Elderly inpatients’ risk of receiving potentially inappropriate medication (PIM) varies widely depending on where in the country they’re hospitalized and the specialty of their attending physicians, according to a study in the March-April edition of the Journal of Hospital Medicine.
Hospitalists may be encouraged by the fact that they, along with geriatricians, internists, and family physicians, were less likely than cardiologists to prescribe PIMs. Still, the major take-home message of the study is to “examine your individual practice and think about whether it’s appropriate to prescribe these medications,” says lead author Michael Rothberg, MD, assistant professor of medicine at Tufts University School of Medicine in Boston.
PIM use was highest in hospitals in the South. There, 55% of elderly patients received at least one PIM, compared with 34% of patients in Northeastern hospitals, where PIM use was lowest. The exact reason for this discrepancy is not known, but Dr. Rothberg hypothesizes that “we tend to prescribe like people in our hospital and like people in our region.” In other words, “it has to do with learning from the people around us.”
Most interesting to him is the wide variation in prescribing practices among individual doctors—even within the same specialty. “The decision to prescribe a drug is based on the individual provider and has to do with how you as a doctor feel about these drugs,” he explains. Although nearly half of all of the patients had received at least one PIM, there were seven hospitals in which those drugs never were prescribed. Somehow, “they found a way to care for people without [those medications],” he points out.
PIM use has been examined among elderly outpatients and nursing home residents, but only a handful of small studies have looked at the problem in hospital inpatients, says Dr. Rothberg. He and his coauthors used data from hospitals across the United States participating in Perspective, a database developed by Charlotte, N.C.-based Premier to measure quality and healthcare utilization.
The survey included patients 65 years or older admitted between Sept. 1, 2002, and June 30, 2005. Their principal diagnoses were acute myocardial infarction, chronic obstructive pulmonary disease, chest pain, community acquired pneumonia, congestive heart failure, ischemic stroke, or urinary tract infection. Surgical patients were excluded. Using the 2002 update of the Beers criteria for PIM use in older adults, the authors identified the total number of PIMs administered to each patient during his or her hospital stay. They further classified each PIM as high- or low-severity, based on the expert consensus expressed in the 1997 update of the Beers criteria.
Data were available on 493,971 patients from 384 hospitals. Of those individuals, 49% received at least one PIM, and 6% received three or more. Thirty-eight percent of patients received at least one PIM with a high severity rating.
The three agents most likely to be prescribed were promethazine, diphenhydramine, and propoxyphene—probably because these drugs treat the problems most commonly encountered in hospitals, such as allergies, sleep problems, nausea, and pain, Dr. Rothberg says.
Hospital region emerged as the most important predictor of PIM use. Compared with patients in the Midwest, patients in the South had an odds ratio of 1.63 of receiving a high-severity PIM. The odds ratio for patients in the West was 1.43. Patients in the Northeast had an odds ratio of 0.85.
The median rate of prescribing high-severity PIMs was lowest among geriatricians, at 24%. Rates among hospitalists, internists, and family physicians were 33% to 36%. Cardiologists had the highest rate: 48% prescribed at least one high-severity PIM.
Interestingly, older patient age also was associated with a lower risk of PIM use. Of patients 85 or older, 42% received at least one PIM, compared with 53% of patients age 65 to 74 (p<0.0001). This suggests that “doctors are aware that the older patients are more frail and vulnerable” and take extra care to avoid prescribing PIMs to people in that age range, Dr. Rothberg says. A diagnosis of stroke or chronic obstructive pulmonary disease also was associated with a lower risk of receiving a PIM—further evidence that “doctors were, to some extent, taking patient factors into account” when prescribing medication.
PIM use among inpatients, as reported in this study, far exceeds the rates published for elders dwelling in the community or in nursing homes, writes Daniel S. Budnitz, MD, MPH, in an editorial accompanying the study.
The wide variation in prescribing practices means each facility must monitor its use of PIMs, just as individual hospitals monitor antibiotic use and resistance, advises Dr. Budnitz, a medical officer in the Division of Healthcare Quality Promotion at the Centers for Disease Control and Prevention. He also points out that the evidence that PIMs cause clinically significant adverse events is “weak and based largely on observational studies with inconsistent results.” The drugs in the Beers criteria are “potentially” inappropriate, he says, but some centers have recategorized them as “ ‘always avoid’ medications, ‘rarely acceptable’ medications, and medications which, indeed, have ‘some indications’ for use in older adults.” Thus, some variation among hospitals may be acceptable.
Rather than concentrate on the Beers criteria, hospitalists should focus “on identifying and mitigating the most common and most severe adverse drug events occurring in their hospitals,” such as bleeding from anticoagulants, hypoglycemic events from insulin, and oversedation from opioid analgesics, Dr. Budnitz points out. TH
Norra MacReady is a medical writer based in California.