Mental health treatment sits at an unusual intersection of science, stigma, economics, and deeply personal experience. The available options — therapy, medication, self-help, and combinations thereof — have vastly different levels of evidence, accessibility, cost, and appropriate application. Yet the way they are typically presented to people in distress involves more folk wisdom than data. Medication gets prescribed by primary care physicians without discussion of therapy alternatives. Therapy gets recommended without acknowledgment of the evidence for medication in severe cases. Self-help books promise transformation without noting their limited evidence base for serious conditions. The person trying to decide what to do rarely has a clear map.

The research on each approach is more developed and more specific than most people realize. Cognitive Behavioral Therapy (CBT) has dozens of randomized controlled trials across multiple conditions. SSRIs have some of the largest evidence bases of any medication class. Even structured self-help has been studied in clinical contexts. What the evidence shows is not that one approach wins universally, but that the optimal approach depends on the condition, severity, individual circumstances, and — critically — whether treatment approaches are combined.

This article presents the evidence clearly for each approach, explains when and why combination treatment typically outperforms either modality alone, and provides practical guidance for making treatment decisions at different severity levels and life circumstances.

"The honest answer to 'which treatment works best' is: it depends on what you have, how severe it is, and what resources you can access — but combination treatment consistently outperforms either medication or therapy alone for moderate to severe presentations." — Pim Cuijpers, Annual Review of Clinical Psychology (2019)

---

Key Definitions

Cognitive Behavioral Therapy (CBT) — A structured, evidence-based form of therapy that targets the relationship between thoughts, behaviors, and feelings. It is time-limited (typically 12-20 sessions), present-focused, and involves homework assignments and skill practice between sessions. CBT has the most extensive evidence base of any psychotherapy approach across anxiety disorders, depression, OCD, PTSD, and eating disorders.

Antidepressants (SSRIs/SNRIs) — Selective serotonin reuptake inhibitors and serotonin-norepinephrine reuptake inhibitors are the most commonly prescribed medications for depression and anxiety. They work by modulating neurotransmitter availability. Common examples include fluoxetine (Prozac), sertraline (Zoloft), escitalopram (Lexapro), and venlafaxine (Effexor). Effects typically develop over 2-6 weeks.

Guided self-help — A structured self-help program — typically based on CBT principles, delivered through a workbook or app — with light-touch professional support (brief weekly or biweekly check-ins). NICE guidelines recommend this as a first-line treatment for mild depression and anxiety.

Effect size — A standardized measure of how large a treatment's effect is relative to control conditions. In mental health research, an effect size of 0.2 is small, 0.5 medium, and 0.8+ large. Understanding effect sizes allows comparison across different treatments and conditions.

Relapse rate — The proportion of patients who recover from an episode of depression or anxiety and subsequently experience another episode. Relapse is common in depression — approximately 50% of people who have one depressive episode will have another. Treatment approaches vary substantially in their impact on relapse risk.

Stepped-care model — The clinical framework in which the least intensive, least costly effective treatment is offered first, with escalation to more intensive interventions if lower-level treatment fails. NICE guidelines and NHS IAPT services are organized around this model.

Behavioral Activation (BA) — A component of CBT for depression that focuses on increasing engagement with rewarding activities and reducing avoidance. BA alone has been shown to be as effective as full CBT for depression in some trials, suggesting that behavioral change is a powerful lever independent of cognitive restructuring.

Exposure and Response Prevention (ERP) — The CBT-based treatment specifically developed for OCD. Involves deliberate, graduated exposure to feared triggers with prevention of the compulsive response, extinguishing the anxiety-compulsion cycle. ERP is the first-line treatment for OCD and substantially more effective than medication alone for many patients.

---

Treatment Comparison at a Glance

The following table summarizes the key characteristics of each approach across dimensions most relevant to treatment decision-making.

Dimension	CBT / Therapy	Antidepressants (SSRIs)	Guided Self-Help
Evidence strength (depression)	High (effect size 0.6-0.8)	High (Cipriani 2018 meta-analysis)	Moderate (mild-moderate only)
Evidence strength (anxiety)	High (often > depression)	High (first-line for GAD, social anxiety)	Moderate
Speed of effect	Slower (weeks 4-8)	Moderate (weeks 2-6)	Variable
Durability after treatment	High — skills persist	Lower — relapse common after discontinuation	Moderate
Relapse rate post-treatment	20-30% (CBT)	50-60% (medication only)	Unknown / less studied
Cost (out of pocket)	High ($100-$200/session US)	Low (generic SSRIs < $20/month)	Low ($15-40 workbook or free app)
Access barriers	High (waitlists, cost)	Low (primary care prescription)	Low
Appropriate for severe cases	Yes (may need medication first)	Yes (often prioritized for severity)	No
Side effects	Psychological discomfort (therapeutic)	Sexual dysfunction, weight, discontinuation	Minimal
Best suited for	Durable change, skill-building	Acute stabilization, severe presentations	Mild presentations, first step

---

The Evidence for Therapy

CBT: The Gold Standard

Cognitive Behavioral Therapy is the best-studied psychotherapy approach, with hundreds of randomized controlled trials supporting its efficacy across a wide range of conditions. For depression, meta-analyses consistently show effect sizes of 0.6-0.8 compared to control conditions — moderate to large effects by research standards. For anxiety disorders including GAD, social anxiety, panic disorder, and PTSD, effect sizes are often even larger.

The critical advantage of CBT over medication is durability. A comprehensive meta-analysis by Cuijpers and colleagues found that patients treated with CBT maintained their gains significantly better after treatment ended than patients treated with medication alone. Relapse rates in the year following successful CBT treatment for depression are roughly 20-30%, compared to 50-60% for patients who responded to medication and then discontinued it. CBT appears to teach skills that provide ongoing protection against recurrence — the benefit continues after the treatment ends.

The mechanism is partly cognitive and partly behavioral. Cognitive restructuring — learning to identify and challenge distorted automatic thoughts — changes the habitual patterns of thinking that maintain depression and anxiety. Behavioral components — activity scheduling, exposure exercises, behavioral experiments — produce direct change in mood and anxiety through action rather than insight. Most CBT practitioners now recognize that the behavioral components are often the workhorse of the treatment.

Other Evidence-Based Therapies

CBT is not the only effective therapy. Behavioral Activation (BA), Interpersonal Therapy (IPT), and Acceptance and Commitment Therapy (ACT) all have meaningful evidence bases. Psychodynamic therapy has a growing evidence base for depression and some anxiety disorders, though the research quantity is smaller than for CBT.

EMDR (Eye Movement Desensitization and Reprocessing) is specifically well-evidenced for PTSD, with systematic reviews showing large effect sizes comparable to trauma-focused CBT. The mechanism remains debated, but the efficacy for trauma is supported by multiple randomized trials.

Mindfulness-Based Cognitive Therapy (MBCT) has been shown in multiple trials to reduce relapse risk for people with recurrent depression — specifically, those who have had three or more depressive episodes. MBCT combines mindfulness meditation practice with CBT elements and is now recommended by NICE for preventing relapse in this population.

For most people with common mental health conditions accessing the evidence-based treatment landscape, CBT or a CBT-adjacent approach is the place to start due to the volume and quality of supporting research.

Access and Cost Barriers

Therapy's primary limitation is access. A 50-minute CBT session in the US typically costs $100-$200 out of pocket. Insurance coverage is variable and often inadequate. Waitlists for NHS therapy in the UK can be months long despite the free access. In rural areas, specialized therapists may simply not be available locally, making telehealth the practical option.

A critical practical development is the strong evidence for internet-delivered CBT (iCBT). A systematic review by Andrews and colleagues found that iCBT produced effect sizes comparable to face-to-face CBT for depression and anxiety, with the substantial advantage of eliminating geographic and time barriers. Several iCBT programs (Beating the Blues, SilverCloud, This Way Up) have been tested in randomized trials. iCBT substantially changes the access equation.

---

The Evidence for Medication

When Antidepressants Work

For moderate to severe depression, SSRIs and SNRIs have strong evidence. A landmark meta-analysis by Cipriani and colleagues published in The Lancet in 2018, covering 522 trials and over 116,000 patients, concluded that all 21 antidepressants studied were more effective than placebo, with effect sizes ranging from modest to substantial depending on the drug and population.

For anxiety disorders, SSRIs are first-line treatments recommended by major clinical guidelines (NICE, APA). They reduce acute anxiety symptoms reliably and often faster than therapy alone. For social anxiety disorder, evidence for sertraline and escitalopram is particularly strong.

Medication's primary advantage over therapy is speed and accessibility. A primary care appointment can result in a prescription the same day. Medication begins producing effects within 2-6 weeks. For a person in severe distress who cannot function well enough to engage productively in therapy, medication can stabilize symptoms and create the conditions for therapy to work.

An important nuance in the Cipriani data: effect sizes were larger in patients with more severe depression and smaller in mild presentations. The relative benefit of medication over placebo increases with symptom severity — which fits the clinical intuition that medication is most clearly indicated when symptoms are serious.

The Relapse Problem

The critical limitation of medication used in isolation is relapse when it is discontinued. The majority of people who recover from depression on antidepressants and then stop taking them relapse within 6-12 months. Long-term continuation reduces this risk during the period of medication use, but does not eliminate the underlying vulnerability.

This is not a reason to avoid medication — in severe cases the short-term relief is necessary and the alternative is worse — but it argues for using medication as a bridge to therapy rather than as a permanent standalone solution for most patients. A person who stabilizes on medication and uses that stability to complete a course of CBT leaves treatment with both the symptom relief and the skills to prevent recurrence.

Side Effects and Discontinuation

SSRIs have a well-known side effect profile including sexual dysfunction (affecting roughly 30-40% of users), initial increase in anxiety or agitation (the 'SSRI activation' effect in the first 1-2 weeks), sleep disturbance, and weight changes. Discontinuation syndrome — a set of flu-like symptoms, dizziness, and emotional disturbance when stopping medication — is real and can make tapering difficult.

The discontinuation issue is clinically underappreciated. Patients who stop medication abruptly (or even taper too quickly) often experience symptoms that can be mistaken for relapse, making it difficult to distinguish medication withdrawal from returning illness. Slow tapering protocols, supervised by a prescriber, substantially reduce discontinuation symptoms.

These are genuine considerations, not reasons to avoid medication when it is indicated, but factors that should be honestly discussed before starting — particularly the importance of not stopping abruptly.

Medication for Specific Conditions

Different conditions have different optimal pharmacological approaches:

• OCD: Requires higher doses of SSRIs than depression; clomipramine (a tricyclic antidepressant) shows the largest effect sizes but has more side effects
• PTSD: SSRIs (sertraline, paroxetine) are FDA-approved; prazosin has evidence specifically for trauma-related nightmares
• Bipolar disorder: Lithium and anticonvulsant mood stabilizers are the foundation; SSRIs alone are potentially destabilizing and require careful specialist management
• Panic disorder: SSRIs are first-line; benzodiazepines provide rapid relief but carry dependence risk and do not address underlying anxiety
• Social anxiety disorder: SSRIs and the SNRI venlafaxine have strong evidence; propranolol is used for situational performance anxiety

---

The Evidence for Self-Help

What Works and What Does Not

The term 'self-help' covers a vast range of quality, from rigorous CBT-based workbooks to pseudoscientific affirmation content with no evidence base. The research-supported distinction is between structured self-help (guided by clinical principles, often with professional check-ins) and unstructured self-help (reading or consuming content without a program).

Guided self-help based on CBT principles has genuine evidence for mild to moderate depression and anxiety. A systematic review by Coull and Morris found effect sizes for bibliotherapy (structured workbook-based self-help) comparable to brief face-to-face CBT for mild presentations. NICE recommends it as a step one intervention in the stepped-care model for depression and anxiety.

Specific books with evidence behind their approaches include:

• Mind Over Mood by Greenberger and Padesky — a structured CBT workbook with multiple clinical trials supporting its use
• Feeling Good by David Burns — studied in randomized trials; Burns (1999) found that reading the book alone produced measurable symptom reduction in mild-to-moderate depression
• The Anxiety and Worry Workbook by Clark and Beck — CBT-based, well-structured for generalized anxiety

Digital CBT tools with clinical trial support include Beating the Blues (UK NHS-endorsed), SilverCloud, and Woebot (AI-based CBT chatbot). A 2017 meta-analysis by Firth and colleagues found that smartphone mental health apps produced significant reductions in depression and anxiety compared to control conditions, though effect sizes were modest.

Where Self-Help Is Not Sufficient

Self-help is not an appropriate primary intervention for severe depression, bipolar disorder, psychotic disorders, PTSD from severe trauma, active suicidality, or eating disorders requiring medical monitoring. In these cases it may be a useful supplement but cannot substitute for professional care.

The danger of self-help promotion is that it may delay appropriate treatment seeking for people with serious conditions who misclassify their symptoms as mild. A clear signal that self-help is not sufficient: if symptoms have persisted for more than 4-6 weeks without meaningful improvement, if functioning at work or in relationships is significantly impaired, or if any thoughts of self-harm are present — professional assessment is indicated.

The Role of Exercise as Self-Help

Exercise deserves special mention as an evidence-based self-help intervention with a growing and impressive evidence base. A 2023 umbrella review by Singh and colleagues in the British Medical Journal, covering 97 systematic reviews and 1,039 trials, concluded that exercise was more effective than medication or counseling for reducing depression and anxiety. The effect size was particularly large for walking, yoga, and strength training.

Exercise increases BDNF (Brain-Derived Neurotrophic Factor), promotes hippocampal neurogenesis, and modulates HPA axis stress reactivity — the same biological systems implicated in depression and anxiety. The prescription that emerges from the evidence: 30-45 minutes of moderate aerobic exercise at least 3-4 times per week produces clinically meaningful mental health benefits.

---

Why Combination Treatment Often Wins

The most compelling evidence in mental health treatment research is for combination approaches. A frequently cited study published in JAMA by Keller and colleagues compared antidepressant alone, cognitive behavioral analysis system of psychotherapy (CBASP) alone, and the combination for chronic major depression. Response rates were 55% for antidepressant alone, 52% for CBASP alone, and 85% for the combination — a substantial and clinically meaningful advantage.

The mechanisms are additive, not redundant. Medication reduces neurobiological dysregulation — the biochemical aspects of depression and anxiety that make thinking clearly and engaging with the world difficult. Therapy addresses the cognitive patterns, behavioral habits, and interpersonal dynamics that maintain mental health problems and that persist after medication is discontinued.

The combination advantage has been replicated across conditions:

• Treatment-resistant depression: The CoBalT trial (Wiles et al., The Lancet, 2013) found that adding CBT to usual care including medication significantly improved outcomes in patients who had not responded to antidepressants alone — 46% response rate vs 22% for usual care
• OCD: Combination of ERP and SSRIs outperforms either alone; the SSRI reduces anxiety enough to engage with exposure exercises
• Panic disorder: SSRI plus CBT shows faster onset and better long-term maintenance than either alone

A person who stabilizes on medication and uses that stability to engage in CBT is building skills that will protect them after medication ends. This sequencing — medication first, then therapy, then medication taper while maintaining therapy gains — is increasingly standard clinical practice for moderate to severe presentations.

---

Decision Framework by Severity

Mild symptoms (functioning well, intermittent low mood or worry, no impairment): Start with structured self-help — a validated workbook, guided self-help program, or digital CBT tool. Exercise and lifestyle factors (sleep, alcohol reduction) have meaningful evidence at this level. If no improvement in 4-6 weeks, step up.

Moderate symptoms (noticeable impact on daily functioning, persistent symptoms over several weeks): CBT or another evidence-based therapy is the first-line recommendation. Medication is a reasonable alternative if therapy access is limited or if the person prefers it. Many people at this level benefit from combination.

Severe symptoms (significant functional impairment, inability to work or maintain relationships, severe distress): Medication is typically prioritized for speed of relief, combined with therapy as soon as the person is able to engage. Combination treatment should be the explicit plan.

Specific conditions: Bipolar disorder and psychotic disorders require specialist psychiatric management and medication as the foundation. OCD has the strongest evidence for CBT specifically (Exposure and Response Prevention) and often requires high doses of SSRIs. PTSD is best addressed by trauma-focused CBT or EMDR.

---

What to Expect from Each Treatment

Starting CBT

CBT typically begins with psychoeducation — learning the model, understanding how thoughts, feelings, and behaviors interact, and beginning to identify your own patterns. Early sessions focus on assessment and goal-setting. The first few sessions often feel more intellectually engaging than therapeutically powerful.

The therapeutic work typically intensifies in the middle phase: cognitive restructuring, behavioral experiments, exposure exercises. This phase can be uncomfortable — the therapy is designed to challenge avoidance and bring difficult material into focus. Discomfort during this phase is a sign the treatment is engaging, not failing.

Homework assignments between sessions are not optional — they are where the learning consolidates. Research consistently shows that CBT homework compliance predicts outcome. A therapist who does not assign between-session work is delivering a less effective version of the treatment.

Starting Medication

The first 1-2 weeks on an SSRI often produce side effects before benefits: mild nausea, increased anxiety or agitation, disturbed sleep. These initial side effects typically resolve within 1-2 weeks and are not signs that the medication is wrong for you. The therapeutic benefit builds gradually over weeks 2-6.

Patients frequently discontinue medication in the first two weeks because the side effects have arrived but the benefit has not. This is one of the most common failure modes in pharmacological treatment. Being prepared for this window — and having discussed it with the prescriber beforehand — substantially improves adherence.

If no meaningful benefit is apparent at 6-8 weeks at therapeutic dose, this is the appropriate point to discuss switching medication or adding therapy — not a signal to give up on treatment. It is normal to try more than one medication before finding the right fit.

Tracking Progress

Use a standardized scale — the PHQ-9 for depression or GAD-7 for anxiety — to objectively monitor progress rather than relying on impressionistic self-assessment. These validated tools provide a score from 0-27 (PHQ-9) or 0-21 (GAD-7) that can be tracked over time and shared with providers. Many people in treatment have a poor sense of their own trajectory; objective measurement provides a more reliable signal.

---

Practical Recommendations

Do not wait for symptoms to become severe before seeking help. Earlier intervention at the mild to moderate stage is associated with faster recovery, lower relapse rates, and less total treatment duration.

If cost is a barrier to therapy, investigate: university training clinics (supervised graduate students at reduced rates), Open Path Collective ($30-$80 sessions), employer EAP programs (typically 6-12 free sessions), and NHS IAPT services if you are in the UK. Internet-delivered CBT through validated programs is substantially cheaper than face-to-face therapy and has comparable evidence.

If you start medication, discuss a therapy plan with your prescriber. Medication alone is an incomplete treatment plan for most people with depression or anxiety. The goal should be using medication to stabilize while building therapeutic skills.

Give treatments adequate time. Medication typically takes 4-6 weeks to reach therapeutic effect. CBT courses are typically 12-20 sessions. Abandoning treatments in the first few weeks because you do not feel immediately better is a common and costly mistake.

If one therapist is not working, try another. Therapeutic alliance — the quality of the working relationship between therapist and patient — is one of the strongest predictors of psychotherapy outcome. If after 4-6 sessions you do not feel heard, understood, or engaged, finding a different therapist is a reasonable clinical decision, not a failure.

---

References

1. Cuijpers, P., Berking, M., Andersson, G., Quigley, L., Kleiboer, A., & Dobson, K. S. (2013). A meta-analysis of cognitive-behavioural therapy for adult depression, alone and in comparison with other treatments. Canadian Journal of Psychiatry, 58(7), 376-385.
2. Cipriani, A., Furukawa, T. A., Salanti, G., et al. (2018). Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder. The Lancet, 391(10128), 1357-1366.
3. Keller, M. B., McCullough, J. P., Klein, D. N., et al. (2000). A comparison of nefazodone, the cognitive behavioral-analysis system of psychotherapy, and their combination for the treatment of chronic depression. New England Journal of Medicine, 342(20), 1462-1470.
4. Vittengl, J. R., Clark, L. A., Dunn, T. W., & Jarrett, R. B. (2007). Reducing relapse and recurrence in unipolar depression: A comparative meta-analysis of cognitive-behavioral therapy effects. Journal of Consulting and Clinical Psychology, 75(3), 475-488.
5. National Institute for Health and Care Excellence. (2022). Depression in adults: Treatment and management (NICE guideline NG222). NICE.
6. Coull, G., & Morris, P. G. (2011). The clinical effectiveness of CBT-based guided self-help interventions for anxiety and depressive disorders: A systematic review. Psychological Medicine, 41(11), 2239-2252.
7. Burns, D. D. (1999). Feeling Good: The New Mood Therapy (rev. ed.). Harper.
8. Hofmann, S. G., Asnaani, A., Vonk, I. J., Sawyer, A. T., & Fang, A. (2012). The efficacy of cognitive behavioral therapy: A review of meta-analyses. Cognitive Therapy and Research, 36(5), 427-440.
9. Firth, J., Torous, J., Nicholas, J., Carney, R., Rosenbaum, S., & Sarris, J. (2017). Can smartphone mental health interventions reduce symptoms of anxiety? A meta-analysis of randomized controlled trials. Journal of Affective Disorders, 218, 15-22.
10. Cuijpers, P., Donker, T., van Straten, A., Li, J., & Andersson, G. (2010). Is guided self-help as effective as face-to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychological Medicine, 40(12), 1943-1957.
11. Segal, Z. V., Williams, J. M., & Teasdale, J. D. (2013). Mindfulness-Based Cognitive Therapy for Depression (2nd ed.). Guilford Press.
12. Wiles, N., Thomas, L., Abel, A., et al. (2013). Cognitive behavioural therapy as an adjunct to pharmacotherapy for primary care based patients with treatment resistant depression: Results of the CoBalT randomised controlled trial. The Lancet, 381(9864), 375-384.
13. Singh, B., Olds, T., Curtis, R., et al. (2023). Effectiveness of physical activity interventions for improving depression, anxiety and distress: An overview of systematic reviews. British Journal of Sports Medicine, 57(18), 1203-1209.
14. Andrews, G., Cuijpers, P., Craske, M. G., McEvoy, P., & Titov, N. (2010). Computer therapy for the anxiety and depressive disorders is effective, acceptable and practical health care. PLoS ONE, 5(10), e13196.

Frequently Asked Questions

Is CBT more effective than antidepressants for depression?

They are roughly equivalent for moderate depression, but CBT has a clear durability advantage: relapse rates after CBT are 20-30% versus 50-60% after medication alone. For severe depression, medication is often prioritized first.

Does self-help actually work for mental health problems?

Structured, CBT-based self-help has genuine evidence for mild to moderate depression and anxiety — NICE recommends it as a first-line intervention. Unstructured motivational content has almost no clinical evidence behind it.

When should medication be the first choice for mental health?

When symptoms are severe enough to impair daily functioning, when therapy engagement is not possible, or for conditions like bipolar disorder and OCD where pharmacological treatment is foundational. Medication can stabilize symptoms enough for therapy to then become effective.

Why does combination treatment often outperform either alone?

Medication addresses neurobiological dysregulation quickly; therapy builds the cognitive and behavioral skills that prevent relapse. The Keller et al. JAMA study found 85% response rates for combination versus 55% for either treatment alone in chronic depression.

What should I do if I cannot afford therapy?

University training clinics, Open Path Collective (\(30-\)80 sessions), employer EAP programs, and NHS IAPT (UK) are all lower-cost options. Internet-delivered CBT programs have evidence comparable to face-to-face therapy at a fraction of the cost.