An 87-year-old man went to the hospital with COVID-19. His memory was a little shaky on a good day. Doctors gave him steroids to treat the illness.
While in the hospital, he grew confused. At night, he tried to climb out of bed. He fought the staff. A sitter was posted at his bedside.
Dr. George Hennawi, a geriatrician who directs the Center for Successful Aging at MedStar Good Samaritan Hospital in Baltimore, was called in. The man had already been given a dose of medicine to control the agitation.
Hennawi’s team looked at the history of his stay. They lowered the steroid dose and moved it to the morning.
“Gradually, he returned to baseline, and we were able to discontinue the sitter and get him up and walking again,” Hennawi told The Epoch Times in an email.
The choice between treating the new symptom and looking to the last prescription is a familiar fork in geriatric care.
A study of nearly 2.3 million older adults, published Thursday in The BMJ, quantified how often these patterns appear. Researchers started with 65 suspected drug pairings. They found 24 in which one medicine was often followed by another used to treat a known side effect of the first.
The strongest link mirrored Hennawi’s patient: a steroid, then an antipsychotic.
The records show which drug came first, not why the second one was prescribed. Researchers could not say how many of these pairings were true prescribing cascades. However, they appeared often enough in that order to deserve a closer look.
A prescribing cascade happens when a side effect from one medicine is mistaken for a new health problem and treated with another drug.
Hennawi drills the same rule into his residents: When a new symptom follows a new prescription, look at the medicine first.
“These sequences of events are common but often missed in clinical practice,” Dr. Paula Rochon, the study’s lead author and a geriatrician at Sinai Health in Toronto, said in a statement.
What Researchers Found in 2.3 Million Older Adults
Ontario’s public drug program keeps a record of prescriptions dispensed to older residents. Researchers used those records to study nearly 2.3 million community-dwelling adults 66 and older.
They started with 65 possible prescribing cascades identified by an international panel of experts. Then they asked three questions of each pairing: Is the first drug widely used? Does the second drug follow often? And does it follow more often than expected?
Twenty-four combinations passed all three tests—and many of them are in American medicine cabinets.
Iron pills topped the list. Nearly 12 percent of older adults who started iron received a prescription laxative within a year—about one in eight.Statins came next: 10.9 percent of new users, about 1 in 9, were given a prescription pain reliever. Statins were also among the most widely used drugs in the study, dispensed to 55 percent of the older adults.
Third was a pairing that will matter to families caring for someone with dementia. About one in 10 people who started a cholinesterase inhibitor, a class of drugs that includes donepezil, were prescribed a sleep aid within a year. That matters because some sleep medicines can themselves impair cognition in older adults.
The sequence caught the attention of Judith Beizer, a clinical pharmacist and professor at St. John’s University. Donepezil can interfere with sleep in some patients, she told The Epoch Times, and moving the dose to the morning may help before adding another medicine.
Other patterns will sound familiar to anyone who has watched a parent’s pill organizer fill up. About 7.5 percent of people who started a calcium channel blocker such as amlodipine—a common blood-pressure drug that can swell the ankles—were prescribed a water pill within a year. Nearly one in 18 new steroid users were given a sleep aid. About one in 15 people who started an anti-inflammatory such as naproxen got a stomach protector, and 3 percent got a new blood-pressure medication.
Hennawi has seen these chains for years.
“What surprised me most was not the specific cascades themselves, but the scale of the problem,” he said. The study “moved prescribing cascades from a conceptual issue to a measurable one.”
The study also leaves part of the medicine cabinet invisible. It did not count over-the-counter drugs, which the researchers said may mean cascades are undercounted. In both Canada and the United States, common pain relievers, acid reducers, and iron supplements can be bought without a prescription.
“Most people are not prescribed iron,” Beizer said. “The doctor says, go to the pharmacy and pick up a bottle of iron.” Those purchases leave no prescription trail.
The same is true of pain medicine. Someone who develops aching muscles after starting a statin may simply buy ibuprofen. The study would never see it. Asked how much larger the anti-inflammatory cascades might be once over-the-counter use is counted, Aaron Tejani, a pharmacist and researcher at the University of British Columbia, told The Epoch Times, “I would suspect it is much bigger.”
Why the Link Gets Missed
In older adults, side effects often look like aging.
“Dizziness, swelling, urinary symptoms, fatigue, insomnia, and constipation are all common complaints,” Hennawi said. “Without careful medication review, it’s easy to assume a new diagnosis is responsible when the medication list may hold the answer.”
The statin finding drew more skepticism. Nearly 11 percent of people who started a statin later received a prescription pain reliever.
“I think the one that surprised me the most was the statin to pain reliever,” Beizer said.
Dr. Ann Marie Navar, a cardiologist at UT Southwestern Medical Center, cautioned against reading that number as evidence that statins were causing muscle pain. Musculoskeletal complaints are common in older adults, she said, and nearly as many people in the study had started a pain reliever before the statin as after it.
“I think it is a stretch to consider those things to be causal,” Navar told The Epoch Times in an email.
That uncertainty cuts both ways. The study also missed over-the-counter pain relievers, meaning the prescription records cannot capture the full picture.
For a patient, the practical question is simpler. If aching legs begin after starting a statin, the timing is worth noting rather than assuming the pain is simply part of getting older.
Other familiar side effects slip through. A dry cough from an ACE inhibitor such as lisinopril is one of the reactions taught early in medical training. Yet 2.1 percent of new users in the study later received a cough remedy.
Dr. Aaron Troy, a clinical and research fellow in cardiology and geriatrics at Johns Hopkins School of Medicine, told The Epoch Times that patients who develop a cough after starting lisinopril should ask whether the drug could be responsible—and whether an alternative, such as losartan, would be a better fit.
Some symptoms are simply hard to trace. Dr. Nancy Schoenborn, a geriatrician at Johns Hopkins, pointed to several cascades that end with a nausea drug. Nausea “is a pretty nonspecific symptom,” she told The Epoch Times in an email, particularly when the first medication was started months earlier.
Depression may be another blind spot. About one in 16 new opioid users and one in 19 new beta-blocker users were prescribed an antidepressant within a year. Clinicians readily think of fatigue or dizziness with those drugs, Schoenborn said, but “depression is less likely thought of as a side effect.”
The calendar hides the trail, too. The researchers noted that some cascades emerge quickly while others take months to appear.
“Time is probably the biggest barrier,” Hennawi said. Older adults may have long medication lists and several specialists prescribing at different times. Reconstructing when a symptom began, and what changed just before it, can be difficult in a short office visit.
The records can be scattered, too. In the United States, one patient may fill prescriptions at a neighborhood pharmacy, through mail order, and through the Department of Veterans Affairs.
“It would be so hard to investigate if the information was in many places,” Tejani said. Even Ontario’s record, he added, does not contain all the clinical detail needed to determine whether a possible cascade really was one.
Someone is supposed to be watching the whole list, Dr. Dave Rakel, chair of family medicine at the University of Wisconsin School of Medicine and Public Health, told The Epoch Times.
That job belongs to the primary care doctor, “the quarterback of care,” who can see the big picture and remove drugs that conflict or are no longer needed, Rakel said. His clinic tries to do that review at every visit. Having a pharmacist on the primary care team, he added, helps.
When the Second Drug Makes Sense
The prescription records cannot show why the second drug was added. “This study identifies patterns and associations, not individual prescribing decisions,” Hennawi said.
Sometimes the second drug is exactly what the patient needs.
Beizer pointed to iron. Constipation is expected, she said, so a laxative may be added on purpose. The same often happens with opioids after surgery.
Other times, the better move is to revisit the first drug. If amlodipine is causing swollen ankles, Schoenborn would rather avoid adding a diuretic.
“I’d rather see if we can get rid of the swelling by making a switch than adding another medication,” she said.
But the swelling may have another cause. Shortness of breath or worsening swelling should prompt a closer look rather than assuming amlodipine is to blame, Troy said.
Sometimes there is no good substitute for the original drug. Then the question becomes whether the side effect is bothersome enough to treat.
“Sometimes people prefer not to ‘rock the boat,’” Schoenborn said.
The Holes in the Safety Net
The study’s authors want prescribing software to flag possible cascades and pharmacists to review them. In practice, both face limits.
“I don’t think we have prescribing cascades built into some of our prescribing software at this point,” Beizer said. “We have interactions.” More alerts can create another problem: “There are also too many alerts, and then you get alert fatigue.”
A pharmacist review could catch what software misses, but many older adults never receive one. Medicare Part D offers Medication Therapy Management to certain eligible patients, but not everyone qualifies. The program has fallen short of the kind of pharmacist-led medication review many hoped it would provide, often focusing instead on adherence and cost, Beizer said.
Even when a possible cascade is flagged, someone has to investigate it. Pharmacists may simply be “too busy to investigate.” He said paying them specifically for that work could help, Tejani said.
What Can a Patient Do?
When a new symptom appears, start by asking: What changed just before it?
A drug started days or weeks earlier may be easy to forget. Hennawi tells patients and families to ask one simple question: “Is it possible that my or my loved one’s symptom is related to a medication?”
Then look at the whole list. Patients can ask for an appointment specifically to review their medications, Rakel said.
Hennawi recommends bringing the actual bottles, including over-the-counter drugs and supplements, at least once or twice a year and whenever a new symptom appears. Know why each one is there, when it was started, and whether it is still doing a job.
He has seen why it matters. One patient who came in with weakness was taking two cholesterol medicines, even though he had been told to stop one. Hennawi has found old prescriptions still in use and products the medical team knew nothing about.
“The clinicians are your partners,” he said. “This is a good way to help us rather than burden us.”
Sometimes the useful information is a few numbers. Blood-pressure drugs were the most widely used medicines in the study, taken by two-thirds of the older adults.
Anyone who feels lightheaded after starting one should first sit down and drink water to avoid a fall, Troy said. Then check your blood pressure while the dizziness is happening, and again first thing in the morning and last thing at night for three days. Send those readings to your clinician, along with anything that seemed to trigger the spell.
That record, he said, lets the doctor adjust or stop the medication “rather than adding another.”
The 70-year-old with the aching legs has options, too. Beizer’s advice is neither to grin and bear it nor to quit the pill. Ask the doctor about a lower dose or a different statin; the drugs should work the same on paper, she said, but people react to them differently. If no statin is tolerable, other cholesterol drugs exist.
The questions can start at the pharmacy counter, before the first pill. Ask which side effects are most likely and what can be done if one appears. Rakel distills it down to one question: “What are the side effects and how can I best manage them?”
Using one pharmacy when possible also gives the pharmacist a better chance of seeing the full picture. If some drugs arrive by mail, make sure the local pharmacy knows about them.
If you suspect a drug is causing a problem, “don’t stop it on your own,” Beizer said. “Talk to your pharmacist. Talk to your physician.” A doctor may change the medication or lower the dose. Or, in some cases, determine that the original medicine still matters enough to keep.
That is the balance Hennawi wants patients to understand.
“Many medicines are extremely beneficial,” he said. “The goal is to ensure that symptoms are interpreted correctly and that every medication continues to provide more benefit than harm.”