U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Gaynes BN, Asher G, Gartlehner G, et al. Definition of Treatment-Resistant Depression in the Medicare Population [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2018 Feb 9.

Cover of Definition of Treatment-Resistant Depression in the Medicare Population

Definition of Treatment-Resistant Depression in the Medicare Population [Internet].

Show details

Background and Objectives

Clinical and Epidemiological Issues

Depressive episodes can be seen in patients with either major depressive disorder (MDD) or bipolar disorder. In 2015, 6.6 percent of adults in the United States experienced a depressive episode in the past year.1 The bulk of these episodes are part of MDD, experienced by more than 13 million U.S. residents each year.2 Of these individuals, one-half seek help for this condition; one in five of those seeking help receive adequate acute-phase treatment.3 Even for patients receiving adequate treatment, only 30 percent (i.e., 3% of patients with MDD) reach the treatment goal of full recovery or remission.4

The remaining 70 percent of MDD patients will either respond without remission (about 20%) or not respond at all (50%).4 Patients whose depressive disorder does not respond satisfactorily to adequate treatment clearly have harder-to-treat depression,5 which is generally (albeit not uniformly) referred to as treatment-resistant depression (TRD). Although often broadly defined this way, TRD is a complex phenomenon that is influenced by heterogeneity in depressive subtypes, psychiatric comorbidity, and coexisting medical illnesses.6 It poses a common, challenging presentation to psychiatric and primary care clinicians,7 one in which decision-making is guided by both available data and patient preference.

Although TRD is most commonly associated with MDD, treatment-resistant depressive episodes are also seen in the depressed phase of bipolar disorder. Bipolar disorder affects 2.6 percent of the U.S. adult population each year.54 Much like MDD, bipolar depression can be treatment resistant, and TRD can sometimes result from a failure to properly diagnose a patient presenting with depression as in fact suffering from bipolar disorder. More than 30 percent of those suffering from bipolar disorder and receiving treatment do not experience sustained remission of depressive symptoms.8 Even among those who do achieve recovery for lengthy periods, depressive relapses are common; more than 20 percent of individuals with successfully treated bipolar depression will experience a depressive relapse within a year.8

TRD has substantial effects on patients and major impacts on families, communities, and society at large, most of which have been described for MDD patients. TRD represents the highest direct and indirect medical costs among those with MDD.9 These costs increase with the severity of TRD.55 Individuals with TRD are twice as likely to be hospitalized; the cost of this hospitalization is more than six times the mean total cost for depressed patients who are not treatment resistant.10 TRD can nearly double both direct and indirect 2-year employer medical expenditures relative to expenditures for patients whose MDD responds to treatment ($35,500 for those with TRD and $18,600 for those with MDD).11

TRD is especially relevant for Medicare beneficiaries, for whom unsuccessfully treated depression has harmful sequelae. Mood disorders, which consist primarily of MDD and bipolar disorder, are the second leading cause of disability in Medicare patients under the age of 65.12 Furthermore, depression in the elderly is more associated with suicide than at any other age;13 although adults 65 or older make up 12 percent of the population, they constitute 16 percent of all suicide deaths.14 Indeed, the decrease in average life expectancy for those with depressive illness, including Medicare beneficiaries, is 7 to 11 years, similar to that in elderly smokers.15

Finally, depression is a major predictor of the onset of stroke, diabetes, and heart disease.16 Being depressed increases patients’ risk of developing coronary heart disease,16 and it raises the risk of dying from a heart attack nearly threefold.17

Rationale for Review

No universally accepted operational definition exists for TRD. Criteria for TRD have been variably defined in clinical research and practice,18 reflecting numerous difficulties and controversies about its definition. These definitional dilemmas limit the ability of systematic reviewers or other experts to synthesize information and generalize the findings of many TRD studies to the array of patient populations encountered in daily practice. In addition, they restrict the application of clinical research findings to clinical practice, including community populations of TRD patients.

Moreover, these varying conceptualizations of TRD have made translating research findings or systematic reviews into clinical practice guidelines challenging and inconsistent. Treatment guidelines reflect this variability: their definitions of TRD differ, agreement on what constitutes prior treatment adequacy is lacking, and recommended “next step” interventions can diverge.19-21,23,56

The purpose of this report is to examine comprehensively the study design issues affecting both outcomes and bias in studies of TRD; we are not in this review determining outcomes associated with specific treatments of TRD. Our specific aims are two-fold: (1) to inform future discussions and decisions about how to define TRD and the important outcomes measured in research studies, and (2) to clarify how trials or observational studies might best be designed and conducted to guide clinical practice and health policy.

Key Questions

Narrative Review Questions

We first performed a narrative review of relevant literature to address Key Questions (KQs) 1 through 5. The narrative review questions were more qualitative and less structured than a systematic review; therefore, we used a more contextual narrative review approach. Our work is based on searches of consensus statements, guidelines, materials from the U.S. Food and Drug Administration, the U.S. National Institutes of Health (including the National Institute of Mental Health), and the U.S. Substance Abuse and Mental Health Services Administration; systematic reviews; and a review of UpToDate®, an evidence-based, peer-reviewed clinical information source. In addition, we used information from the Medicare Evidence Development & Coverage Advisory Committee panel meeting on April 27, 2016,57 to augment our reporting on TRD definitions, study design issues, and the related topics.

The specific narrative KQs are:

  1. What definitions of TRD are found in this literature? What consensus, if any, exists about the best definition(s) for this condition?
  2. What methods do investigators use to diagnose this condition in clinical research? What consensus, if any, exists about the best measure(s) to use? Does the setting of the medical visit influence the choices that investigators make about the diagnostic tool they use?
  3. What measures have been developed to determine the success and failure of treatment in clinical research studies of TRD?
    1. What consensus, if any, exists about the best measure(s) to investigate treatments for TRD? What are the main points of agreement about such measures?
    2. Are these measures physician reported or patient reported? What are the psychometric properties of these measures? Is the minimum significant clinical difference defined for these measures?
    3. Compare and contrast these measures in how they describe:
      1. Change in depression scores as measured by depression scales
      2. Change in depressive symptomatology (e.g., sleep disorders, fatigue, weight change, cognition)
      3. Change in measures of anhedonia
      4. Change in measures of functional capacity (e.g., physical functioning, ability to care for self)
      5. Change in measures of quality of life
      6. Change in measures of suicide ideation
      7. Change in suicide attempts
      8. Other
  4. What types of research designs are used to study TRD?
    1. What consensus, if any, exists about the type of study design that best minimizes bias and the placebo effect in this field?
    2. If no consensus exists about study designs to accomplish these goals, what are the trends in study designs for assessing interventions for TRD? Do these trends reflect long-lasting (e.g., traditional) designs or short-lived, evolving, or newly emerging designs?
    3. What consensus, if any, exists about the appropriate length of a trial?
  5. What are the risk factors for TRD?

Systematic Review Questions

From a systematic literature search for individual studies on TRD, we address KQs 6 through 11 with their subquestions as listed below.

6.

What are the inclusion criteria for patients in these studies? Specify at least the factors listed below.

  1. Patient characteristics:
    1. Age
    2. Type of depressive episode (unipolar, bipolar, psychotic, atypical, other)
    3. Number of depression relapses and time to relapse
    4. Psychiatric comorbidities
    5. Medical comorbidities (e.g., diabetes, cardiac disease, renal disease, dementia, and other cognitive abnormalities)
    6. Suicidal ideation
    7. Suicide attempts
    8. Duration of symptoms
    9. Screening tools used to make the diagnosis
    10. Diagnostic tools to confirm the diagnosis
  2. Prior treatments:
    1. The number, duration, dosage, or classes of antidepressants attempted for each trial of therapy
    2. The number of failed trials of adequate therapy
    3. The number of prior treatment trials that patients did not tolerate The use of augmentation and combination pharmacological therapies for each attempted treatment trial
    4. The use of electroconvulsive therapy
    5. The use of psychotherapy
  3. Diagnostic characteristics
    1. The use of structured versus unstructured diagnostic assessments
    2. Scores on standardized and validated depression rating instruments
    3. Setting in which the diagnosis was made (i.e., primary care, generalized psychiatric setting, specialty psychiatric setting, other)

7.

How do these inclusion criteria compare or contrast with the definition(s) of TRD noted in the narrative questions?

8.

What were primary characteristics of included studies?

  1. What was the main design of each included study (e.g., randomized controlled trial with blinding; interrupted time series; use of placebo, wait-list, or sham procedure)?
  2. Were run-in or wash-out periods (or both) used in included studies? If so, how long were they?
  3. How long was each included study?

9.

How were included studies designed to account for the risk factors for TRD (see (Narrative Review KQ 5)? If the following characteristics are not noted above as risk factors, how did included studies account for at least the following: age, sex, race, socioeconomic status, duration of symptoms, disease severity, coexisting medical and psychiatric conditions, and placebo effect?

10.

What are relationships between risk factors and various results of included studies?

  1. Using regression analysis or other statistical techniques, determine whether the risk factors for Narrative Review KQ 5 and Systematic Review KQ 9 can be correlated with study results (i.e., the magnitude of treatment effects)?
  2. What is the influence of placebo response on the magnitude of treatment effects for different types of interventions?
  3. Does study duration moderate the influence of placebo response?

11.

What variables or information did included studies report? Specifically:

  1. What measures are used to define end points in these TRD trials?
  2. In addition to the measures noted for Narrative Review KQ 3, did these studies record:
    1. Adherence to treatment
    2. Attrition from care
    3. Changes in patient-selected factors of importance (i.e., outcome measures identified by patient as important)
    4. Changes in employment or disability status
    5. Changes in use of medical resources (e.g., hospitalizations, emergency room or physician visits)
    6. Time to relapse

Organization of This Report

Apart from this introduction, this Technology Assessment report has the following chapters. Next is the Methods chapter, which covers all the steps used to address KQs 1 through 11. Following that, for ease of presentation and readability, we have split Results into two chapters; the first concerns the narrative review KQs and the second addresses the systematic review KQs. We conclude with the Discussion chapter, which addresses findings, strengths, and limitations of the project and the evidence base, applicability, and similar topics for all 11 KQs. The appendices document the following methods or data: A, Search Strategy; B, List of Excluded Studies; C, Evidence Tables; and D, Risk of Bias Ratings.

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.1M)

In this Page

Other titles in this collection

Recent Activity

ClearTurn OffTurn On

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...