CURRENT LITERATURE: CLINICAL SCIENCE

Headache Currents

Stress Management for Migraine:

Recent Research and Commentary head_1478 1395..1398

Donald B. Penzien, PhD

samples, improved reporting).3 In addition, efforts to develop

Cognitive-behavioral therapy versus temporal pulse
and validate an algorithmic approach to prescribing behavioral
amplitude biofeedback training for recurrent headache
headache treatment have lagged owing to insufficient evidence
Martin PR, Forsyth MR, Reece J. Behav Ther.
about which headache patients are responsive to behavioral thera-
2007;38:350-363.
pies and under what circumstances.
Abstract
With the research abstracted above, Martin et al4 have made a
Sixty-four headache sufferers were allocated randomly to
solid contribution to the literature addressing behavioral head-
cognitive-behavioral therapy (CBT), temporal pulse
ache treatments. They applied current methodological standards
amplitude (TPA) biofeedback training, or waiting-list
in gathering new evidence pertaining to the effectiveness of a
control. Fifty-one participants (14M/37F) completed the
well-established behavioral headache therapy (cognitive-
study, 30 with migraine and 21 with tension-type headache.
behavioral therapy) as well as a long extant but less extensively
Treatment consisted of 8, 1-hour sessions. CBT was highly
studied behavioral approach (temporal pulse amplitude biofeed-
effective, with an average reduction in headaches from pre-
back). The treatment gains for both conditions were sizeable (eg,
to posttreatment of 68%, compared with 56% for
77.8% and 63.2% of patients demonstrating a clinically mean-
biofeedback, and 20% for the control condition. Headaches
ingful headache reduction at posttreatment vs 23.1% of con-
continued to decrease to 12-month follow-up for CBT.
trols). In addition, the authors examined a variety of potential
Improvement with CBT was associated with baseline coping
mediators and predictors of treatment outcome, and they suc-
skills, social support, and physiological measures at rest and
ceeded in identifying several psychological and psychophysiologi-
in response to stress, particularly TPA. Changes on some of
cal correlates of treatment outcome – although the observed
these measures were correlated with changes in headaches.
correlates were not necessarily the obvious ones. The latter evi-
No significant predictors of response to biofeedback
dence may well make a contribution to future efforts to develop
emerged.
algorithms for optimally prescribing headache therapies. While
this trial examined a larger sample (n = 64) than many of the
Commentary
studies published to date, the modest sample size per condition
A sizeable evidence base has established the effectiveness of
nevertheless precluded meaningful statistical analysis of a
behavioral treatments for migraine; to a greater or lesser extent,
number of key questions. Until such time as research funding is
all of the established behavioral headache treatments focus upon
made more readily available for headache research (including
management of headache-related stress and stress responding.1
public and private sources), modest sample sizes unfortunately
With the overwhelming majority of published trials yielding
are likely to be the order of the day for behavioral headache
positive outcomes, the evidence long ago led many professional
research.
organizations to endorse behavioral interventions alongside phar-
macotherapy for treatment of primary headaches (including the
Headache Treatment Guideline Consortium).2 Migraine patients as trainers of their fellow patients in
Nevertheless, much remains to be learned about the behavioral non-pharmacological preventive attack management:
factors influencing headache and behavioral approaches to head- Short-term effects of a randomized controlled trial
ache therapy. This includes the need for replication and extension Mérelle SYM, Sorbi MJ, van Doornen LJ, Passchier J.
of the published research using updated standards and more Cephalalgia. 2008;28:127-138.
rigorous methodology (eg, current diagnostic standards, larger Abstract
In conformity with current views on patient empowerment,
From the University of Mississippi Medical Center, Jackson, MS, USA. we designed and evaluated the effects of home-based
Address all correspondence to D.B. Penzien, University of Mississippi Medical Center, PO Box
behavioral training (BT) provided by lay trainers with
4772, Jackson, MS 39296-4772, USA. migraine to small groups of fellow patients. The primary
Accepted for publication April 21, 2009.
.............
Headache
© 2009 the Author .............
Journal compilation © 2009 American Headache Society Conflict of Interest: None.

1395
 Headache Currents
1396 | Headache | October 2009

aims of BT were to reduce attack frequency and increase Conclusion.—Lay trainers with migraine can train small
perceived control over and self-confidence in attack groups of fellow patients at home in behavioral attack
prevention. In a randomized controlled trial, the BT group prevention. At 6-month follow-up, attack frequency and
(n = 51) was compared with a waitlist-control group quality of life were significantly but modestly improved and
(WLC), receiving usual care (n = 57). BT produced a minor feelings of control and self-confidence remained strongly
(-21%) short-term effect on attack frequency and clinically improved.
significant improvement in 35% of the participants.
Covariance analysis showed a nonsignificant trend Commentary
(P = 0.07) compared with WLC. However, patients’ Traditionally, behavioral headache therapies have been adminis-
perceived control over migraine attacks and self-confidence tered in clinical settings by professional therapists in weekly,
in attack prevention increased significantly with large effect face-to-face, individual therapy sessions. But this approach to
sizes. Patients with high baseline attack frequency might headache therapy presents drawbacks for many sufferers (eg, cost
benefit more from BT than those with low attack and inconvenience for patients, lack of trained behavioral thera-
frequency. In conclusion, lay trainers with migraine pists, poor reimbursement for mental health services). As a means
strengthened fellow patients’ perceived control, but did not of overcoming these obstacles, researchers in the 1980s began to
induce a significant immediate improvement in attack systematically examine alternate formats for delivery of behav-
frequency. ioral treatment. The most extensively studied alternate format
has been the minimal-therapist contact or “home-based” formats
Lay trainers with migraine for a home-based behavioral that provide treatment components similar to their clinic-based
training: A 6-month follow-up study counterparts, with the evidence consistently indicating the 2
Mérelle SYM, Sorbi MJ, van Doornen LJ, Passchier J. treatment formats yield similar outcomes.
Headache. 2008;48:1311-1325. Considerably less attention has been devoted to development
Abstract of headache therapies as administered by nonprofessionals. As an
Objective.—To evaluate the changes at 6-month follow-up alternative or supplement to therapy for chronic illness by a
after a home-based behavioral training (BT) provided by lay health professional, the self-management literature for many
trainers with migraine to small groups of fellow patients. years has emphasized use of therapy groups led by trained and
Background.—The need for self-management programs and professionally supervised lay leaders for disorders such as arthri-
cost-effective treatments gave rise to this study. tis, asthma, and diabetes. But parallel work in the headache
Methods.—In a previous randomized controlled trial, we arena has only recently begun with encouraging the initial
compared the BT group with a waitlist-control group, results.5,6
receiving usual care. The control group was trained directly In 2006, Mérelle et al from the Netherlands published the
after their waitlist period. The present study examined the first in a series of reports presenting their efforts to design and
follow-up results in both groups and measurements were evaluate a home-based behavioral self-management training
held pre BT, post BT, and at 6-month follow-up. program administered by nonprofessional trainers with
Results.—Six months after BT, 42% was categorized as migraine to small groups of fellow patients. With encouraging
responders (ⱖ-50% decrease in attack frequency), 42% did results from their uncontrolled pilot study,5 Mérelle et al initi-
not change (-49 to 49%), and 16% responded adversely ated a randomized, controlled trial to further examine this
(ⱖ50% increase). In the group as a whole (n = 95), attack therapeutic regimen; the findings of the latter trial along with a
frequency significantly decreased from 3.0 attacks at second report presenting 6-month follow-up data are abstracted
baseline to 2.5 post BT and to 2.3 at 6-month follow-up above.7,8 In addition, these authors have now published a fourth
(-23%, medium effect size 0.6). The strong improvements paper in this program of research, with a report of their study
of perceived control over and self-confidence in attack examining the personal characteristics of 13 nonprofessional
prevention were maintained at follow-up. Disability and trainers and how those characteristics relate to self-management
health status were unchanged but quality of life significantly outcomes.9
improved over time (P = 0.007). BT was more beneficial for At the initial outcome assessment, attack frequency was mod-
patients who entered the training with a high attack estly reduced by the lay-led self-management program (21%
frequency. Linear regression analysis demonstrated that a decrease), with a clinically meaningful improvement (ⱖ50%
stronger belief at baseline that the occurrence of migraine is reduction in attack frequency) reported for 35% of treated
due to chance (external control) significantly predicted a patients. On the other hand, patients’ perceived control over
lower attack frequency at follow-up. migraine and their sense of self-efficacy for preventing migraine
attacks increased substantially at posttreatment. By their
 Headache Currents
1397 | Headache | October 2009

6-month follow-up, not only were the enhancements in per-

treatments, to identify moderators of efficacy, and to
ceived control and self-efficacy well maintained, but quality of
determine effects of treatment on other health-related
life and headache frequency measures demonstrated continuing
variables such as quality of life.
gains (31% reduction on the Migraine Specific Quality of Life
Questionnaire; 42% of patients reported clinically meaningful Feasibility assessment of telephone-administered
reduction in headache frequency). The authors observed that behavioral treatment for adolescent migraine
their lay-led therapy was more beneficial for patients who initi- Cottrell C, Drew J, Gibson J, Holroyd K, O’Donnell F.
ated therapy with a “high attack frequency.” It is worthy of note Headache. 2007;47:1293-1302.
that the investigators opted to enroll only migraineurs with a Objectives.—To examine the feasibility of administering
baseline attack frequency of 1-6 headaches per month, and they behavioral migraine management training by telephone
excluded patients with headache more than 15 days/month. (TAT) and the acceptability of TAT to adolescents with
While excluding patients with higher attack frequency likely episodic migraine.
excluded many among the most refractory headache patients (eg, Methods.—A total of 34 adolescents (M = 14 years) with
those with chronic migraine), the relatively low attack frequency migraine (M = 3.6 migraines/month; M = 29.2-hour
in this sample may well have presented a “floor effect” making it duration) were randomly assigned to a 2-month telephone
difficult to meaningfully demonstrate treatment gains. Given administered behavioral migraine management program
these promising findings, future research is encouraged with the (TAT) or to a standard Triptan Treatment (TT). Outcome
hopes that investigators also will enroll migraineurs who experi- was assessed at 3- and 8-month evaluations. Participants
ence more frequent migraine. Doing so not only will overcome completed a daily migraine diary that yielded information
the potential floor effect, but also will enhance generalizability of about number, duration and severity of migraines and
these findings to patients with chronic migraine – a very migraine-related disability, as well as the Migraine Specific
common constituent of headache specialty services. Quality of Life Questionnaire – Adolescent. In addition,
TAT participants evaluated key aspects of the TAT program
Psychological treatment of recurrent headache in using 5-point Likert-like rating scales. Lastly, the ability of
children and adolescents – A meta-analysis adolescents to demonstrate specific headache management
Trautmann E, Lackschewitz H, Kröner-Herwig B. skills following TAT was assessed.
Cephalalgia. 2006;26:1411-1426. Results.—All 15 adolescents who entered TAT successfully
Abstract demonstrated either full or partial mastery of 2 or more
Psychologically based interventions such as relaxation skills and nearly half demonstrated at least partial mastery
training, biofeedback, and cognitive-behavioral therapy are of all 4 skills evaluated. Some 93% of the TAT participants
increasingly discussed as options for the treatment of reported having a positive relationship with their phone
migraine and tension-type headache in children and counselor. They also reported a preference for the
adolescents. In order to determine the state of evidence telephone-based treatment over in-clinic visits and rated the
regarding the efficacy of these treatments, a meta-analysis of manual and tapes as helpful. Treatment effects (in terms of
randomized controlled studies was conducted. In a percent improvement) ranged from consistently large across
comprehensive literature search including data from 1966 to both evaluations for improvement in number of migraines
2004, 23 studies were found meeting the inclusion criteria. (54% and 71%), disability equivalent hours (80% and
Due to the application of the random effects model, 63%), and quality of life (44% and 48%), to moderate or
generalization of the results is possible. Specific statistical variable for migraine duration (35% and 23%) and severity
procedures were used to account for a possible publication (30% and 34%). The TT group also showed clinically
bias. Significantly more patients improved to a clinically meaningful reductions in headache parameters and
relevant extent (headache reduction ⱖ50%) in treatment improvements in quality of life.
conditions compared with waiting-list conditions (high Conclusions.—Completion rates for TAT were high;
effect sizes). Long-term stability was also confirmed. The adolescents evaluated their experience with TAT positively
analyzed treatments lead to improvement (up to 1 year) in and were able to exhibit key behavioral headache
headache status in children and adolescents with primary management skills following treatment. While clinically
headache. However, more well-designed studies are needed significant improvements in migraine and migraine-related
to support and consolidate the conclusions of this disability/quality of life were observed with both TAT and
meta-analysis and to compare the effects of psychological treatment as usual (triptan therapy), the small study size
treatment with those of prophylactic medical interventions and the absence of a control group do not permit
(in migraine), to examine potential differences between conclusions about the effectiveness of either treatment.
 Headache Currents
1398 | Headache | October 2009

delivery of behavioral therapy is feasible and effective, but patient

Nonetheless, these results indicate TAT may be a promising
reports indicated that the substantial majority of patients would
treatment format for improving access to behavioral
prefer this treatment strategy to face-to-face contact with their
treatments for underserved adolescents and justifies further
therapist (80%) and perceived that they had an effective collabo-
evaluation of TAT both alone and in combination with
rative relationship with their counselor (90%). Hopefully, this
drug therapy. study and the work by Mérelle et al will encourage continued
innovation in the delivery of behavioral self-management thera-
Commentary pies to primary headache sufferers.
The evidence pertaining to the effectiveness of behavioral self-
management therapies for children and adolescents is more
limited than it is for adults. Nevertheless, some reviewers have References
concluded not only that these therapies can prove beneficial for 1. Rains JC, Penzien DB, McCrory DC, Gray RN. Behavioral head-
young headache patients, but that behavioral treatments may be ache treatment: History, review of the empirical literature, and
even more effective for children and adolescents than for adults. methodological critique. Headache. 2005;45:S91-S108.
2. Campbell JK, Penzien DB, Wall EM. Evidence-based guidelines for
Based upon their carefully designed, executed, and reported
migraine headaches: Behavioral and physical treatments. 2000.
meta-analysis of this literature, Trautmann et al10 drew that same Available at: http://www.aan.com/professionals/practice/pdfs/
conclusion. They reported, for example, that behavioral treat- gl0089.pdf (accessed May 29, 2009)
ment yielded a large treatment effect (ES = 0.87) that corre- 3. Penzien DB, Andrasik F, Freidenberg BM, et al. Guidelines for
sponded to a clinically meaningful improvement in 70% of trials of behavioral treatments for recurrent headache. Headache.
patients vs only 30% of controls, and that these treatment gains 2005;45:S109-S131.
are well maintained for up to 1-year posttreatment. They 4. Martin PR, Forsyth MR, Reece J. Cognitive-behavioral therapy
reported, however, that despite a relatively large number of ini- versus temporal pulse amplitude biofeedback training for recurrent
tially identified trials, only a small number (23) were eligible for headache. Behav Ther. 2007;38:350-363.
inclusion in this meta-analysis. In addition, many of the pub- 5. Mérelle SYM, Sorbi MJ, Passchier J. The preliminary effectiveness
lished trials have enrolled only small samples. Thus, Trautmann of migraine lay trainers in a home-based behavioural management
training. Patient Educ Couns. 2006;61:307-311.
et al10 have made the essentially requisite call for more and better
6. Rothrock JF, Parada VA, Sims C, Key K, Walters NS, Zweifler RM.
designed trials to support their conclusions. The impact of intensive patient education on clinical outcome in a
Fortunately, a noteworthy recent trial published by Cottrell clinic-based migraine population. Headache. 2006;46:726-731.
et al11 not only adds valuable evidence in this arena, but also 7. Mérelle SYM, Sorbi MJ, van Doornen LJ, Passchier J. Migraine
examines application of an alternative format for delivery of patients as trainers of their fellow patients in non-pharmacological
behavioral treatments to adolescent patients. They randomized preventive attack management: Short-term effects of a randomized
34 adolescents either to an 8-week telephone-supervised home- controlled trial. Cephalalgia. 2008;28:127-138.
based headache treatment program or to triptan therapy. The 8. Mérelle SYM, Sorbi MJ, van Doornen LJ, Passchier J. Lay trainers
telephone-administered therapy fared well, yielding 54% reduc- with migraine for a home-based behavioral training: A 6-month
tion in migraines at the 3-month follow-up and 79% reduction follow-up study. Headache. 2008;48:1311-1325.
at the 8-month follow-up, with 80% of patients showing a clini- 9. Mérelle SYM, Sorbi MJ, Duivenvoorden HJ, Passchier J. Qualities
and health of lay trainers with migraine for behavioral attack pre-
cally meaningful reduction in headache (ⱖ50% decrease in
vention. Headache. 2008;48:1311-1325.
attack frequency) at both follow-up assessments. The triptan 10. Trautmann E, Lackschewitz H, Kröner-Herwig B. Psychological
therapy likewise produced clinically meaningful improvements treatment of recurrent headache in children and adolescents – A
within the same range demonstrated by patients in the behavioral meta-analysis. Cephalalgia. 2006;26:1411-1426.
therapy condition; the small sample size did not permit mean- 11. Cottrell C, Drew J, Gibson J, Holroyd K, O’Donnell F. Feasibility
ingful statistical comparisons between the groups, however. This assessment of telephone-administered behavioral treatment for
study demonstrated that not only this alternate format for adolescent migraine. Headache. 2007;47:1293-1302.