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 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.
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.
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.
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 is specifically well-evidenced for PTSD.
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. These barriers are real and account for a significant portion of the gap between what evidence supports and what people actually receive.
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.
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. 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.
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. These are genuine considerations, not reasons to avoid medication when it is indicated, but factors that should be honestly discussed before starting.
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, 'Feeling Good' by David Burns (which has been studied in randomized trials), and 'The Anxiety and Worry Workbook' by Clark and Beck. Digital CBT tools like Beating the Blues and SilverCloud have been tested in clinical trials.
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.
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.
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. Neither fully addresses what the other targets. A person who stabilizes on medication and uses that stability to engage in CBT is building skills that will protect them after medication ends.
For severe anxiety and depression, for OCD, and for panic disorder, combination treatment is now the standard recommendation in major clinical guidelines — not as a last resort after simpler approaches fail, but as the primary approach 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.
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.
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.
Track your symptoms. Use a standardized scale — the PHQ-9 for depression or GAD-7 for anxiety — to objectively monitor your progress rather than relying on impressionistic self-assessment. This also helps communicate with providers.
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.
References
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.
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.
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.
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.
National Institute for Health and Care Excellence. (2022). Depression in adults: Treatment and management (NICE guideline NG222). NICE.
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.
Burns, D. D. (1999). Feeling Good: The New Mood Therapy (rev. ed.). Harper.
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.
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.
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.
Segal, Z. V., Williams, J. M., & Teasdale, J. D. (2013). Mindfulness-Based Cognitive Therapy for Depression (2nd ed.). Guilford Press.
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.
Frequently Asked Questions
Is CBT more effective than antidepressants for depression?
Research suggests they are roughly equivalent for moderate depression, with some studies favoring medication for speed and others favoring CBT for durability. A landmark meta-analysis by Cuijpers and colleagues found that CBT and antidepressants produced similar remission rates for major depression. The critical difference is relapse: patients treated with CBT show substantially lower relapse rates after treatment ends compared to patients treated with medication alone — sometimes half the relapse rate. For mild to moderate depression, NICE guidelines recommend CBT as a first-line treatment. For severe depression, medication is typically recommended first, often alongside therapy.
Does self-help actually work for mental health problems?
For mild to moderate anxiety and depression, structured self-help has meaningful evidence behind it. Guided self-help based on CBT principles — where a therapist provides brief check-ins on your progress through a workbook — is recommended by NHS guidelines for mild depression and anxiety. Unguided self-help (books, apps, no professional contact) shows weaker but still positive effects. The key word is 'structured' — working through a program like David Burns' 'Feeling Good' workbook or a validated app like Woebot differs meaningfully from reading motivational content. Self-help is not a substitute for professional care in severe cases, but it is a legitimate first step for mild presentations.
When should medication be the first choice for mental health?
Medication is generally the first-line recommendation when symptoms are severe — when depression or anxiety is significantly impairing daily functioning, when there is active suicidality, or when the severity makes engaging with therapy too difficult. Bipolar disorder, schizophrenia, and severe OCD have strong evidence bases for pharmacological intervention as the foundation of treatment. For panic disorder, SSRIs often produce faster response than therapy alone in the acute phase. The threshold is roughly: if you cannot reliably attend and engage with therapy sessions due to symptom severity, medication may need to stabilize you first before therapy becomes effective.
Why does combination treatment often outperform either alone?
Medication can reduce symptom severity quickly, lowering the baseline of suffering enough that a person can actively engage with therapy. Therapy then provides cognitive and behavioral tools that remain after medication is discontinued, reducing relapse risk. A widely cited study in JAMA compared combination treatment to each alone for chronic depression and found the combination produced significantly higher remission rates — 85% vs around 55% for either treatment alone. The mechanisms are complementary: medication addresses neurobiological dysregulation while therapy addresses the cognitive patterns and behavioral avoidance that maintain mental health problems. Neither alone addresses both layers.
What should I do if I cannot afford therapy?
Several options exist at lower cost. Community mental health centers offer sliding-scale fees based on income. Training clinics at universities provide supervised therapy from graduate students at reduced rates. Open Path Collective connects clients to therapists offering sessions at \(30-\)80. Many employers offer Employee Assistance Programs (EAPs) with 6-12 free therapy sessions. For mild to moderate anxiety and depression, structured self-help workbooks and validated apps provide real benefit. Online therapy platforms like BetterHelp and Talkspace cost less than traditional therapy though quality varies. The NHS in the UK provides free access to IAPT services with self-referral. Cost should not be an absolute barrier — lower-cost options have genuine evidence.