A newer set of criteria used to flag potentially inappropriate medicines prescribed in older patients was better at identifying avoidable adverse drug-related events that led or contributed to hospitalizations than was a preexisting tool, according to an Irish study of ambulatory people aged 65 years and older.
STOPP (Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions) could "be highly valuable as a routine screening tool," concluded Dr. Hilary Hamilton and her coauthors at University College Cork, Ireland, in the June 13 issue of Archives of Internal Medicine (2011;171:1013-19). The researchers belong to the same study group that devised and validated STOPP in the late 2000s.
In the more recent, prospective study, they compared use of the two sets of criteria in identifying potentially inappropriate medicines that could cause adverse drug events in 600 patients admitted to a university hospital with an acute illness over a 4-month period. Their median age was 77 years, and they were on a median of seven medications (range 1-27).
The STOPP criteria are organized by physiological system. Devised in the late 2000s, the tool flags possible drug interactions, duplicate drug-class prescriptions, and the use of certain medications in patients with current falls, among other potential problems.
In contrast, the latest (2003) version of the Beers criteria, first published in 1991, uses two lists of drugs that should be avoided in older people, one that is independent of the diagnosis and one that considers the diagnosis. Dr. Hamilton and her coauthors pointed out that the lists include drugs that are not available in European countries, and that the results of studies on the association between the potentially inappropriate medicines identified with the Beers criteria and adverse drug events have been mixed.
Of the 600 patients, 158 (26%) had 329 adverse drug events. Of these 329 events, 36 (11%) were considered to be the main cause of the hospitalization, and 183 (56%) were considered a significant contributory factor to the hospitalization. Of these 219 causal or contributory cases, 151 (69%) were determined to be "avoidable or potentially avoidable."
Significantly more of the adverse drug events that were due to potentially inappropriate medicines were identified using the STOPP criteria: Of the 329 adverse drug events, 170 (52%), compared with 67 (20%) that were identified with the Beers criteria.
Of the 329 adverse drug events, 235 or 71% were classified as avoidable or potentially avoidable, of which 68% were identified as potentially inappropriate medicines by the STOPPS criteria compared with 29% of cases using Beers criteria.
In addition, the likelihood of patients experiencing an adverse drug event was nearly 85% higher if they were prescribed a medicine that was identified as being potentially inappropriate with the STOPP criteria than a medicine not identified as potentially inappropriate with these criteria, after adjusting for age, sex, comorbidity, chronic cognitive impairment, baseline activities of daily living, and number of medications.
However, treatment with a medication deemed potentially inappropriate using the Beers criteria was not associated with a significant increase in adverse-event risk, they said.
The results indicate that the STOPP criteria are more sensitive than the Beers criteria for identifying those potentially inappropriate medicines that result in adverse drug events, "and are therefore more clinically relevant," the authors concluded.
The researchers wrote that not including over-the-counter medications was among the study’s limitations, and that the criteria are not meant to replace clinical judgment but "are designed to enhance clinical evaluation of pharmacotherapy in older patients."
They also wrote that the STOPP criteria, which address common avoidable cases of inappropriate prescribing, are designed to be used with the START (Screening Tool to Alert Doctors to Right Treatment) criteria, which represent "the more common instances of inappropriate omission of potentially beneficial medication for no valid clinical reasons." The authors referred to the "substantial" prevalence of potentially inappropriate omission of beneficial drugs in older people in both primary care (23%) and hospital settings (58%).
In an accompanying editorial, Dr. Jeffrey Schnipper, of the division of general internal medicine, Brigham and Women’s Hospital, Boston, said the study is important because "it facilitates the design of better interventions to improve medication safety among ambulatory elderly patients." He added that incorporating this information into electronic prescribing and outpatient pharmacy systems would be "the most obvious" application. He had no financial disclosures.
Funding for the study was provided by the Health Research Board of Ireland and Enterprise Ireland. The authors had no financial disclosures.