in all U.S. nursing homes." Such technologies have already been shown to prevent serious medication errors in the hospital setting.

The University of Massachusetts Medical School is one of the fastest growing medical schools in the country, attracting more than $93 million in research funding annually. A perennial top ten finisher in the annual US News & World Report ranking of primary care medical schools, UMMS comprises a medical school, graduate school of nursing, graduate school of biomedical sciences and an active research enterprise, and is a leader in health sciences education, research, and public service. It is the academic partner of UMass Memorial Health Care, and a proud partner of Fallon Health Care System in the Meyers Primary Care Institute.

Related websites:
National Institutes of Health (www.nih.gov) This link goes to another web server
National Institute on Aging (www.nih.gov/nia) This link goes to another web server
Institute of Medicine (www.iom.edu) This link goes to another web server
Agency for Healthcare Research and Quality (www.ahcpr.gov) This link goes to another web server
Fallon Health Care System (www.fallon-clinic.com) This link goes to another web server