INAPPROPRIATE prehospital medication management may be a factor in the hospitalisation of up to 25% of older patients, according to new research published in the MJA
The researchers, from the University of Adelaide and BUPA Health Foundation, called for routine prospective monitoring according to medication-related clinical indicators for older patients as a means for quality improvement in the management of common chronic conditions.
However, there has been a mixed reaction to the clinical indicators used in the study, with a leading GP expressing “serious doubts” about their validity.
The researchers analysed Department of Veterans’ Affairs claims data for 164 813 hospitalisations over 5 years and found that one in four hospitalisations for conditions listed in the clinical indicator set was preceded by suboptimal medication management. The median age of the patients studied was 81 years.
The evidence-based, medication-related indicators of suboptimal processes of care preceding hospitalisation were developed by the researchers and validated by an expert panel.
Dr Evan Ackermann, chair of the RACGP’s national standing committee on quality care, said the data used in the study was “tenuous”. He said data sourced from a financial claims set, particularly in an elderly veterans population, was unlikely to reflect the many variations in this group.
“They are complex patients, who often have multiple morbidities, and really they don’t follow guidelines”, Dr Ackermann told MJA InSight. “To get good, patient-centred care, you have to optimise therapy according to the patient, not these guidelines. These sorts of indicators won’t be applicable to that population.”
The authors noted several limitations in their study, such as not assessing whether implementation of appropriate care processes might have reduced hospitalisations, and an inability to distinguish between diastolic and systolic heart failure in the available data.
While Dr Ackermann said efforts to improve medication management, particularly in elderly patients, were needed, he had “serious doubts” about the use of clinical indicators that had not been validated in a clinical setting.
He said an example was the small number of patients in the study with acute coronary syndrome (ACS) not taking aspirin, beta-blockers, ACE (angiotensin-converting enzyme) inhibitors or angiotensin II receptor blockers and statin, who were likely to have had contraindications to these drugs.
Associate Professor Ian Scott, director of the department of internal medicine and clinical epidemiology at Brisbane’s Royal Princess Alexandra Hospital, said although there were some limitations to the study, it did signal that poor medication use might be a contributing factor in a significant proportion of hospitalisations in this elderly population.
Professor Scott said the clinical indicators used in the study were adequate, as more than 60% were based on level 1 evidence and all had been reviewed by an expert panel.
He said he was in favour of routine monitoring of medication management preceding hospitalisation, but the data would be most effective if it could be delivered to practitioners at the local population level.
“If you are going to encourage [clinicians] to change behaviour, you have to drill down to a local population level so doctors can see what is happening with their own patients”, Professor Scott said, adding that Primary Health Networks would be well placed to take on this role.
Professor Ric Day, professor of clinical pharmacology at the University of NSW, said the clinical indicators had the potential to improve medication management and reduce hospitalisations associated with these “big ticket” conditions, many of which were national health priority areas.
“The fact that there is probably about a 25% identification of suboptimal prescribing practices for people who were hospitalised is worthy of attention”, he said.
Professor Day said the approach used in the study was “not perfect”, also pointing to the use of administrative data for clinical purposes, and that further studies were needed to determine if improved medication management would reduce the number of hospitalisations.
However, he said there were positive messages in the study, including the low rate of suboptimal medication use in ACS patients.
“It’s good to note that practitioners are doing pretty well”, he said. “There are a few caveats, but as a tool, I think it’s a good one.”
Professor Dimity Pond, professor of general practice at the University of Newcastle, said she was also “pleasantly surprised” at the low levels of hospitalisations associated with suboptimal medication use in some categories. For instance, she said, only 0.62% of patients with a gastrointestinal bleed had been taking a non-steroidal anti-inflammatory drug and had not been on a gastroprotective agent.
However, the high use of falls-risk medications in patients who were later hospitalised for a fracture (85.4%) was cause for concern.
Professor Pond said a data extraction tool that flagged patients who might require a medication review would be helpful. “I am very much in favour of the involvement of the pharmacist with the GP in reviewing [medications], particularly for some of these older people.”
(Photo: Monkey Business Images / shutterstock)