Are Psychological Interventions Efficacious for Adults With Migraine?

A Systematic Review and Meta-Analysis

Joanne Dudeney PhD; Louise Sharpe PhD; Sarah McDonald PhD; Rachel E. Menzies PhD; Brian McGuire PhD

Headache. 2022;62(4):405-419.

Abstract and Introduction

Abstract

Objective: To determine the efficacy of psychological interventions in studies of adults with migraine, in reducing pain, and functional and psychosocial difficulties.

Background: A recently published Cochrane Review showed no strong evidence for psychological intervention for adults with migraine. However, this review was limited by stringent inclusion criteria, potentially resulting in low power. The current review examines the stability of these findings.

Methods: CENTRAL, MEDLINE, Embase, PsycINFO, and CINAHL databases were systematically searched from inception to July 31, 2021. Reviewers independently conducted the search and extraction process. Risk of bias was conducted, and we provide GRADE ratings of our confidence in the evidence. We included 39 randomized controlled trials in the meta-analyses (n = 3155); 31 included data on migraine only and eight combined data from adults with migraine or tension-type headache (TTH).

Results: In contrast to the Cochrane Review, for studies including adults with migraine only, we found psychological interventions had a small to medium beneficial effect on improving migraine frequency, pain intensity, and disability post-treatment, compared to controls (Cohen's d range 0.23 to 0.33), and disability at follow-up (d = 0.44). We found no evidence of a beneficial effect on quality-of-life post-treatment, or mood at post-treatment or follow-up. Our sensitivity analyses on studies with mixed headache populations (migraine and combined migraine/TTH) found a similar magnitude of effects for most outcomes (d range 0.25 to 0.36), except for mood, which produced a small to medium beneficial effect (d = 0.32).

Conclusion: We found evidence that adults with migraine may benefit from psychological intervention, and psychological treatment may equally benefit adults with a range of headache conditions. However, as with previous reviews, quality of the evidence was low, findings should be cautiously interpreted, and this field would benefit from future high-quality trials.

Introduction

Migraine is a leading cause of pain in adults, with prevalence rates of approximately 15%.^[1,2] The individual and societal burden is extreme, with approximately 28% reporting substantial headache-related disability,^[1,2] along with increased psychiatric difficulties, reduced health-related quality of life, greater work absenteeism, and increased emergency department and long-term health-care use.^[1,3–6]

Prophylactic and acute medications are a first-line treatment.^[7] However, data on the use of preventative medication shows less than 25% adherence rates after 12 months.^[8] Factors identified for low adherence include medication cost, contraindication (e.g., pregnancy, diabetes, heart conditions),^[9] adverse effects,^[10,11] intolerability of frequency and dose,^[12] perceived delay in therapeutic benefits,^[13,14] or decreased efficacy over time.^[15]

Accordingly, there has been interest in reducing or preventing migraine by non-pharmacological methods such as psychological intervention. These interventions aim to reduce the psychological processes (e.g., cognitions, behaviors, attention) that can maintain migraine pain and associated disability and distress. However, periodic evidence-based reviews on psychological approaches have reported contrasting findings. Reviews in favor of psychological intervention for adults with migraine are limited by qualitative methodology,^[16,17] precluding the calculation of treatment efficacy, or included inadequate study designs (e.g., uncontrolled, non-randomized studies, small sample sizes) which potentially leads to an overestimation of the observed effects.^[18–20] In contrast, a recent Cochrane Review published by members of our research group on randomized controlled trials (RCTs) showed no strong evidence for psychological intervention.^[21] There are some possible reasons for the contradictory findings. First, Cochrane Reviews are stringent in how data are reported in the eligible studies, only including outcome data reported as means and standard deviations. This leads to a small number of studies being eligible for meta-analysis, potentially reducing power. Second, the Cochrane Review included studies with migraine only, excluding those in which participants with migraine and other headache conditions (e.g., tension-type headache [TTH]) were combined in analyses. This decision mitigated heterogeneity, however, as there is currently no evidence to indicate whether migraine and other headache conditions respond equivalently or not to psychological interventions, and several studies include combined populations; removing these combined studies again likely reduced the power of the overall effects.

Our aim for the current review was to extend and update the Cochrane Review, broadening the range of studies investigating psychological treatments for adults with migraine, while maintaining the scientific rigor of the review and meta-analytical approach. Through this, we aimed to determine the overall efficacy of psychological interventions for adults with migraine in reducing migraine frequency, pain, and functional and psychosocial difficulties.

1 2 3 4

Next Section

Abstract and Introduction
Method
Results
Discussion
References

Table 1. Study characteristics of studies including adults with migraine only
#	First author	Year	N	Diagnosis	Aura	TTH	Psychological intervention	Duration	Setting	Delivery	Control
1	Bakhshani¹⁵	2016	37	ICHD	n.r.	Chronic	MBSR	8 weeks	F2F	Group	Pharmacotherapy
2	Basler⁴⁸	1996	88	ICHD	n.r.	n.r	CBT	12 sessions, 12 weeks	F2F	Group	Standard care
3	Bhombal^{a 49}	2014	90	ICD	n.r.	–	Behavioral	1 session	F2F	Individual	Pharmacotherapy
4	Blanchard⁵⁰	1990	46	Other	n.r.	Episodic or chronic	Biofeedback + Cognitive	5 sessions + 2 phone calls	F2F + Remote	Individual	Standard care
5	Bromberg^{a 51}	2012	144	ICHD	w/& w/o	–	Behavioral	8 sessions	Remote	Individual	Waitlist
6	Calhoun^{a 52}	2007	36	Other	n.r.	–	Behavioral	1 session	F2F	Individual	Waitlist
7	Cousins^{a 53}	2015	53	ICHD	w/& w/o	–	CBT	3 sessions + 2 phone calls	F2F + Remote	Individual	Standard care
8	Devineni⁵⁴	2005	86	Other	w/& w/o	Chronic	CBT	4 weeks	Remote	Individual	Symptom monitoring
9	Dindo⁵⁵	2020	103	Other	n.r.	–	ACT	1 session	F2F	Group	Psychoed + support
10	D'Souza^{a 56}	2008	90	ICHD	n.r.	–	Behavioral	4 sessions, 2 weeks	F2F	Individual	Time management
11	Flynn⁵⁷	2019	40	Other	n.r.	–	Hypnosis	4 weeks	Remote	Individual	Waitlist
12	Fritsche^{a 58}	2010	115	ICHD	w/& w/o	–	CBT	5 sessions	F2F	Group	Brochure
13	Grazzi⁵⁹	2021	35	ICD	w/out	–	ACT	8 sessions	F2F	Group	Standard care
14	Hedborg^{a 60}	2011	76	ICHD	w/& w/o	–	Behavioral	n.r	Remote	Individual	Massage + relaxation
15	Holroyd⁶¹	1988	37	Other	n.r.	Episodic or chronic	Biofeedback + relaxation	3 sessions + 2 phone calls, 6 weeks	F2F + Remote	Individual	Pharmacotherapy
16	Holroyd^{a 62}	2010	86	ICHD	w/& w/o	–	Behavioral	4 sessions	F2F	Individual	Placebo pill
17	Kang^{a 63}	2009	32	ICHD	w/& w/o	–	Biofeedback	8 sessions, 4 weeks	F2F	Individual	Monitoring
18	Kaushik^{a 64}	2005	192	ICHD	w/& w/o	–	Biofeedback + relaxation	10 sessions	F2F	Individual	Pharmacotherapy
19	Kleiboer^{a 65}	2014	368	ICHD	w/& w/o	–	Behavioral	8 lessons	Remote	Individual	Standard care
20	Kohlenberg^{a 66}	1981	51	Other	n.r.	–	CBT + biofeedback	10 weeks	Remote	Individual	Standard care
21	Mahmoudzadeh-Zarandi^{a 31}	2016	83	ICHD	w/& w/o	–	Behavioural	4 sessions	F2F	Both	Standard care
22	Marcus^{a 67}	2008	43	ICHD	n.r.	–	EMDR	1 session	F2F	Individual	Standard care
23	Martin⁶⁸	1989	55	Other	n.r.	Episodic or chronic	CBT	12 sessions, 12 weeks	F2F	Group	Self-management training
24	Martin⁴⁰	2014	48	ICHD	w/& w/o	Episodic or chronic	Behavioral	8 sessions	F2F	Individual	Avoidance training
25	Merelle^{a 69}	2008	108	ICHD	w/& w/o	–	Behavioral	7 sessions, 10 weeks	Remote	Group	Standard care
26	Meyer^{a 70}	2016	35	ICHD	w/& w/o	–	Behavioral	6 sessions, 6 weeks	F2F	Group	Waitlist
27	Minen⁷¹	2020	139	ICHD	n.r.	–	Behavioral	Avg 11 mins over 19 days	Remote	Individual	Standard care
28	Minen⁷²	2021	50	ICHD	n.r.	–	Biofeedback	Avg 29 sessions over 36 days	Remote	Individual	Waitlist
29	Mullally²⁹	2009	50	ICHD	n.r.	Episodic	Biofeedback + behavioral	10 sessions	F2F	Individual	Relaxation
30	Rashid-Tavalai^{a 73}	2016	35	ICHD	n.r.	–	Behavioral	7 sessions	F2F	Group	Pharmacotherapy
31	Richardson^{a 34}	1989	47	Other	n.r.	–	CBT	8 sessions, 8 weeks	F2F	Individual	Waitlist
32	Rothrock^{a 74}	2006	100	ICHD	n.r.	–	Psychoeducation	3 sessions	F2F	Group	Standard care
33	Sargent^{a 75}	1986	136	Other	n.r.	–	Biofeedback	22 sessions, 36 weeks	F2F	Individual	Standard care
34	Seminowicz⁷⁶	2020	98	ICHD	w/& w/o	–	MBSR	12 sessions, 16 weeks	F2F	Group	Stress management
35	Seng⁷⁷	2019	60	ICHD	n.r.	–	MCBT	8 sessions, 8–10 weeks	F2F	Individual	Standard care
36	Simshauser⁷⁸	2019	44	ICD	w/& w/o	–	MBSR	8 sessions, 8 weeks	F2F	Group	Psychoed + relaxation
37	Tehrani⁷⁹	2021	86	n.r	n.r.	–	Biofeedback	10 sessions, 5 weeks	F2F	Individual	Pharmacotherapy
38	Wachholtz⁸⁰	2008	42	ICHD	n.r.	–	Behavioral	1 group session + 4 weeks of practice	F2F	Both	Relaxation
39	Wells⁸¹	2021	89	ICHD	w/& w/o	–	MBSR	8 sessions, 8 weeks	F2F	Group	Headache education

Note: (–) = data not available as study included data for participants with migraine only.
Abbreviations: CBT, cognitive behavioral therapy; EMDR, eye movement desensitization and reprocessing; F2F, face to face; ICHD, International Classification of Headache Disorders; MBCT, mindfulness-based cognitive therapy; MBSR, mindfulness-based stress reduction; n.r., not reported; w/& w/o, included patients with migraine with and without aura.
^aDenotes study was also included in the Cochrane Review.³³

Table 2. Risk of bias assessment for all included studies
	Selection bias		Detection bias	Attrition bias	Reporting bias	Other methodological bias		Total score^a
	Random sequence generation	Allocation concealment	Blinding of outcome assessment	Complete outcome data	Selective reporting	Treatment integrity	Adequate sample size	Low = 0; Unclear/High = 1
Bakhshani 2016¹⁵	Unclear	Unclear	Unclear	Low	Unclear	High	High	6
Basler 1996⁴⁸	Unclear	Unclear	Unclear	High	Unclear	High	High	7
Bhombal 2014⁴⁹	Low	Low	High	High	Unclear	Low	High	4
Blanchard 1990⁵⁰	Low	Unclear	Unclear	High	Unclear	High	High	6
Bromberg 2012⁵¹	Low	Unclear	Unclear	High	Unclear	Low	Unclear	5
Calhoun 2007⁵²	Low	Unclear	High	High	Unclear	High	High	6
Cousins 2015⁵³	Low	Low	Low	High	Unclear	High	High	4
Devineni 2005⁵⁴	Unclear	High	Low	High	Unclear	Low	High	5
Dindo 2020⁵⁵	Low	Low	Low	High	Low	Low	Unclear	2
D'Souza 2008⁵⁶	Low	Low	Low	Low	Unclear	Low	High	2
Flynn 2019⁵⁷	Low	Low	Low	High	Unclear	High	High	4
Fritsche 2010⁵⁸	Low	Unclear	Unclear	High	Unclear	Low	Unclear	5
Grazzi 2021⁵⁹	Low	Low	Low	Low	High	High	High	3
Hedborg 2011⁶⁰	Low	Unclear	Low	Unclear	Unclear	High	High	5
Holroyd 2010⁶¹	Low	Low	High	High	Low	Low	Unclear	3
Holroyd 1988⁶²	High	Unclear	Unclear	High	Unclear	High	High	7
Kang 2009⁶³	Unclear	Unclear	High	Unclear	Unclear	High	High	7
Kaushik 2005⁶⁴	Low	Low	Low	Low	Unclear	High	Unclear	3
Kleiboer 2014⁶⁵	Low	Low	High	High	Unclear	Low	Unclear	4
Kohlenberg 1981⁶⁶	Unclear	Unclear	Unclear	High	Unclear	Low	High	6
Mahmoudzadeh-Zarandi 2016³¹	Unclear	Unclear	Unclear	Low	Unclear	High	High	6
Marcus 2008⁶⁷	Unclear	Unclear	Low	High	Unclear	Low	High	5
Martin 1989⁶⁸	Unclear	Unclear	Unclear	High	Unclear	High	High	7
Martin 2014⁴⁰	Low	Low	Low	Unclear	Unclear	Low	High	3
Merelle 2008⁶⁹	Low	Unclear	Low	Low	Unclear	High	Unclear	4
Meyer 2016⁷⁰	High	Unclear	Low	High	Unclear	High	High	6
Minen 2020⁷¹	Low	High	High	High	Low	Low	Unclear	5
Minen 2021⁷²	Unclear	Unclear	Unclear	Low	High	Low	Unclear	5
Mullally 2009²⁹	Unclear	Unclear	Low	Low	Unclear	High	High	5
Rashid-Tavalai 2016⁷³	Unclear	Unclear	Unclear	High	Unclear	High	High	7
Richardson 1989³⁴	Unclear	Unclear	Unclear	Unclear	Unclear	High	High	7
Rothrock 2006⁷⁴	Unclear	Unclear	Low	Low	Unclear	High	Unclear	5
Sargent 1986⁷⁵	Low	Low	Unclear	High	Unclear	High	High	5
Seng 2019⁷⁷	Low	Low	High	Low	High	Low	High	3
Seminowicz 2020⁷⁶	Low	Low	Low	Low	Low	Low	High	1
Simshauser 2019⁷⁸	Low	Low	Unclear	High	Unclear	High	Unclear	5
Tehrani 2021⁷⁹	Unclear	Unclear	Low	High	Unclear	High	High	6
Wachholtz 2008⁸⁰	Low	Low	Unclear	Low	Unclear	High	Unclear	4
Wells 2021⁸¹	Low	Low	Low	Low	High	High	High	3
Low risk	k = 23	k = 16	k = 17	k = 13	k = 4	k = 15	k = 0
High risk	k = 2	k = 2	k = 7	k = 22	k = 4	k = 24	k = 12
Unclear	k = 14	k = 21	k = 15	k = 0	k = 31	k = 0	k = 27

^aScores are out of a possible of 7, with higher scores indicating a higher risk of bias.

Table 3. Summary of findings and GRADE ratings
Outcomes	Anticipated absolute effectse (95% CI)		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
Outcomes	Risk with control	Risk with psychological intervention	Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
Migraine frequency		Mean 0.28 fewer (0.12 fewer to 0.44 fewer)	–	1402 (16 RCTs)	⊕○○○ VERY LOW^a,b,c	Psychological intervention may result in a reduction in migraine frequency, post-treatment
Treatment responders	240 per 1000	450 per 1000	RR 1.82 (1.19, 2.80)	435 (6 RCTs)	⊕○○○ VERY LOW^a,c,d	Adults who received psychological intervention were more likely to have a 50% or more reduction in migraine frequency, post-treatment
Pain intensity		Mean 0.31 lower (0.11 lower to 0.51 lower)	–	1105 (12 RCTs)	⊕○○○ VERY LOW^a,b,c	Psychological intervention may reduce migraine intensity, post-treatment
Mood		Mean 0.31 lower (0.63 lower to 0.01 higher)	–	733 (10 RCTs)	⊕○○○ VERY LOW^a,c,d,e	There was no evidence on the benefit or harm on mood, post-treatment
Disability		Mean 0.33 lower (0.16 lower to 0.49 lower)	–	1663 (15 RCTs)	⊕○○○ VERY LOW^a,c,e	Psychological intervention may reduce pain-related disability, post-treatment
Quality of life		Mean 0.23 lower 0.47 lower to 0.02 higher)	–	1008 (10 RCTs)	⊕○○○ VERY LOW^a,c,d,e	There was no evidence of benefit or harm on quality of life, post-treatment
Medication usage		Mean 0.30 lower (0.05 lower to 0.55 lower)	–	247 (4 RCTs)	⊕○○○ VERY LOW^a,c,d	Psychological Intervention may reduce medication usage, post-treatment

Note: GRADE Working Group grades of evidence: High: We are very confident that the true effect lies close to that of the estimate of the effect; Moderate: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different; Low: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect; Very low: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.
Abbreviations: CI, confidence interval; RCT, randomized controlled trial; RR, risk ratio.
^aThe majority of studies had unclear or high risk of bias.
^bHeterogeneity (I ²) was 46% to 65%.
^c70%–100% of studies eligible for inclusion did not report data for this outcome.
^dThis outcome had a low overall sample size.
^eHeterogeneity (I ²) was 66% to 100%.
*The risk in the intervention group is based on the assumed risk in the comparison group and the relative effect of the intervention.

Table 4. Sensitivity analysis on post-treatment outcomes for all studies
Outcome	k	d (95% CI) or RR (95% CI)	p	I ² (%)
Headache frequency	17	d = 0.25; 95% CI = 0.09, 0.41	<0.001	60
Treatment responders	9	RR = 1.98; 95% CI = 1.35, 2.92	0.001	55
Pain intensity	15	d = 0.34; 95% CI = 0.15, 0.53	<0.001	64
Mood	13	d = 0.32; 95% CI = 0.07, 0.56	0.011*	70
Pain-related disability^a	17	d = 0.33; 95% CI = 0.18, 0.48	<0.001	55
Medication usage	9	d = 0.36; 95% CI = 0.18, 0.53	<0.001	0

Abbreviations: CI, confidence interval; RR, risk ratio.
^aEffect was non-significant in the main analyses of studies with migraine only.
^bWith outlier removed.

processing....

Are Psychological Interventions Efficacious for Adults With Migraine?

A Systematic Review and Meta-Analysis

Abstract and Introduction

Abstract

Introduction

Tables

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Most Popular Articles

Email This

Feedback