In children, biofeedback works at least as well as in adults, and probably better. It is a well-tolerated, medication-free option for migraine and tension-type headache. Neurofeedback, by contrast, has been tested in only one small, older study.
AAPB2Possibly efficaciousNeuroLogic2Possibly efficacioussame vs AAPB
Level 2 (AAPB level assigned jointly to EEG neurofeedback and electrodermal biofeedback, which AAPB describes as the 'less supported' modalities; it prints no separate level for them.) A single study: Siniatchkin 2000, an exploratory non-randomised comparison of 10 children with migraine without aura trained over 10 sessions in 8 weeks to regulate slow cortical potentials (Cz/Fz, contingent negative variation paradigm), 10 matched migraine children on a waitlist and 10 non-migraine controls; migraine frequency and migraine index fell significantly in the trained group. Well-documented procedure, small sample, non-randomised assignment. The 2022-2026 search finds no paediatric neurofeedback trial for headache (Connelly 2023 tested the acceptability of extended-reality relaxation with a wearable neurofeedback headband in 10- to 17-year-olds with migraine, without efficacy data). Possibly efficacious; no replication in 25 years.
Biofeedback
Peripheral thermal and surface EMG biofeedback (migraine and tension-type headache)
Ages
Child (under 12) · Adolescent (12-17)
Techniques
Thermal biofeedback, EMG biofeedback
AAPB4EfficaciousNeuroLogic4Efficacioussame vs AAPB
Level 4 maintained. Randomised trials from independent groups: Bussone 1998 (30 consecutive patients aged 11-15, 10 sessions of frontal EMG biofeedback plus relaxation vs a relaxation placebo control: significant pain reduction in the biofeedback arm, EMG change tracking clinical change, gains up to 12 months); Kröner-Herwig 1998 (n = 50, ages 8-14, 12 EMG sessions vs progressive relaxation vs waitlist: both treatments responded, biofeedback with the greater reduction in headache duration and the largest effect sizes); Allen & Shriver 1998 (n = 27, thermal biofeedback with or without parent-mediated pain-behaviour management, both arms improved over one year, the parent arm more); Guarnieri & Blanchard 1990 (n = 16, clinic vs home thermal training, no significant difference); Fentress 1986 (n = 18, EMG plus relaxation vs relaxation, no added benefit). Single-arm series with 1- to 3-year follow-up (Grazzi 1990, 2001; Hermann 1997, n = 32, younger children improving more). The 2022-2026 window adds no paediatric trial: Treadwell 2025 (13 trials, n = 1444) concludes only that CBT plus biofeedback plus relaxation may reduce migraine frequency and disability more than education (low strength of evidence); Schwarz 2025 (n = 74, retrospective, post-traumatic headache: 42 % responders). Level 5 is not assigned: superiority over relaxation, shown in two trials, is contradicted by Fentress 1986 and not tested against a credible placebo in a second setting.
AAPB2Possibly efficaciousNeuroLogic2Possibly efficacioussame vs AAPB
Level 2 (AAPB level assigned jointly to EEG neurofeedback and electrodermal biofeedback, the AAPB's 'less supported' modalities.) A single single-arm pilot: Shiri 2013, 10 children aged 10-17 with chronic tension-type or migraine headache (9 completers), 10 sessions of virtual-reality electrodermal biofeedback in which skin conductance drives the child's own images of emotional states; pain intensity fell at post-test and at 1- and 3-month follow-up. No control group, no later study in the 2022-2026 window, no meta-analysis isolates it. Held at the AAPB's level for a modality it lists as less supported; a strict reading of a nine-child pilot would give Level 1.
In short
Clinical reading
AAPB Level 4 confirmed by NeuroLogic for peripheral thermal and surface EMG biofeedback: independent RCTs (Bussone 1998, Kröner-Herwig 1998, Allen & Shriver 1998) and series with follow-up to 3 years; superiority over relaxation in two trials, not replicated against placebo. SCP neurofeedback (Siniatchkin 2000, non-randomised) and EDA biofeedback (Shiri 2013, pilot): Level 2. No new paediatric trial in 2022-2026; Treadwell 2025 concludes only for the CBT + biofeedback + relaxation package over education.
Protocols
Digital thermal and frontalis EMG biofeedback, 6-10 sessions with home practice; growing interest in HRV for paediatrics.
Limits
Relaxation alone and self-administered interventions also work, which raises the question of biofeedback's added value against cheaper options; older studies sometimes report incomplete data; no paediatric RCT since 1998 and no HRV trial in childhood headache; neurofeedback limited to one non-randomised study.
Study base
Five RCTs and six quasi-experimental studies in the AAPB base, plus meta-analyses spanning adults and children; 2022-2026 base: 4 publications indexed in the archive (two mixed adult/child meta-analyses, one systematic review, one retrospective cohort), no new trial.
Brendan's perspective
Level 4, the same as the AAPB, and it is deserved: Bussone 1998, Kröner-Herwig 1998 and Allen & Shriver 1998 are independent, randomised and followed children for a year or more. What should stop a reader is the dates. No paediatric randomised trial since 1998 — the evidence base is older than the patients. Level 5 is not given because superiority over relaxation, shown twice, is contradicted by Fentress 1986. The SCP row rests on one non-randomised study, Siniatchkin 2000, unreplicated in twenty-five years; an absence of replication that long is a finding, not a gap. In clinic I would train finger temperature and frontal EMG with breathing, six to ten short sessions, parents in the room, daily practice at home, and I would add HRV before I added EEG. Tell the family plainly that relaxation alone also helps, and that nobody has tested the difference in a generation.
Bussone et al. (1998) Biofeedback-assisted relaxation training for young adolescents with tension-type headache: a controlled study doi:10.1046/j.1468-2982.1998.1807463.x
Kröner-Herwig, Mohn & Pothmann (1998) Comparison of biofeedback and relaxation in the treatment of pediatric headache and the influence of parent involvement on outcome doi:10.1023/a:1022267104369
Allen & Shriver (1998) Role of parent-mediated pain behavior management strategies in biofeedback treatment of childhood migraines doi:10.1016/S0005-7894(98)80044-0
Stubberud et al. (2016) Biofeedback as prophylaxis for pediatric migraine: A meta-analysis doi:10.1542/peds.2016-0675
Siniatchkin et al. (2000) Self-regulation of slow cortical potentials in children with migraine: An exploratory study doi:10.1023/a:1009581321624
Treadwell et al. (2025) Behavioral interventions for migraine prevention: A systematic review and meta-analysis doi:10.1111/head.14914
Dormal et al. (2021) Is heart rate variability biofeedback useful in children and adolescents? A systematic review doi:10.1111/jcpp.13463
