In recent years, there has been much talk about drug reduction in nursing homes. We hear about “GDR meetings” and even a new verb, as in “We are ‘GDRing’ our residents’ medications.” 

As the Centers for Medicare & Medicaid Services has pushed harder to stigmatize antipsychotics, stepped-up gradual dose reduction and deprescribing actions are all the rage lately. 

After 40-plus years in long-term and post-acute care as a medical director, and extensive personal involvement in helping manage many thousands of patients, the whole topic has left an indelible impression. Medications have revolutionized care and are often highly beneficial. But medication-related misadventures and complications have always remained at the top of my list of concerns. It is not a new issue, it is possible to do it right, but only some out there are doing it right or giving good advice.  

Over the years, I have often heard something like this, “This week, we will do antipsychotics because CMS wants to see less of them. Let’s go down the list of people on antipsychotics and see who we can cut back. Next month, we will do cholinesterase inhibitors and then proton pump inhibitors and then opioids, etc.” 

I remember a supervisor from a psychiatric consultant group insisting that their nurse practitioners be allowed to control all the reviews and changes of psychopharmacologic medication under the guise of “medication management.”

Here is a common example: someone with behavior or mood issues has been getting – among other medications – valproic acid, trazodone, alprazolam, citalopram and risperidone. We may be inclined to immediately conclude that the antipsychotic should be reduced and then removed first, while keeping the rest. 

All of this (GDR, deprescribing, medication management, etc.) sounds very efficient and proficient until we realize that it is highly superficial and artificial. While individual medications may target different organs and symptoms, they never act independently. All organ systems (heart, lungs, brain, etc.) interact continually. The ultimate goal is to try to optimize internal balances (“homeostasis”) in the face of constant change. Therefore, we must always consider the broader impact of treatments and treatment changes on the patient. Truly comprehensive care is a biological necessity, not just a platitude. 

Like so many things (for example, artificial intelligence or AI) that become the latest obsession, Gradual Dose Reduction (GDR) is nothing new; instead, it recycles old ideas in new packaging. More than 30 years ago, along with several other physicians, I helped Dr. Sam Kidder, CMS’s original Consultant Pharmacist, develop the original version of the OBRA-related GDR requirements related to psychopharmacologic medication. It hasn’t exactly turned out as we anticipated. 

For decades, CMS’ OBRA-related guidance has required attempted Gradual Dose Reduction (GDR) unless contraindicated. The guidance has also reiterated the same thing: “Medication management is based in the care process and includes recognition or identification of the problem/need, assessment, diagnosis/cause identification, management/treatment, monitoring, and revising interventions, as warranted as well as documenting medication management steps.”

If this is correct (and it is!), then medications are not the principal focus of the medication “management” process. There is no care process step called “GDR.” Instead, it is just one phase (monitoring) of the entire care delivery process. The RAI manual Chapter 4, Table 2, explains this further. 

Is there a better way?

There is an optimal and universally applicable way to review and adjust any person’s treatment regimen and decide on next steps. First, with rare exceptions, GDR and deprescribing should be done patient-by-patient, looking at each person’s total situation, not drug-by-drug or category-by-category for multiple patients. Don’t make decisions based on predetermined regulatory quotas for the “right” number of any medication, including (but not limited to) antipsychotics. 

Before working on someone’s medications, routinely summarize their physical, functional and psychosocial status, including medical and psychological stability. Define their issues clearly and in detail – not just the symptoms (e.g., headaches, pain, agitation). Always consider that existing medications could be causing symptoms that are being chased by other medications. 

Diagnosis and treatment invariably involve some uncertainty. Trained clinicians are good at predicting the likely consequences of clinical decisions. Others just guess. Do not let well-intentioned but clinically flawed guessing based on limited understanding be part of your facility’s culture. 

Always be prepared to address the results of modifying a medication regimen, especially when the situation worsens or fails to improve as anticipated. Don’t be in a hurry to do GDRs quickly to “get it out of the way.” Sometimes the brain or other organs need days or weeks just to settle down in a new steady state after medication changes occur. Barreling down the GDR track can too easily become like racing a car with a stuck accelerator and failed brakes down a dead-end street and over a cliff. 

When reviewing medications, don’t make unwarranted assumptions ahead of time about what should stop, continue or change. Previous misdiagnosis and one or more inadvisable medications may have already precipitated a problem. Multiple medications often augment or counteract one another; for example, too many medications (multiple antidepressants, opioids, plus trazodone for “behavior”) that stimulate serotonin in the brain can cause major symptoms of “serotonin syndrome.”

Choosing the wrong medication to taper or stop may exacerbate the situation. Not recognizing when tapering or stopping something has caused an undesired return of symptoms can lead down a different but equally bad or worse track.

Always use reliable references (Medscape’s medication reference is my favorite). Don’t rely too much on any one discipline or person to tell you what to do and how to do it. Knowing a lot about drugs is not the same as knowing how to analyze and manage specific cases involving medications. 

This begs a simple question: how can so many licensed healthcare practitioners appear to be so remiss after so many decades about properly performing the most basic tasks associated with prescribing medications safely and effectively? Why haven’t regulatory efforts addressed the problem by now? 

While there are many reasons involving many parties, I will mention three reasons that I have identified over the decades. One, it is not nearly as simple as it may seem, even though following a consistent process makes it easier. Two, it takes more than a snippet of time and conversation to do it right. Third, many licensed healthcare practitioners actually have never learned how to prescribe and manage medications appropriately and safely by following the full care delivery process. 

Just doing GDR is not virtuous unless it is done right. Really, we all need to rethink how we are approaching this whole issue before we continue barreling down the track to trying to fix it. However, well-intended, current “solutions” are not really cutting it.

Steven Levenson, MD, CMD, has spent 43 years as a physician and medical director in long-term and post-acute care settings in Maryland. He is a pioneer in medical direction, including his groundbreaking book, “Medical Direction in Long-term Care.” For 25 years (1988-2013), he served as a CMS advisor and consultant, helping develop clinically relevant surveyor guidance and training in many aspects of OBRA.

The opinions expressed in McKnight’s Long-Term Care News guest submissions are the author’s and are not necessarily those of McKnight’s Long-Term Care News or its editors.

Have a column idea? See our submission guidelines here.