American Health & Medical Research Association
Mind-body practices

The relaxation methods guidelines name, and the anxiety they can trigger

Four named relaxation methods carry very different guideline standing, a documented paradoxical anxiety effect, and insurance coverage far narrower than their reputation suggests.

By the Health Research USA editorial team

Published Last updated October 2026

Written from the sources listed below; links checked October 2026. It touches on medicines, doses or health conditions and has not yet been reviewed by a licensed clinician, so use it to prepare for a conversation with yours. How we research and review

Sources28, listed below Reading time16 minutes
On this page
  1. The Four Methods, in a Few Sentences Each
  2. The Side Effect: Relaxation-Induced Anxiety
  3. Where Guidelines Place Each Method
  4. Head to Head With Cognitive Behavioral Therapy
  5. Biofeedback: The Regulated End
  6. Who Delivers the Other Three
  7. Questions to Ask a Relaxation Program
  8. The Takeaway

Here is a finding that runs against intuition. In a 1983 study that gave the problem its name, researchers taught relaxation methods to people with chronic tension, and some of them grew more anxious, not less. By a later paper’s account of that study, about three in ten of those learning progressive muscle relaxation had the reaction, and more than half of those learning mantra meditation did.

That paradox matters because relaxation training is not only something people do on their own at home. In England’s national guideline for generalized anxiety disorder, a structured method called applied relaxation sits beside cognitive behavioral therapy (CBT) as an equal choice. It is delivered from a treatment manual, by trained practitioners, in 12 to 15 weekly hour-long sessions. In US sleep medicine, relaxation therapy carries a formal, if conditional, recommendation for chronic insomnia. And at the device end of the field, biofeedback instruments are regulated by the Food and Drug Administration.

This article covers four named methods as treatments with a track record: progressive muscle relaxation, autogenic training, applied relaxation and biofeedback. It doesn’t teach any of them, and it leaves the general biology of stress to other parts of the site. What follows is where each method stands in guidelines, what it has matched and lost to in head-to-head trials, who delivers it, what insurers pay for, and what it can do wrong.

The four methods, in a few sentences each

Progressive muscle relaxation (PMR) has a person tense and then release muscle groups in a set order. The 2020 German medical history cited in the next entry names Jacobson as its founder.

Autogenic training uses repeated self-suggestions of heaviness, warmth and other body sensations. Johannes Heinrich Schultz developed it from experiments with hypnosis in Jena around 1920. He published his first book on it in 1932, according to a 2020 history in Deutsches Ärzteblatt, a German medical journal. The same article reports that it became a benefit of German statutory health insurance around 1967, with prevention courses covered from 1989. Schultz, it adds, wanted the people teaching it to hold a license to practice, professional experience and solid further training.

Applied relaxation was described by Öst in a 1987 paper. It begins with progressive muscle relaxation. Over several weeks, it then trains people to relax quickly and to use the skill as a portable coping response when anxiety starts, according to a 2019 description by researchers at Penn State.

Biofeedback uses an instrument to show a person a real-time signal from their own body, such as muscle activity, skin temperature or heart rhythm, so they can learn to change it. The American Academy of Sleep Medicine (AASM) describes it as a variant of relaxation training that adds a monitoring device.

Federal sources often list these alongside guided imagery and self-hypnosis. Why imagery resists separate grading is laid out in our guided imagery coverage, and self-hypnosis falls under the library’s hypnosis article.

The side effect: relaxation-induced anxiety

The paradox in the opening has a name, relaxation-induced anxiety, and a research record running from 1983 to the present. It belongs before any discussion of benefits.

The term comes from a 1983 paper by Heide and Borkovec in the Journal of Consulting and Clinical Psychology. The figures most often cited for it come from a 2019 summary by Hanjoo Kim and Michelle Newman of Penn State, published in the Journal of Affective Disorders. They report that among the original study’s participants with chronic tension, 30.8 percent of those doing PMR and 53.8 percent of those doing mantra meditation experienced it. Those numbers are the 2019 authors’ account of the older study.

Kim and Newman describe the reaction as a spike in anxiety, muscle tension or anxious thoughts and images that happens while a person is trying to relax. They point out a double irony: the people most prone to it are those with anxiety disorders, the very people relaxation training is meant to help. Citing earlier treatment studies, they report that the reaction predicted smaller improvements in anxiety and depression 12 months after treatments that included applied relaxation. Patients with lower peak reactions, they add, did better after therapy. Their own experiment analyzed 89 undergraduates, none seeking treatment: 28 who screened positive for generalized anxiety disorder, 32 for depression and 29 healthy controls. It linked the reaction to discomfort with sudden shifts from calm into negative emotion. The Penn State authors concluded that when applied relaxation ignores that sensitivity, it “could be a poisoned chalice to those in need of help.”

Other studies point the same way, with small numbers. A 1985 study followed 38 people who said they felt anxious at least half the day. During training, 2 of 20 doing PMR and 3 of 18 doing a meditation procedure showed rising heart rates. Earlier reports described panic attacks during relaxation in people with panic disorder. A 1997 study from the University of Ottawa complicates that picture: when panic disorder patients listened to relaxation tapes, their panic-symptom ratings did not change. A 1988 theory paper proposed that people who chronically over-breathe can tip into panic sensations when they settle down to relax, because their bodies produce less carbon dioxide while their breathing stays the same. That is a hypothesis, not a settled mechanism.

The National Center for Complementary and Integrative Health (NCCIH), on a relaxation page last updated in June 2021, acknowledges the issue in general terms. It records occasional reports of increased anxiety, intrusive thoughts or fear of losing control, and rare reports of worsened symptoms in people with epilepsy, certain psychiatric conditions, or a history of abuse or trauma.

Signs that call for a licensed clinician: panic attacks during relaxation practice, anxiety that reliably climbs rather than falls during sessions, intrusive memories, or a diagnosed anxiety, panic or trauma-related disorder. Kim and Newman argue that therapists using applied relaxation should assess the reaction during treatment rather than assume relaxation is working.

Where guidelines place each method

Read claim by claim, the guideline record looks like this.

Generalized anxiety disorder: an equal alternative to CBT. NICE, England’s guideline body, covers this in its guideline on generalized anxiety disorder (GAD) and panic disorder in adults. If a person with GAD chooses a high-intensity psychological treatment, NICE says, they should be offered either CBT or applied relaxation. That recommendation dates from 2011. NICE specifies that applied relaxation should follow the manuals used in the clinical trials, be delivered by trained and competent practitioners, and usually run 12 to 15 weekly sessions of one hour each. For panic disorder, NICE’s psychological recommendations name only CBT. This is the strongest guideline position this article found for any relaxation method: recommended by name in a clinical guideline, as one of two equal high-intensity psychological options, for one condition.

Chronic insomnia: conditional. The AASM’s 2021 guideline suggests relaxation therapy as a single-component treatment for chronic insomnia in adults. That is a conditional recommendation, on evidence the panel rated low quality because of bias and imprecision; the panel identified 12 randomized trials, five of which could be pooled. It judged that the modest benefits likely outweigh the minimal harms and burdens. Multicomponent CBT for insomnia received the only strong recommendation. Biofeedback got no recommendation either way, because fewer than three qualifying studies existed. The academy’s announcement of a newer guideline, published in April 2026 on combining CBT for insomnia with medication, does not mention relaxation.

Chronic low back pain: on a menu. The American College of Physicians’ 2017 guideline strongly recommends that people with chronic low back pain start with a nondrug treatment. The choice comes from a long list that includes progressive relaxation and EMG biofeedback, meaning feedback from muscle-activity sensors. The evidence for those two options was rated low quality. The strong grade attaches to choosing from the list, not to PMR itself.

Cancer: undefined. The 2023 cancer guideline on anxiety and depression, issued by the Society for Integrative Oncology with ASCO, the American Society of Clinical Oncology, mentions relaxation therapies twice. They may be offered for anxiety during active cancer treatment, on intermediate evidence with moderate strength. They may also be offered for depression during treatment, though there the evidence is low and the strength weak. ASCO’s summary table doesn’t say which techniques count. It does name biofeedback and autogenic training, but only in lists of approaches whose evidence it calls inconclusive, for anxiety and for depression alike.

Hospitals: named, not required. The report explaining the Joint Commission‘s pain standard, effective in 2018, lists relaxation therapy among examples of nondrug pain treatments, while a Joint Commission FAQ says no specific option is mandated.

Hot flashes: a federal and professional split. NCCIH’s June 2021 page says a 2019 review found overall evidence that paced breathing, a slow and regular breathing pattern, could significantly improve hot flashes. The North American Menopause Society’s 2023 nonhormone position statement went the other way, listing paced respiration as not recommended at its highest evidence level and relaxation as not recommended at its second level. The society, which now goes by the Menopause Society, did list clinical hypnosis as recommended. The NCCIH page predates the 2023 statement, and the two positions stand unreconciled.

Head to head with cognitive behavioral therapy

Guideline wording only goes so far. Direct comparisons with CBT show where relaxation holds its own and where it falls short.

  • Insomnia: In a 2001 JAMA trial of 75 adults with long-standing trouble staying asleep, CBT cut time awake after first falling asleep by 54 percent, PMR by 16 percent and a placebo therapy by 12 percent. On that measure, PMR barely beat the placebo. At the end of treatment, the PMR group still logged about 43 minutes awake in the middle of the night, against about 27 with CBT. PMR patients slept 78.8 percent of their time in bed; CBT patients slept 85.1 percent.
  • Generalized anxiety: A 2007 meta-analysis found cognitive therapy and relaxation therapy equivalent for GAD.
  • Panic disorder: The same analysis found cognitive therapy, which included deliberate exposure to feared body sensations, better than relaxation on every panic measure. A 2018 meta-analysis of 50 studies with 2,801 patients found no significant difference for panic but noted that relaxation might be less effective at one year. The two analyses don’t fully agree.
  • All anxiety disorders together: That 2018 analysis favored cognitive and behavioral therapies overall by a small margin. It found no difference for GAD, panic, social anxiety or specific phobias, and relaxation fared worse for post-traumatic stress disorder and obsessive-compulsive disorder.
  • Autogenic training: A 2002 meta-analysis of 60 studies published between 1952 and 1999 found medium effects against control conditions, but mostly no effect or small negative effects when autogenic training was compared with other psychological treatments. Medium effects appeared in at least three studies each for a long list of conditions. They included tension headache and migraine, mild to moderate high blood pressure, coronary heart disease, asthma, somatoform pain, Raynaud’s disease, anxiety disorders, mild to moderate depression and functional sleep problems. Its search ended in 1999, so that list describes an old literature.

One federal summary goes further than its source’s abstract. NCCIH’s relaxation page says a 2018 review of 16 trials found relaxation therapy more effective than CBT for reducing anxiety. The review’s published abstract, in Archives of Psychiatric Nursing, does not support that reading. It reports a medium-to-large overall effect and says only that the size of the effect differed depending on whether relaxation was compared with usual care or with CBT. The full paper, which was not available for this article, may break that comparison down in a way that backs NCCIH’s summary. The larger 2018 analysis of 50 studies, which NCCIH also summarizes, points the other way.

On this library’s evidence scale, relaxation for GAD has positive trial evidence and a matching guideline position. PMR alone for insomnia is measured but small next to the treatment guidelines prefer. And the claim that relaxation matches CBT for PTSD and OCD has been tested and not supported.

Biofeedback: the regulated end

Biofeedback is the one relaxation method with a federal regulatory identity, and that identity is narrow.

The device. Under 21 CFR 882.5050, a biofeedback device is an instrument that gives a visual or sound signal tied to a body measure, such as brain-wave activity, muscle activity or skin temperature. The point is to let the patient learn to control that measure. It is a Class II device. A prescription, battery-powered unit intended for relaxation training and muscle reeducation is exempt from premarket notification, the process known as 510(k). The rule dates to 1979 and was amended in 1998.

What clearance says. The regulation is built around relaxation training and muscle reeducation, and a clearance can name relaxation as the only use. One EEG neurofeedback system cleared in February 2013, according to FDA’s records as reproduced by a third-party database, is labeled for general relaxation training, with no disorder named. One breathing-based device was cleared through 510(k) in August 2018 as an add-on treatment for symptoms of panic disorder or PTSD. A November 2024 evidence review for New York State’s Medicaid program found three uncontrolled studies with 176 participants and said its reviewers had very low certainty in the results. Of the 13 private insurers and 10 Medicaid programs it checked, one covered the device, and, according to the same review, the 2023 VA and Department of Defense PTSD guideline found insufficient evidence to recommend for or against it. A 510(k) clearance means FDA judged the device substantially equivalent to one already sold, which is not proof that it works; the library’s device-regulation coverage explains that difference.

What Medicare pays for. Medicare’s national coverage determination for biofeedback therapy covers it only when it is reasonable and necessary for retraining specific muscle groups or treating abnormal muscle states, such as spasticity, incapacitating spasm or weakness. Even then, more conventional treatments like heat, cold, massage and exercise must have failed first. Medicare’s determination adds that biofeedback is “not covered for treatment of ordinary muscle tension states or for psychosomatic conditions.”

The evidence. A 2017 meta-analysis of heart rate variability biofeedback, 24 studies with 484 participants, found a large reduction in self-reported stress and anxiety compared with control conditions, while its authors called for more well-controlled studies. NCCIH’s June 2021 page adds limits elsewhere. A 2018 review it describes found biofeedback for anxiety and depression in children promising but not ready for clinical use, and a 2020 evaluation found no clear benefit for fibromyalgia.

The practitioner. The Biofeedback Certification International Alliance (BCIA) runs the main certification, a private credential. Entry-level certification requires at least a bachelor’s degree in an approved health field, a 42-hour didactic course and an anatomy and physiology course. Candidates also need 20 contact hours with an approved mentor, 50 client sessions, 10 written case studies and a proctored exam of about 100 questions. Someone without a state health license must work under a credentialed professional’s supervision when treating a medical or psychological disorder.

This library’s credential test has four buckets, running from titles defined in regulation through government and private certification to titles nobody defines. The device sits in the first. The person operating it holds, at most, a private certificate from the third.

Who delivers the other three

None of the sources reviewed for this article describes a state license specific to PMR, autogenic training or applied relaxation, and relaxation therapist appears to be an undefined title, the fourth bucket. What protects you is the delivering clinician’s underlying license, if there is one.

NICE’s standard for GAD shows what a treatment-grade version looks like. It calls for trained and competent practitioners working from the manuals used in the trials. Its broader guidance also asks that people delivering high-intensity psychological treatments get regular supervision to keep them faithful to the treatment model. That describes a structured course of therapy, closer to a series of appointments than to a class.

Germany offers a contrast in how a health system can absorb one of these methods. There, autogenic training became an insurance benefit decades ago, and its founder argued that the people teaching it should be licensed, experienced and further trained.

Questions to ask a relaxation program

  1. Which method is this, by name, and is it delivered from a published manual?
  2. How many sessions, how long, and who supervises the person delivering it?
  3. What happens if my anxiety goes up during practice instead of down?
  4. For a biofeedback device: what exactly was it cleared for, relaxation training or a named condition?
  5. If I have a diagnosed condition, how does this method compare with CBT for that specific condition?

The Takeaway

Relaxation training is not one thing, and its standing depends on the method and the problem. Applied relaxation for generalized anxiety disorder, delivered from a manual over 12 to 15 weekly sessions, has the strongest record and a guideline seat beside CBT. For insomnia, PTSD and OCD, it trails CBT, and for panic disorder the evidence leans the same way. For hot flashes, the Menopause Society’s 2023 statement does not recommend it. And for a meaningful minority of people, especially those with anxiety disorders, relaxation itself can provoke anxiety. That is why Kim and Newman want therapists to check for the reaction during treatment, and why relaxation is better understood as a therapy than as a harmless extra.

Sources

  1. Journal of Affective Disorders. “Kim H, Newman MG. "The paradox of relaxation training: Relaxation induced anxiety and mediation effects of negative contrast sensitivity in generalized anxiety disorder and major depressive disorder." J Affect Disord, 2019. PubMed. DOI: 10.1016/j.jad.2019.08.045.” pubmed.ncbi.nlm.nih.gov
  2. Journal of Consulting and Clinical Psychology. “Heide FJ, Borkovec TD. "Relaxation-induced anxiety: paradoxical anxiety enhancement due to relaxation training." J Consult Clin Psychol, 1983. PubMed. DOI: 10.1037//0022-006x.51.2.171.” pubmed.ncbi.nlm.nih.gov
  3. Journal of Behavior Therapy and Experimental Psychiatry. “Norton GR, Rhodes L, Hauch J, Kaprowy EA. "Characteristics of subjects experiencing relaxation and relaxation-induced anxiety." J Behav Ther Exp Psychiatry, 1985. PubMed. DOI: 10.1016/0005-7916(85)90065-5.” pubmed.ncbi.nlm.nih.gov
  4. Journal of Anxiety Disorders. “Knott V, Bakish D, Lusk S, Barkely J. "Relaxation-induced EEG alterations in panic disorder patients." J Anxiety Disord, 1997. PubMed. DOI: 10.1016/s0887-6185(97)00016-9.” pubmed.ncbi.nlm.nih.gov
  5. Journal of Behavior Therapy and Experimental Psychiatry. “Ley R. "Panic attacks during relaxation and relaxation-induced anxiety: a hyperventilation interpretation." J Behav Ther Exp Psychiatry, 1988. PubMed. DOI: 10.1016/0005-7916(88)90054-7.” pubmed.ncbi.nlm.nih.gov
  6. National Center for Complementary and Integrative Health, National Institutes of Health. “Relaxation Techniques: What You Need To Know (content Last Updated June 2021; safety statements, paced breathing and hot flashes, anxiety reviews).” nccih.nih.gov
  7. National Institute for Health and Care Excellence (UK). “Generalised anxiety disorder and panic disorder in adults: management (CG113), section 1 Guidance (applied relaxation recommendations [2011]).” nice.org.uk
  8. American Academy of Sleep Medicine (Journal of Clinical Sleep Medicine). “Edinger JD, Arnedt JT, Bertisch SM, et al. "Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline." J Clin Sleep Med, 2021 (Recommendation 5; Table 3). DOI: 10.5664/jcsm.8986.” pmc.ncbi.nlm.nih.gov
  9. American Academy of Sleep Medicine. “New guideline provides recommendations on combining treatments for chronic insomnia (April 22, 2026; guideline published April 13, 2026).” aasm.org
  10. Annals of Internal Medicine (American College of Physicians). “Qaseem A, Wilt TJ, McLean RM, Forciea MA. "Noninvasive Treatments for Acute, Subacute, and Chronic Low Back Pain: A Clinical Practice Guideline From the American College of Physicians." Ann Intern Med, 2017. PubMed. DOI: 10.7326/M16-2367.” pubmed.ncbi.nlm.nih.gov
  11. American Society of Clinical Oncology. “Integrative Oncology Care of Symptoms of Anxiety and Depression in Adults with Cancer: SIO-ASCO Guideline (2023) Summary Table, Recommendations 1.4 and 3.4.”
  12. The Joint Commission. “R3 Report Issue 11: Pain assessment and management standards for hospitals (August 29, 2017).” jointcommission.org
  13. The Joint Commission. “Standards FAQ: Acceptable non-pharmacologic pain treatment modalities (last reviewed November 17, 2022).” jointcommission.org
  14. Menopause (The North American Menopause Society, now The Menopause Society). “"The 2023 nonhormone therapy position statement of The North American Menopause Society." Menopause, 2023. PubMed. DOI: 10.1097/GME.0000000000002200.” pubmed.ncbi.nlm.nih.gov
  15. The Menopause Society (formerly The North American Menopause Society). “Clinical Hypnosis vs. Cognitive Behavioral Therapy: What's Better for Managing Hot Flashes? (press release, September 2024; confirms current name).” menopause.org
  16. JAMA (American Medical Association). “Edinger JD, Wohlgemuth WK, Radtke RA, Marsh GR, Quillian RE. "Cognitive behavioral therapy for treatment of chronic primary insomnia: a randomized controlled trial." JAMA, 2001. PubMed. DOI: 10.1001/jama.285.14.1856.” pubmed.ncbi.nlm.nih.gov
  17. Journal of Consulting and Clinical Psychology. “Siev J, Chambless DL. "Specificity of treatment effects: cognitive therapy and relaxation for generalized anxiety and panic disorders." J Consult Clin Psychol, 2007. PubMed. DOI: 10.1037/0022-006X.75.4.513.” pubmed.ncbi.nlm.nih.gov
  18. Psychological Medicine. “Montero-Marin J, Garcia-Campayo J, López-Montoyo A, Zabaleta-Del-Olmo E, Cuijpers P. "Is cognitive-behavioural therapy more effective than relaxation therapy in the treatment of anxiety disorders? A meta-analysis." Psychol Med, 2018. PubMed. DOI: 10.1017/S0033291717003099.” pubmed.ncbi.nlm.nih.gov
  19. Archives of Psychiatric Nursing. “Kim HS, Kim EJ. "Effects of Relaxation Therapy on Anxiety Disorders: A Systematic Review and Meta-analysis." Arch Psychiatr Nurs, 2018. PubMed. DOI: 10.1016/j.apnu.2017.11.015.” pubmed.ncbi.nlm.nih.gov
  20. Applied Psychophysiology and Biofeedback. “Stetter F, Kupper S. "Autogenic training: a meta-analysis of clinical outcome studies." Appl Psychophysiol Biofeedback, 2002. PubMed. DOI: 10.1023/a:1014576505223.” pubmed.ncbi.nlm.nih.gov
  21. Behaviour Research and Therapy. “Öst LG. "Applied relaxation: description of a coping technique and review of controlled studies." Behav Res Ther, 1987. PubMed. DOI: 10.1016/0005-7967(87)90017-9.” pubmed.ncbi.nlm.nih.gov
  22. Deutsches Ärzteblatt PP. “Husmann B. Autogenes Training: Potentes Psychotherapeutikum (issue 6/2020, online June 9, 2020, updated April 18, 2024).” aerzteblatt.de
  23. Legal Information Institute, Cornell Law School (Code of Federal Regulations). “21 CFR 882.5050 Biofeedback device (44 FR 51730, Sept. 4, 1979, as amended at 63 FR 59229, Nov. 3, 1998).” law.cornell.edu
  24. Innolitics (third-party mirror of FDA 510(k) data). “K122879, EEGer4 Model 4.3, decision date February 6, 2013, product code HCC, indication: general relaxation training.” fda.innolitics.com
  25. New York State Department of Health, Medicaid Evidence-Based Benefit Review Advisory Committee. “Freespira health technology assessment (meeting of November 7, 2024).” healthweb-back.health.ny.gov
  26. Centers for Medicare & Medicaid Services. “National Coverage Determination: Biofeedback Therapy (NCD 30.1, Publication 100-3; longstanding NCD, Version 1).” cms.gov
  27. Psychological Medicine. “Goessl VC, Curtiss JE, Hofmann SG. "The effect of heart rate variability biofeedback training on stress and anxiety: a meta-analysis." Psychol Med, 2017. PubMed. DOI: 10.1017/S0033291717001003.” pubmed.ncbi.nlm.nih.gov
  28. Biofeedback Certification International Alliance. “BCIA Biofeedback Certification: Entry Level requirements.” bcia.org