Depression/Anxiety: Dimensional Treatment Decisions

Treatment decisions lie on a continuum—befitting a dimensional approach to diagnosis.


(This is the second of a 2-part series. The first article discussed how categorization bias favors the DSM, and how to avoid the consequent narrowing of treatment options.—Ed.)

In part 1 of this series, I discussed cognitive biases related to inaccuracies in the DSM and some solutions for improving the process of shared treatment decision-making with patients. I noted that categorical DSM diagnosis and dimensional spectrum-based diagnosis can be used simultaneously, depending on which is most useful.

Sometimes categorical diagnosis is better—eg, for legal purposes or for communication with other medical colleagues. Dimensional assessment more closely matches mood phenomena (see part 1). But is dimensional assessment difficult for clinicians to use? Are binary diagnoses necessary for binary treatment decisions?

No. Treatment decisions are often not binary, but instead lie on a continuum themselves. Here are 3 treatment-decision spectra:

1. Degree of certainty regarding a patient’s position on the mood spectrum

2. Number of previous treatments, and response thereto

3. Acuity versus chronicity and impact on long-term risk preference

1. Certainty

Take Ms Smith, who presented in Part 1 with depression, anxiety, and a history of trauma. After consideration of thyroid, other medical issues, and substance use, the categorical differential for a mixture of depression and anxiety symptoms includes posttraumatic stress disorder (PTSD); major depression with mixed features (or comorbid generalized anxiety disorder); bipolar II; and borderline personality disorder. These diagnoses can be extremely difficult to differentiate, as suggested by the overlapping symptoms shown in the Table.1

What is a clinician to do? Likely what many readers have been doing for years: Choose treatment options by degree of certainty. Let’s examine this process using the hypothetical Ms Smith as an example.

If Ms Smith describes intrusive reexperience and avoidance, and if a PTSD-specific psychotherapy (eg, cognitive processing therapy or eye movement desensitization and reprocessing) is available to her, then your PAR (procedures, alternatives, and risks) discussion would emphasize these specialized therapies. But if PTSD is not clearly the basis of her depression and anxiety (her trauma might be predisposing, but not causative), then antidepressants and more widely available therapies would become more prominent in your PAR. As your certainty about the role of trauma in her presentation shifts, so do your treatment recommendations: not from black to white, but gradually from a strong preference for one treatment option toward another.

If Ms Smith’s history includes cutting behavior, suicidal ideation, and dramatic interpersonal chaos, dialectic behavior therapy is an obvious treatment option. If Ms Smith meets full DSM criteria for borderline personality disorder, then antidepressants are unlikely to be of significant benefit.2 But to the extent that borderlinity is not certain, an antidepressant becomes a more prominent option and psychotherapy options broaden; perhaps cognitive behavioral therapy (CBT) would have more value than dialectical behavior therapy (DBT). Your shared decision-making would be strongly influenced by how Ms Smith reacts to your description of antidepressants’ risk/benefit ratios, and the 2 psychotherapies’ relative strengths—a reaction that itself lies on a continuum from strongly favoring one treatment to strongly favoring another.

If borderlinity is not prominent, then the decision-making focus shifts to Ms Smith’s depression and anxiety symptoms. Your PAR discussion would include a comparison of psychotherapies:

CBT for depression, or a transdiagnostic CBT3 that can address her anxiety as well. The strength of her anxiety symptoms relative to her depression symptoms, which also lies on a continuum, would determine your emphasis.

Finally, if a history of hypomania/mania is obvious, then one would target Ms Smith’s depression using a mood stabilizer with antidepressant effects (MSAE): interpersonal and social rhythm therapy (IP/SRT)4 and bipolar-specific CBT5; and/or lamotrigine, lithium, quetiapine, olanzapine, or lurasidone. But if her bipolarity is uncertain, then antidepressants and a more standard psychotherapy—CBT or interpersonal therapy—become options. The choice would depend on the question, “how bipolar is she?”—just the question advocated years ago by the STEP-BD research program’s principal investigator.6

Is this process really any different from just making a DSM diagnosis and recommending the appropriate treatments? Yes. At minimum, one approaches one’s patients differently—not saying, “this is what you have, and this is the treatment for it,” Rather, one is in a humbler position vis-à-vis the patient, and vis-à-vis what we really know about psychiatric diagnoses. To Ms Smith, for example, you might say: “We cannot know for sure where on the mood spectrum you are, or how much your awful experiences account for the symptoms you have. But the treatment options are much the same: psychotherapy (not Freud, not about your parents, no couch), and/or a medication for the mood part. We can start with either or both. May I tell you more?”

In shared decision-making, one would then contrast the forms of psychotherapy to be considered and the medication options. As you know from experience, this choice becomes particularly complex around 1 variable: bipolarity. Will you suggest an antidepressant or an MSAE? This thorny decision is strongly influenced by a second continuum: your patient’s previous treatments and responses.

2. Previous Treatments

Many patients who come to see you with depression and anxiety will have already seen their primary care provider and been given a prescription for an antidepressant, or several, and sometimes even an atypical antipsychotic. Or you may have tried these medications and now must follow the old dictum: “When your treatment’s not working, reconsider your diagnosis.” Your next recommendation depends on the number of previous treatments and response thereto.

Consider Ms Smith again. If she is “treatment naïve,” many symptoms or markers of bipolarity would be needed to justify using an MSAE instead of an antidepressant. One might even require that she meet DSM criteria for bipolar II. But if she has had 2 antidepressants without sustained benefit, an MSAE is more worthy of consideration. And if she has had 3 or more antidepressants, and perhaps a reliable course of CBT, and is still coming to you for help, then you might consider an MSAE as one of her best options even if she has very few bipolar markers. This continuum is shown schematically in Figure 1.

(Lamotrigine is a first-line option for bipolar depression in 2 guidelines, from the Royal Australian and New Zealand College of Psychiatrists,7 and from the Canadian Network for Mood and Anxiety Treatments and the International Society for Bipolar Disorders (CANMAT-ISBD).8 Lithium is also first-line in these guidelines. The logic behind the positioning of these 2 medications is summarized on page 18 of the latter guidelines.)

3. Acuity, Severity, and Chronicity

A third continuum: If Ms Smith has been experiencing depression (perhaps in episodes) for years, she is more likely to accept that her treatment may also need to go on for a long time, for the prevention of recurrence. For many patients, this long-view perspective increases their preference for treatments with very low risk and very few side effects.

Conversely, a very acute or very severe episode may prompt a wish for a treatment with the strongest evidence for benefit, even if side effects or risks or costs are high. This continuum is shown schematically in Figure 2.

Concluding Thoughts

Dimensional diagnosis is not inherently difficult for clinicians, because clinicians are not making binary treatment decisions. Rather, their treatment recommendations lie on at least 3 continua. Categorical diagnosis is not required to guide treatment.

In this series, I have used the example of a patient with depression, anxiety, and trauma to examine 1) the cognitive biases that sustain the DSM and 2) the dimensional nature of treatment decision-making. I chose a mood disorder case to illustrate these ideas because that is the field I know best. But I believe the logic presented here applies across almost all psychiatric illnesses. Most are better discerned using a dimensional perspective as well as categorical one, and treatment decisions are better understood as continua than as binary choices.

We need not abandon the DSM to move forward—we simply need to add a dimensional view of psychiatric illness alongside our categorical system. Use the one that most helps you help your patient.

Dr Phelps is research editor at the Psychopharmacology Institute, emeritus faculty at Samaritan Mental Health in Corvallis, Oregon, and founder of He is the bipolar disorder section editor for Psychiatric TimesTM and the author of A Spectrum Approach to Mood Disorders for clinicians and Bipolar, Not So Much for patients and their families.


1. Perugi G, Angst J, Azorin JM, et al. Mixed features in patients with a major depressive episode: the BRIDGE-II-MIX study. J Clin Psychiatry. 2015;76(3).

2. Lieb K, Völlm B, Rücker G, et al. Pharmacotherapy for borderline personality disorder: Cochrane systematic review of randomised trials. Br J Psychiatry. 2010;196(1):4-12.

3. Barlow DH, Farchione TJ, Bullis JR, et al. The unified protocol for transdiagnostic treatment of emotional disorders compared with diagnosis-specific protocols for anxiety disorders: a randomized clinical trial. JAMA. 2017;74(9):875-884.

4. Interpersonal and social rhythm therapy. IPSRT. Accessed January 26, 2022.

5. Aiken C. Depression and Bipolar Workbook: 30 Ways to Lift Your Mood & Strengthen the Brain. PESI Publishing & Media; 2020.

6. Saenger E. The Bipolarity Index as a tool for assessment and creating rapport: an expert interview with Gary Sachs, MD. Medscape Psychiatry & Mental Health. 2005;10(1).

7. Malhi GS, Bell E, Boyce P, et al. The 2020 Royal Australian and New Zealand College of psychiatrists clinical practice guidelines for mood disorders: Bipolar disorder summary. Bipolar Disord. 2020;22(8):805-821.

8. Yatham LN, Kennedy SH, Parikh SV, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) 2018 guidelines for the management of patients with bipolar disorder. Bipolar Disord. 2018;20(2):97-170.