250 Headaches
2.Severity of attacks, rated on either (a) a 4-point scale,
where 0 no headache, 1 mild headache (allowing
normal activity), 2 moderate headache (disturbing but
not prohibiting normal activity, bed rest is not necessary),
and 3 severe headache (normal activity has to be dis-
continued, bed rest may be necessary) or (b) a visual ana-
logue scale, wherein one end is anchored as •noneŽ and
the other as •very severe.Ž
3.Headache duration in hours.
4.Responder rate„the number or percentage of patients
achieving a reduction in headache days or headache dura-
tion per day that is equal to or greater than 50%. (This is
in accord with the recommendations of Blanchard &
Schwarz, 1988.)
Several modi“cations to the intensive, hourly recording
format have been proposed in order to improve adherence
and accuracy. Epstein and Abel (1977) directly observed in-
patients and noticed that most did not record continuously;
rather, they periodically omitted recordings and completed
them later by recall. Their modi“ed procedure asked patients
to make ratings only four times per day: wakeup/breakfast,
lunch, evening meal, and bedtime. These events tend to occur
at fairly regular times that are easily discriminated.
Although a time-sampling format (such as four times per
day) is less demanding for patients and is likely to yield more
reliable and valid data, it has certain shortcomings. In this ap-
proach, it is not possible to obtain true measures for headache
frequency and duration. Chronic or unwavering pain lends
itself quite nicely to either format, but the clinician might want
to make alterations for people with infrequent, but discrete,
prolonged migraine attacks. In the latter case, the patient
could make ratings repeatedly throughout an attack or, alter-
natively, could note the time of onset and offset and then per-
form a single rating of peak headache intensity. This would
allow the therapist to keep track of all key parameters. We
have used this procedure successfully with pediatric mi-
graineurs (Andrasik, Burke, Attanasio, & Rosenblum, 1985).
If patients resist recording on multiple occasions throughout
the day, then a single recording at the end of the day is most ad-
visable. Occasionally a patient•s symptoms will display •reac-
tivityŽ when being recorded systematically and worsen be-
cause of this increased symptom focus. These reactions are
typically shortlived.
A critical concern with any type of daily monitoring record
is the level of patient adherence. In an analog sample of col-
lege students, approximately 40% of subjects evidenced
some degree of nonadherence. The most common form of
noncompliance involved subjects recalling and completing
ratings at a later time (Collins & Thompson, 1979). Review-
ing pain records regularly, socially praising efforts to comply
(yet refraining from punishing noncompliance), anticipating
problem areas, and having the patient mail records to the
of“ce when gaps between appointments are large may help
emphasize the importance of and facilitate accurate record
keeping (Lake, 2001).
It is common for clinicians to have their patients monitor
headaches on a systematic basis during treatment but to con-
duct follow-up evaluations by interview or questionnaire
completion. Several studies have examined correspondence
between these two approaches: prospective, daily monitoring
versus retrospective, global determinations (Andrasik &
Holroyd, 1980a; Andrasik et al., 1985; Cahn & Cram, 1980).
Very different results emerge, with the latter believed to yield
biased overestimates of improvement. Clinicians and re-
searchers alike need to be aware of the potential for bias
when it is necessary to alter measures midstream and not be
lulled into uncritical acceptance of global reports of bene“t
that might be in”ated.Supplementary ApproachesA number of supplementary and alternative approaches have
been developed to assess headaches, and these are reviewed
in greater depth in Andrasik (2001a). Four approaches may
be easily adopted by practitioners and researchers:1.Measurement of multiple aspects of pain, speci“cally
affective/reactive as well as sensory/intensity.
2.Social validation of patient improvement.
3.Measurement of pain behavior or behavior motivated by
pain, including medication consumption.
4.Impact on other aspects of functioning, such as general
health or overall quality of life, physical functioning, emo-
tional functioning, cognitive functioning, role function-
ing, and social well-being (see Andrasik, 2001a, 2001b;
Holroyd, in press).HEADACHE TREATMENTPharmacological TreatmentMost individuals will experience a headache from time to
time, yet few of these individuals seek regular treatment from
a health care provider, even when headaches are severe and
disabling (Mannix, 2001; Michel, 2000). More typically,
headaches are tolerated, treated symptomatically with over-
the-counter analgesics, or managed by •borrowingŽ pre-
scribed medications from friends and family members. When