Science1 publisher2 min readPublished
Ontario prescription records narrow 65 suspected prescribing cascades to 24 worth checking
Prescribing cascades are easy to describe and hard to check for. A BMJ team screened Ontario prescribing records and came out with 24 specific drug pairs common enough, and lopsided enough in order, to be worth a look.
The Scientist · Science desk
What happened
- The BMJ paper worked through prescription patterns for nearly 2.3 million Ontario adults aged 66 and older and came out with 24 drug combinations its authors say warrant a second look.
- The starting list was 65 possible cascades assembled in earlier research by 12 international experts in prescribing and the care of older adults.
- The data came from ICES, Ontario's health data institute, and the team examined new prescriptions written during 2022 and 2023.
- The most frequent sequence was iron supplements followed by a laxative: 11.9 percent of people who started iron received a laxative prescription within the year.
- The pattern has a name, potentially inappropriate prescribing cascade, and the paper's worked example is an anti-inflammatory raising blood pressure and a blood pressure drug following it.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability A named list can be coded, and the authors point to pharmacists reviewing medications alongside prescribers and to software flagging a pair before the second drug is written.
- constraint With no indication recorded in the data, a flag can prompt a review but cannot show a prescription was wrong, so any audit built on these pairs will catch legitimate second drugs at a rate nobody has measured.
- decision Each flagged pair puts a choice in front of the prescriber: cut the first drug's dose, swap it for something else, or drop it because it is no longer needed.
- contradiction The paper's stated concern is older women, who take more medications and have more harmful side effects, yet for most pairs the first-to-second link was about as strong in men, so sex is a weak filter for this list.
Counting how often a laxative follows an iron supplement tells you only that the two travel together. The design step that turns it into a test is the ordering comparison. For each candidate pair, the researchers counted how many people used the first drug and how often a second was added within a year [6], then compared how often patients started the drugs in that order against the reverse, adjusting for changes in prescribing over time [7]. Two drugs given to the same sort of patient for unrelated reasons should appear in both orders at similar rates. The lopsided pairs are the ones that survived to the final list of 24 [7].
The records could not show whether the second drug was treating a side effect of the first or a separate health problem [11].
The percentages also belong to a smaller denominator than the cohort. Among people starting statins, 10.9 percent later received a pain reliever [9], roughly one in nine of the people who started a statin [23]. That sequence and the iron one had the highest rates of the 65 pairs examined [10]. Of those 65 candidates, 24 cleared the screen and 41 did not, a survival rate of about 37 percent [22].
Paula Rochon, the lead author, directs research at Sinai Health's Weston and O'Born Centre for Mature Women's Health in Toronto [16]. "These sequences of events are common but often missed in clinical practice," she said in a press release [14].
Timing complicates the one-year window. Some suspected side effects appear soon after treatment begins, while others take months [13]. Everyone in the cohort was living outside a long-term care home when the study began [5].
A pair on the list is not automatically a mistake. Treating a known side effect can be what allows a patient to keep taking a medicine they need [20].
"Our concern is that so often these conversations between the health care prescriber and the patient are being missed, so people don't recognize the sequences of events and that they are connected to one another," Rochon said [15].
What to watch
- Whether Ontario prescribers see these 24 pairs as an e-prescribing alert, and what frequency threshold fires it.
- Whether the pairs replicate in a health system with different prescribing baselines, and in the long-term care residents this cohort excluded.
- Whether follow-up work ties a flagged pair to an outcome, such as a dose reduction or a stopped drug, instead of to a second prescription record.