The Most Dangerous Prescription in Geriatrics Is the One Nobody Remembers Starting

The patient I've never met

I can describe her anyway. She's on almost every complex Medicare panel I look at.

78 years old. Six chronic conditions. 13 medications. Three prescribers. Two ER visits in the past year — both for falls. On paper, "well-controlled."

Run the first 30 minutes of a real medication review on a patient like her and the picture changes fast. Her anticholinergic burden lands well past the threshold where the risk turns clinically meaningful — an Anticholinergic Cognitive Burden score of 3 or higher, the level consistently linked to elevated fall risk, delirium, and accelerated cognitive decline in older adults.

That burden almost never comes from one prescription. In a patient like this, it comes from five:

• Oxybutynin from urology, for overactive bladder.

• A muscle relaxant on the Beers list, still on the regimen long after the orthopedic episode that started it had resolved.

• Diphenhydramine taken nightly for sleep — prescribed by no one, never on the chart.

• Hydroxyzine from a dermatology visit two years ago for hives, never stopped.

• A first-generation antihistamine in the OTC cold medication she took all winter.

Three of those — oxybutynin, diphenhydramine, hydroxyzine — each carry a high anticholinergic score on their own. Stack them with the rest and the cumulative load is the kind that quietly rewrites a person's fall risk and cognition. No single prescriber did anything wrong. No one was looking at the whole list.

The pattern

She isn't unusual. She's the pattern — the one that shows up again and again in complex older adults, and the one the literature has documented for years.

A medication started 8, 10, sometimes 15 years ago for a real reason at the time. A PPI for reflux during a stressful year. A benzodiazepine after a hospitalization. An anticholinergic for bladder symptoms that have since resolved. A muscle relaxant from an old back injury.

The original indication is gone. The prescriber may have retired. The patient assumes it's still necessary because no one has ever said otherwise.

And every year it stays on the list, the risk profile shifts underneath it.

Kidneys decline. Cognition becomes more vulnerable. Fall risk climbs. The Beers list grows. Drug–disease interactions emerge that didn't exist when the script was written.

The medication didn't change. The patient did.

Why it gets missed

This is the work that doesn't fit in a 15-to-20-minute follow-up. It requires sitting with the full list, the full chart, and the full patient — and asking a question most encounters never reach:

Does this still need to be here?

Deprescribing is one of the highest-impact clinical interventions in chronic care. It is also one of the most consistently skipped.

Not because clinicians don't know it matters. Because the structural setup of a follow-up visit — limited time, single-organ focus, no medication reconciliation budgeted into the encounter — makes it nearly impossible to do well. The cardiologist manages the heart. The endocrinologist manages the glucose. The orthopedist manages the knee. Everyone manages their slice. The whole regimen belongs to no one.

That's not a failure of any individual physician. It's a failure of structure. And structural problems need structural solutions.

What a real CCM touchpoint should be doing

A monthly Chronic Care Management touchpoint is the natural home for this work. It's the one recurring encounter built around the whole patient rather than a single complaint.

Done well, that touchpoint is where a clinical pharmacist sits with the full regimen, calculates the anticholinergic burden, surfaces the duplicates and the never-revisited prescriptions, checks whether the metformin dose still matches the patient's current eGFR, flags the 11-year PPI that no one has reassessed and its possible contribution to a slow B12 decline — and then sequences the deprescribing conversation with each prescribing physician.

That is what CCM exists to deliver.

It is also what most CCM programs skip — because the team running the monthly call doesn't have the clinical depth to run the calculation, much less the conversation that follows. The touchpoint becomes a check-the-box call. Time gets logged. A box gets billed. The medication list stays exactly where it was.

A patient like the one I described might be billed for 14 consecutive months. She'd be served for none of them.

The takeaway for practice owners

Here's the part that matters if you run a practice with a Medicare panel.

If your CCM program is generating revenue but not moving fall rates, ER utilization, or quality measures on your highest-risk patients — the deprescribing layer is almost certainly the missing piece. You're billing for the encounter. You're not capturing the clinical work that makes the encounter worth billing.

That gap shows up in your data. It shows up in your HEDIS measures, in your readmission numbers, in the patients who disenroll because the monthly call never actually helped them. And it shows up, eventually, in whether leadership decides CCM was "worth it."

It is not a workflow problem. It is a clinical-depth problem.

And it is solvable.

— Chisom, Esther's Wellness