Certain drug combinations are common but often unrecognized contributors to drug-related harm among older adults, setting off prescribing cascades that add unnecessary costs to the healthcare system, a new cohort study suggests.
"A prescribing cascade occurs when a drug is prescribed, a drug-related adverse event develops that is misinterpreted as a new medical condition, and a second, potentially inappropriate drug is started to manage the adverse event," Paula Rochon, MD, director of research at the Weston and O'Born Centre for Mature Women's Health at Sinai Health in Toronto, and her research manager, Joyce Li, told Medscape News Canada. "While some prescribing cascades are appropriate, potentially inappropriate prescribing cascades can lead to unnecessary medications and medication-related harm."
Last year, the team published what they described as a comprehensive list of 65 potentially inappropriate prescribing cascades (PIPCs) that was developed through a Delphi consensus process with an international panel of experts in geriatric medicine and pharmacology. The list "provides an important expert-informed tool for clinicians and researchers to identify potentially inappropriate prescribing patterns and improve medication safety in older adults," Rochon and Li said.
The current study builds on this work, they said, with an analysis that resulted in a prioritized list of 24 PIPCs that "can help clinicians focus on the most common and potentially harmful patterns."
The study was published online September 10 in BMJ.
'A Prompt to Look More Closely'
The researchers measured the population prevalence of the drugs that started the 65 prescribing cascades identified previously, the incidence of the initial and subsequent drug combination, and the strength of the temporal association between them. The cohort included 2,297,942 community-dwelling adults in Ontario ages 66 years or older (54.3% women; 12.8% aged ≥ 85 years).
Twenty-four of the 65 PIPCs (37%) had high prevalence of the initial drug, high incidence of the prescribing cascade, and high strength of association. One example highlighted in the study involved nonsteroidal anti-inflammatory drugs. Commonly prescribed for pain, these drugs are associated with a rise in blood pressure, which can lead to a new prescription for high blood pressure, rather than a second look at the original pain medication.
The researchers found that cardiovascular drugs were the most common starting point for a PIPC. The prevalence for individual drug classes ranged from 19% for diuretics to nearly 66% for antihypertensives.
The pairs with the highest 1-year incidence of PIPCs were iron supplements followed by laxatives (11.9%), statins followed by pain relievers (10.9%), and cholinesterase inhibitors followed by sleep agents (10.3%).
The strongest timing associations were seen for corticosteroids followed by antipsychotics, laxatives followed by antidiarrheal agents, and cholinesterase inhibitors followed by antiemetics.
One pattern differed by sex. The cascade linking selective serotonin reuptake inhibitor or serotonin-norepinephrine reuptake inhibitor antidepressants to overactive bladder drugs was identified only in men.
"Our findings provide the foundation for tools that could alert prescribers and pharmacists to potentially inappropriate prescribing cascades when a new medication is added or an existing medication is reviewed," Rochon and Li said.
Good documentation can make a difference, they noted. "Recording when a medication was started, why it was prescribed, and who prescribed it makes it much easier to recognize the sequence of events leading to a potentially inappropriate prescribing cascade if a new symptom emerges later," they said.
"These tools and lists are intended to support, not replace clinical judgment," they added. "A prescribing cascade is not always inappropriate; the risks and benefits of continuing the original medication need to be considered in the context of the individual patient and their goals and preferences and discussed through shared decision-making involving the patient and the prescriber.
"Our prioritized list should therefore be viewed as a prompt to take a closer look at the medications, rather than a reason to stop a medication," they said.
'A Pause May Prevent a Cascade'
Commenting on the study for Medscape News Canada, Mark Olfson, MD, MPH, Elizabeth K. Dollard professor of psychiatry, medicine, and law at Columbia University Irving Medical Center in New York, agreed that prescribing cascades are common and clinically important in older adults. Yet he added words of caution. "All the medication sequences identified in this study may not be prescribing cascades. The study demonstrates temporal associations between medications, but that does not establish that the second drug was prescribed to treat an adverse effect of the first." Olfson did not participate in the study.
In addition, he said, "this study does not incorporate information about diagnoses, treatment indications, or patients' comorbidities. These omissions are important because older adults commonly have multiple conditions requiring multiple medications."
A medication started after another medication may reflect treatment of an unrelated condition rather than a prescribing cascade, said Olfson. "The new findings are signals of potential cascades, rather than proof that the sequences were caused by the first medication."
Timing can be "very important," he noted. "When a new symptom appears shortly after a medication is started or the dose is increased, clinicians should pause before starting another prescription. They should ask, 'Why did this symptom appear now?' A pause can sometimes prevent a medication cascade from developing."
Furthermore, he added, medication reviews should consider whether one of the medications already being taken is causing the problem. "When appropriate, reducing the dose, stopping the suspected medication, or switching to an alternative may prevent a prescribing cascade."
This study was supported by a grant from the Canadian Institutes of Health Research. The work of one author was partially supported by Politecnica delle Marche University (Ricerca di Ateneo). All study authors declared support from the Canadian Institutes of Health Research, and Olfson reported having no relevant financial relationships.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.