ECT for Dementia? The Dangers of the BPSD Mindset

[ECT – electroconvulsive therapy;

BPSD – behavioral and psychological symptoms of dementia]

by: G. Allen Power, MD



It is hard to believe that my first blog post on this subject was 8 years ago, colorfully titled ECT for AD and LBD? WTF…( In that post, I criticized researchers for giving electroconvulsive therapy (ECT) to people living with dementia who were exhibiting “severe agitation or aggression.” This topic has jumped back into the limelight recently with the announcement of a new $11.8 million NIA grant—to test ECT on a total of 200 people living with Alzheimer’s disease who have been referred to five geri-psych inpatient centers, including those affiliated with Harvard, Mayo Clinic, and Emory University.

I guess it’s time to update that post.

In recent decades, electroconvulsive therapy has been used mainly for the treatment of severe depression, and has been found to be helpful in many cases where one’s life is truly endangered by the condition. The technique has greatly improved since the One Flew over the Cuckoo’s Nest days: electrode leads are applied to only one brain hemisphere to minimize cognitive changes, and both a general anesthetic and muscle relaxant are used to confine the induced seizure to the brain. In my former geriatrics practice, a few of my patients with very severe depression were treated with this approach over the years (none of them had significant dementia at the time of treatment and all received fewer than a half dozen treatments). Afterward, they showed modest improvement in their mood, but they also showed some lasting cognitive changes, contradicting the claims of some that there are “no” cognitive side effects to ECT.

Now once again, a large study will apply this treatment to people living with dementia. Although I am not privy to all the methodology in the proposed study, the following remarks should fall close enough to the actual design to support my concerns.

Let’s start with the ethical problems.The people who are transferred to such inpatient centers are invariably moved there at someone else’s direction. At this point in the lives of people with Alzheimer‘s disease, a multitude of day-to-day decisions have already been—rightly or wrongly—assumed by family members or other representatives. While advanced directive protocols are necessary in cases of decreased capacity, they were created to protect the person and respect her wishes. This leads one to wonder how such an invasive procedure as ECT, employed without the individual’s consent, has come to be construed as an example of “protection” of that person’s health, well-being, and previously stated wishes.

My recent foray into formal research has taught me that even the simple collection of anonymous data about, for example, antipsychotic use needs to have ethics committee approval. One wonders how surrogate consent to use ECT (usually given a dozen or more times in these studies) ever passed an ethics board.

Actually, one doesn’t wonder. This study is a perfect example of what happens when an overly medicalized view that sees all distress as “behavioral and psychological symptoms of dementia” (or “BPSD”) tries to find “solutions.” My experience and the work of many others has shown that most distress is the result of other underlying factors—relational, environmental, unmet needs, etc.—and that while dementia may affect one’s communication or coping skills, it is rarely the root cause of their distress. With any other disease process, the logical approach would be to address the root cause(s), not simply treat the individual with pills or shocks.

But our BPSD view causes us to medicalize and pathologize the expressions of people living with dementia; in fact, we hold them to a higher emotional standard than we would ourselves. We get angry, sad, frustrated or anxious, but people with dementia have “behaviors.” We go for a walk, do our “steps,” or get bored and leave, but people with dementia “wander” or ”exit-seek.” We get restless when forced to follow other’s rhythms and schedules, but people with dementia “sundown.” We go to Costco and shop in bulk, but people with dementia “hoard.” And we don’t like being locked up, bossed around, or touched by strangers, but people with dementia get “agitated” and aggressive.” If people are upset because they are being undressed by strangers, are locked inside a living area, have no meaningful engagement in their days, or are simply expressing the range of emotions that we all display at times, how does ECT improve that?

These issues also raise the larger question of how the human rights of people living with dementia are potentially being trampled by such a procedure. A 2018 judge’s decision to allow a Massachusetts school to use ECT for students with severe learning disabilities was harshly condemned by human rights advocates (, showing that the rights of those with other disabilities continue to be advanced far ahead of those living with dementia. Viewed from this lens, one could say that using sedating medications or ECT violates one of our most basic human rights in any care environment—the right to have one’s concerns heard and understood, and necessary accommodations made. In fact, this is a central tenet of resident rights regulations throughout aged care.

On the efficacy side, having read and extensively critiqued all of the antipsychotic studies over the years, I can predict that this study will follow most of the same flawed premises: (1) it will view the distress as an internal “symptom,” ignoring everything else surrounding the person, (2) it will measure its primary outcomes via decreased scores on reductionistic, deficit-based scales of “agitation” rather than looking for positive outcomes, such as measures of well-being or improved engagement, (3) it will minimize the reporting of the side effects of the treatment, or paint them as an acceptable tradeoff to obtain a sense of emotional calm (more on this later), and (4) most important, it is designed to compare ECT against “usual care,” meaning that no innovative approaches to improving people’s well-being or transforming the care environment will be applied—in effect, it will be like testing the effect of Bacardi 151 rum on relieving extreme thirst, without trying to offer water.

On the safety side, a lot has been said about how much safer and well-tolerated ECT has become in recent years. However, people with severe depression have brains that are quite structurally intact. How does that degree of safety translate to the people with “moderate to severe Alzheimer’s disease” who will be treated in this study, in whom there has been extensive neuronal damage, particularly to memory centers that are susceptible to the effects of repeated ECT? In 2016, citing the potential for severe side effects, the FDA moved to enact new restrictions on the use of ECT in depression, including holding a detailed risk-benefit discussion with each patient (

Next, let’s look at patient selection. Who is referring the people to the geri-psych units that will be enrolled? Usually it is nursing homes, and if so, what is their current knowledge of innovative, compassionate support for people living with dementia? Are they high-performing homes who are only referring people with truly refractory episodes of distress? Or are they homes who are poorly educated and equipped to offer more individualized approaches? Do they have consistent staffing assignments, or are they rotating people regularly who provide personal care (there is much evidence that such rotation is a major causative factor for distress and resistance)? Are many antipsychotic drugs being used concomitantly with the ECT? And does the study also include community-based referrals from overwhelmed family members who have not been taught how to best support their loved ones, or who have inadequate resources to do so?

My guess is that they will take their referrals as per their usual “agitation” protocols, and will not try to delve deeply into these factors. Here is why that is critically important:

Dr. Angela Norman and her team from the Arkansas Health Care Foundation have launched a large initiative to employ the Well-Being Approach to Distress that I teach in their state’s long-term care communities. After an initial, highly successful pilot with four homes, Dr. Norman brought the approach to the 25 highest-prescribing homes in Arkansas. As of this writing, those homes have cut their antipsychotic use by an average of 49% in only 6 months.

Furthermore, Dr. Norman’s team has been mentoring a group of nearly 100 Arkansas nursing homes with this approach. Over the same period, they have significantly reduced the number of people transferred to hospitals and geri-psych units from those homes.

Consider the implications of this: These people whose well-being has been improved, and who have been kept off of drugs and out of inpatient treatment in Arkansas are the same kinds of people to whom the NIA study will be applying multiple courses of ECT in Michigan, Minnesota, New York, Massachusetts, and Georgia. This suggests that many, if not most of the study subjects will be getting an invasive, potentially harmful treatment that would have been unnecessary if they lived in a place that offered more innovative approaches.

Unfortunately, my own past searches for research grants have found that the NIA rarely funds the kind of work that’s being done in Arkansas. I can only wonder what my late friend and mentor, Dr. T. Franklin Williams (who directed the NIA from 1983-1991) would say of his former organization funding an ECT study such as this. Having had many discussions on dementia with Frank and his wife Carter over the years, I think I know.

Let’s wrap up by getting back to the comment above about “acceptable side effects” to produce calm—we’ve also been down that road with antipsychotic drugs. While I have not yet heard this particular research team say it, there is a common attitude around such interventions that you probably have heard: “Even if there are lasting cognitive effects (from ECT, or antipsychotics), the person is suffering; and having her shift to a more advanced stage where she is less distressed could actually be a blessing.” I have heard this said many times from people who truly feel they are being compassionate—I have surely thought it myself in the distant past. But this is a dangerous way to think.

I’ve discussed some of the many external factors that can serve as a root cause of distress, and ways in which those underlying factors can be successfully addressed without drugs or ECT. So, is a decrease in cognitive function really an acceptable tradeoff? How compassionate is it to render a person no longer capable of expressing those human emotions and unmet needs?

Kate Swaffer, who has lived with a diagnosis of dementia for over a decade, told a 2018 Australian parliamentary advisory group, “Future generations are going to look back at the way we have mistreated our Elders in aged care . . .and will hang their heads in shame at the way we as a ‘civilised society’ have treated them.”

I believe dementia is the greatest shame of modern medicine; not because there have been no significant advances in treatment, but because—from restraints, to locked units, to antipsychotics, to ECT—we have lost our recognition of the humanity of those living with the diagnosis. And there are few signs that we are making much headway in this regard.

I am left to the same conclusion I reached in my 2011 post: ECT for dementia still sounds to me like the “21st Century Lobotomy.”