Home Depression Understanding Treatment-Resistant Depression in Later Life

Understanding Treatment-Resistant Depression in Later Life


In this video, Randall Espinoza, MD, MPH, discusses the epidemiology, impact, and management of treatment-resistant depression in older adults. He recently spoke on the topics at the virtual 2021 annual meeting of the American Association for Geriatric Psychiatry.

Dr. Espinoza is Professor of Clinical Psychiatry and the Muriel Harris Endowed Chair of Geriatric Psychiatry in the Department of Psychiatry and Biobehavioral Sciences at the Geffen School of Medicine at UCLA, Los Angeles, California. He is also Director of the UCLA Geriatric Psychiatry Fellowship Training Program, Medical Director of the ECT and Interventional Psychiatry Program at the Resnick Neuropsychiatric Hospital at UCLA, and Associate Director of the UCLA Longevity Center.

Read the transcript of the video, which has been edited for clarity:

Hello, my name is Randall Espinoza, and I am a professor of clinical psychiatry and the Muriel Harris Chair of Geriatric Psychiatry at the Geffen School of Medicine at UCLA. Today, I will be talking about treatment‑resistant depression in later life, or TRD-LL.

In this talk, I will provide an overview of TRD and briefly point out issues relevant to older adults. Next, I will highlight the impact of TRD in later life, and then I will review the epidemiology of TRD in later life, discussing differences between adult and older adult populations and discussing some challenges to the diagnosis of and risk factors for TRD in later life. Finally, I will share some thoughts on implications for management.

What Is Treatment-Resistant Depression? (0:56)

To begin, let’s spend a few minutes briefly reviewing the concept of treatment‑resistant depression and related issues. Definitions of TRD are wide‑ranging and there is no operational consensus on when or how a person might be identified as suffering from TRD. Various staging methods have been proposed.  Some are based upon the number of treatment modalities attempted or, more specifically, on the number of medication trials of antidepressants used.

However, recognize that medication dosage and duration criteria have also been debated. Importantly, none are explicitly validated in older adult populations, so it is uncertain how well these methods may actually apply.

Further, other questions remain unanswered, such as how long to wait for a response and what level or extent of response is needed. While for younger adults, the typical adequate time-frame is listed as 4 to 6 weeks, for older adults, the time-frame is often extended to between 12 to 16 weeks. However, this variability may not uniformly apply across all older age groups.

In late‑life depression, there also remain questions about the role and relationship among illness course, symptom dimensions, and comorbidity. Studies show that delays in treatment initiation may prolong the episode. Unfortunately, delays in treatment are particularly well‑documented in late‑life depression samples.

It is possible that TRD itself is a depressive subtype and that certain dimensions of illness respond variably to particular interventions, and therefore, current first‑step treatments may not provide uniform benefits.

Given the challenges in definition and identification of TRD, a new term called “Difficult‑to‑Treat Depression” is emerging.  This concept expands the notion of treatment, moving away from an exclusive focus on acute outcomes, which may be arbitrarily defined, to a more integrated approach that includes not only assessment of symptoms over the short term but appreciates the often chronic, recurrent, and multidimensional nature of this illness in older adults.

Pseudoresistant Depression (3:26)

Importantly, not all depressions are truly refractory. Instead, lack of response may be obscured or impacted by other aspects. This issue is termed “pseudoresistant depression.”

Contributions to pseudo-resistance may be a misdiagnosis of the primary mood disorder. For the older adult, it’s important to consider bipolar depression, especially in the presence of another first‑degree relative with bipolar illness, the development of irritability or agitation when taking antidepressants, or evidence of mixed states.

Psychotic depression can be also hard to detect if psychosis is not florid. Agitated catatonia may present a similar challenge. Depression arising from another cause, such as medication, substances, or a general medical condition may also influence depressive presentation.

Depression arising from a difficult or unexpected life situation, where coping skills or resources are overwhelmed, may also confound diagnosis, as can depression arising close to bereavement or complicating grief.

For older adults, subsyndromal and persistent depressive disorder can be very difficult to separate from a major depressive episode. Determining where one ends and the other begins can be a fraught exercise.

Another contributor to pseudo-resistance involves issues of treatment management. For medications in particular, older adults may have been under-dosed out of concerns for side effects, issues of medical comorbidity, or tolerance.  Older adults may have discontinued treatment too early due to lack of understanding of the time needed to achieve full benefit or to the development of discouragement. Older adults may have not adhered to the treatment plan for a variety of other reasons. Finally, there may be concerns arising out of pharmacokinetics or pharmacogenetic matters.

Impact of TRD-LL (5:34)

The impact of treatment‑resistant depression in general is well‑recognized. Like in major depressive disorder, TRD is associated with increases in morbidity and mortality, both medical and psychiatric.  Additionally, there are increases in health costs and utilization. Further, treatment‑resistant depression is associated with decreases in quality of life and productivity in the workplace, home, and other social settings.  Recent studies in the Medicare population support very similar findings of TRD in later life.

We also know that for many patients, including older patients, major depression can be a chronic and/or recurrent illness. Twenty-five to 40 percent may experience a recurrence within 2 years of the index episode.  A substantial portion may experience recurrence after 5 years. A substantial portion of patients who experience one episode of depression will have chronic depression, that is, continuous depression lasting more than a year.

For treatment‑resistant depression in later life, a very similar picture appears. In fact, older adults, for a variety of reasons, may be more prone to develop a chronic, unremitting course or to have more fragile recoveries.

Epidemiology of TRD in Later Life (6:58)

Perhaps owing to the existence of debate on the concept of TRD and on differing definitions, the prevalence of TRD varies widely from about 6 percent to over 40 percent across psychiatric populations and settings.

However, the precise prevalence of TRD in later life is unknown, even though many factors contributing to development of TRD may be more common in older adult populations.

Despite the availability of numerous antidepressant medications, validated psychotherapies, and other neuromodulation approaches for the treatment of major depressive disorder, studies of each modality in treatment‑resistant depression in later life are few.

Therefore, the exact epidemiology of TRD in later life is unclear, owing to a variety of reasons. Depending on the study definition or patient population in query, estimates of TRD prevalence may range broadly from 5 percent to up to 40 percent.

Data from various drug trials, such as STAR*D, VAST, and CATIE place the estimate of lack of response to pharmacotherapy (or antidepressant medication) at between 10 to 30 percent.

Recent studies in the Medicare‑covered population, which includes, predominantly but not exclusively, older adults, place estimates of TRD prevalence at around 10 percent.

However, the true prevalence of TRD in later life has yet to be fully or accurately determined. Indeed, there are multiple factors that may influence the presentation and detection of TRD in older populations.

Possible differences for estimates include different presentations of late‑life depression, differences in illness onset, differences in the illness course and duration, exposures, as well as other treatment issues.

Predictors of Nonresponse to Treatment of TRD-LL (8:56)

Understanding predictors of nonresponse to treatment is very important but challenging. Predictors of nonresponse to a single antidepressant treatment and predictors of resistance to multiple antidepressant agents may differ.

Many of the predictors to nonresponse are considered to additionally predict overall resistance. Further, the intensity and features of the current episode of depression may also be predictive of nonresponse.

Looking at some specific factors, these may be particularly relevant to older adult populations. Both medical and psychiatric comorbidity appear to predict nonresponse. Among psychiatric issues are anxiety disorders, substance abuse disorders, cognitive disorders, and personality disorders.

The literature is replete with the bidirectional relationship between medical conditions and the risk of psychiatric illness, including depression. Further, the presence of any comorbidity increases the risk of delay in accessing or starting treatment and may, in the long run, contribute to chronicity.

In older adults, it is often difficult to ascertain precisely which other first‑degree relatives may have suffered from a psychiatric disorder. A positive family history would appear to confer a greater risk of nonresponse.

Differences in age of onset also appear to play a role in treatment responsiveness, with early onset illness occurring in childhood or adolescence, as well as very late onset, being associated with the risk of development of TRD.

Further, depression subtypes, if overlooked, correlate with the risk of lack of response, at least to monotherapy. Among these subtypes are bipolar depression, psychotic depression, and possibly substance‑induced mood disorder.

Implications for Management of TRD-LL (11:02)

Finally, it is helpful to understand what are some implications for management of TRD in later life. To begin, there are issues related to assessment. We want to consider these 3 areas: a thorough clinical history, evaluation of illness course, and examination of treatment history.

After completion of this assessment, treatment choices are predicated on setting up a plan, and then choosing what to do next.

Thank you very much.


“Beyond ECT – Neuromodulation and Interventional Psychiatry for Treatment Resistant Depression in Later Life.” Presented at the American Association for Geriatric Psychiatry 2021 Annual Meeting: Virtual; March 16, 2021.


This site uses Akismet to reduce spam. Learn how your comment data is processed.