The Clinical Journal of Pain

17:78–93 © 2001 Lippincott Williams & Wilkins, Inc., Philadelphia

International Headache Society Headache Diagnostic Patterns in

Pain Facility Patients
*†‡§David A. Fishbain, M.Sc., M.D., *§Robert Cutler, Ph.D.,
§Brandly Cole, Psy.D., †‡§Hubert L. Rosomoff, M.D., D.Med.Sc., and
†§Renee Steele Rosomoff, B.S.N., M.B.A., C.R.C., C.L.R.S., C.R.R.N.
Departments of *Psychiatry, †Neurological Surgery, and ‡Anesthesiology, University of Miami, School of Medicine, and
§Comprehensive Pain and Rehabilitation Center at South Shore Hospital, Miami Beach, Florida, U.S.A.
Downloaded from https://journals.lww.com/clinicalpain by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3tjcLwhL8g9YAVhcVNn6WC7XTyiviepf0yOAWZjNV2tg= on 07/23/2019

Abstract:
Objective: Previous studies have indicated that many patients with chronic pain
(PWCP) referred to pain facilities for the treatment of neck and/or low back pain
complain of associated headaches. The purpose of this study was to characterize the
nature of these headaches according to International Headache Society (IHS) headache
diagnostic criteria.
Design: In preparation for this study, a questionnaire that reflected IHS headache
diagnostic criteria was developed. All consecutive patients admitted to our pain facility
complaining of headache completed this questionnaire and received a physical and
neurologic examination focused on key aspects of headache. A headache interview was
also conducted, using the questionnaire as a question guide. All questionnaires were
entered in a computerized database, and IHS diagnoses were arrived at for each patient.
As many IHS diagnoses as possible were assigned to each PWCP as long as IHS
criteria were fulfilled. In addition, a frequency distribution for headache precipitants
and neck-associated symptoms was developed and evaluated by discriminant analysis
to determine the diagnostic value of these factors in relation to each IHS diagnostic group.
Setting: Pain facility (multidisciplinary pain center).
Patients: Consecutive PWCP.
Results: Of 1,466 PWCP, 154 (10.5%) were identified as suffering from severe
headache interfering with function. Of these, 55.8% indicated that their headaches
were related to an injury for which they were seeking treatment and 83.7% had neck
pain. Migraine headache represented the most common diagnostic group (90.3%), with
cervicogenic headache representing the second most common (33.8%). Of the total
group, 44.2% had more than one headache diagnosis, that is, there was overlap.
Cervicogenic headache patients had the greatest percentage of overlap (94.2%), with
migraine patients being second (68.3%). The most frequent headache precipitant was
mental stress, followed by neck position and activity/exercise. The migraine and cer-
vicogenic headache groups had a statistically significant greater number of neck-
associated symptoms when compared with the remaining patients. Of the total head-
ache group, 74.6% complained that they had a tender point at the back of their neck.
Cervicogenic, migraine, and tension PWCP had the greatest frequency of head or neck
tender points. The discriminant analysis for neck-associated symptoms yielded the
following symptoms as the most common predictors of headache across IHS diagnos-
tic groups: clues to onset were severe headache beginning at the neck or tender point
and numbness in arms and legs; headache brought on by neck position and arms
overhead; and neck symptoms consisting of a tender point in the neck and feeling
severe headache in the neck.
Conclusions: Headache can and should be considered a frequent comorbid condi-
tion in PWCP. Because of the overlap data, more precise diagnostic criteria may
be required to separate cervicogenic headache from migraine headache. Neck-
associated symptoms seem to be important even to those PWCP diagnosed with
migraine headache.
Key Words: Cervicogenic—Headache—IHS headache diagnostic criteria—
Migraine—Pain facility—Patients with chronic pain

Received March 10, 2000; revised July 17, 2000; accepted Septem-
ber 28, 2000.
Address correspondence and reprint requests to Dr. David A.
Fishbain, University of Miami, Comprehensive Pain and Rehabilitation
Center, 600 Alton Road, Miami Beach, FL 33139, U.S.A.
Pain facility clinicians have long noted that many pa-
tients with chronic pain (PWCP) suffering from neck
pain or low back pain (LBP) complain of headache. In an
early study, Fishbain et al.1 reported that of 48 PWCP

78
 IHS HEADACHE DIAGNOSTIC PATTERNS
with the primary pain complaint of cervical or neck pain,
almost 100% complained of associated headache. This
finding has recently been reinforced by a number of stud-
ies. In the first study, Lebbink et al.2 compared a group
of chronic headache sufferers with age- and sex-matched
controls for a history of neck pain. They found that in the
headache group, the intensity of the neck pain increased
significantly when the headache was present and that this
group had a higher prevalence of head or neck injury and
LBP.2 Similarly, Blau and MacGregor3 reported that of
50 migraine patients, 32 (64%) reported neck pain either
before or during their headache. Finally, in a large epi-
demiologic study of 18,105 people, Hasvold et al.4 found
an extremely strong association between the report of
headache and neck and shoulder pain. These three stud-
ies indicate that neck pain populations may contain sig-
nificant percentages of patients suffering from headache
or that headache populations may contain a significant
percentage of patients complaining of neck pain.
To complicate matters, there is also some evidence
that LBP patients may have significant headache prob-
lems. Three recent studies have presented evidence for
this association. In the first study, Ahern et al.5 demon-
strated that 61.4% of chronic LBP patients referred to a
pain facility reported headache in addition to LBP. Of
those with headache, only 24.8% reported a history of
headache before the development of LBP, and the rest
reported concurrent headache development as a sequela
of LBP. Ahern et al.5 did not specify what percentage of
the chronic LBP patients had cervical pain or the types of
headaches found in their sample. In the second study,
Egan and Betrus6 reported that of 54 chronic LBP pa-
tients referred to an outpatient pain facility, 40% com-
plained of muscle contraction headaches and 2% com-
plained of migraine headache. Here, headache diagnoses
were not made according to consistent criteria but were
obtained from a referral source. In the third study,
Duckro et al.7 isolated a group of 46 PWCP with “pri-
mary” LBP from 100 consecutive patients admitted to a
pain facility. Patients with chronic pain satisfied the cri-
teria for “primary” LBP if LBP was the initial complain;
if there was no injury/trauma history to the upper back,
midback, or neck; and if there was no history of simul-
taneous headache onset or exacerbation. It was found
that of the 46 PWCP, 52.2% developed headache and
15.2% developed a significant exacerbation of an exist-
ing headache condition at some point after (nonsimulta-
neously with) the onset of LBP. Thirteen (28.3%) of
these PWCP and headache met International Headache
Society (IHS) criteria8,9 for migraine without aura.
These studies indicate that headaches are commonly
found in neck pain or LBP PWCP, that these headaches
79
seem to begin after the onset of chronic pain, and that a
significant percentage of these headaches could be clas-
sified as migraines. These studies also raised a number of
questions, including what percentage of PWCP and
headache have neck pain and what is the actual distribu-
tion of IHS8,9 headache diagnoses (migraine with aura;
migraine without aura; and tension, cluster, and other
[cervicogenic, chronic daily] headache diagnoses) within
a sample of pain facility PWCP with headache? The
senior author (D.A.F.) designed a study to address these
research questions. The results of this study are presented
below.
METHODS
For data collection for this study, the senior author
(D.A.F.) developed a questionnaire (Appendix I). Ques-
tions were designed to reflect IHS diagnostic criteria8,9
for various types of headaches (migraine with aura, mi-
graine without aura, tension headache, and cluster head-
ache). In addition, questions were included that reflected
proposed criteria for cervicogenic headache.10 The rela-
tion between the IHS headache criteria, proposed cervi-
cogenic headache criteria, and questionnaire numbers are
presented in Appendices II through VI.
All consecutive patients admitted to the University of
Miami Comprehensive Pain and Rehabilitation Center
who complained of headache completed the above ques-
tionnaire (see Appendix I) on their initial day of admis-
sion. A headache history was also obtained.11 This was
followed by a physical and neurologic examination
focused on the key aspects of the physical examination
for headaches.11 Diagnostic tests such as blood tests, an
electroencephalogram, computed tomography, and mag-
netic resonance imaging were ordered if there were “red
flags”11 in the history or physical examination or because
of other indications. The neurologic and physical exami-
nations were performed by three clinicians working in-
dependently of each other. Thus, each patient received
three independent neurologic examinations and one com-
plete physical examination (including a neurologic com-
ponent). All possible causes of headaches, including par-
oxysmal hemicrania,12 were identified in this way.
A headache interview was then conducted by the se-
nior author (D.A.F.) with each patient, using the ques-
tionnaire as a question guide. All patient responses were
reviewed for correctness of response and to make sure
the patients understood the questions. If necessary, cor-
rections to the questionnaire were made. Special atten-
tion was paid to questions that reflected IHS and cervi-
cogenic headache diagnostic criteria (see Appendices II–
IV). The characteristics of headache (i.e., pain type[s],
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
80 FISHBAIN ET AL.
location, severity, associated symptoms, and aggravating
and ameliorating factors) were sought in detail. Severity
was graded based on answers to questionnaire questions
I through V. Patients with analgesic/ergotamine rebound
headache were identified, using the IHS functional cri-
teria for medication overuse.9 These included at least one
of the following for at least 1 month: simple analgesic
use (>1,000 mg of acetosalicylic acid/acetaminophen)
more than 5 days per week, combination analgesics
(more than three tablets per day) more than 3 days per
week, narcotics (more than one tablet per day) more than
2 days per week, and ergotamine use (1 mg administered
orally or 0.5 mg administered per rectum) more than 2
days per week. It is to be noted that patients were con-
sidered to have analgesic/ergotamine rebound headache
only if they used these medications specifically to con-
trol their headaches. Many of our PWCP used large
quantities of narcotics and other analgesic agents for
their neck pain and/or LBP.
All questionnaires were entered in a computerized da-
tabase. On this basis, PWCP were eliminated from fur-
ther analysis if they had indicated the following in their
responses: if PWCP did not complete the questionnaire
beyond question 3, indicating that they did not get severe
headaches or, alternately, if severe headache frequency
was seldom or never or several times a year (question 1
[1] and [2]), if headache severity was mild or not usually
severe (question 2 [1] and [2]), if severe headache fre-
quency was seldom or several times a year (question 3
[1] and [2]), if headache did not affect function (question
4 [1] and [2]), or if the patients were thought to have
analgesic/ergotamine rebound headache. Through this
procedure, 93 PWCP were eliminated from further
analysis, leaving 154 PWCP who were presumed to be
complaining of severe headaches interfering with func-
tion which were nonanalgesic/ergotamine rebound head-
aches. All further analyses were done on this group of
154 PWCP.
As specified previously, some of the questionnaire
questions represented IHS criteria8,9 for migraine with
aura, migraine without aura, tension or cluster head-
aches, and proposed criteria for cervicogenic headache10
(see Appendices II–VI). Employing these questions and
thereby the IHS criteria as well as the proposed cervico-
genic headache criteria, the PWCP were classified as
belonging to one of the following diagnostic groups: mi-
graine with aura, migraine without aura, tension head-
ache, cluster headache, and cervicogenic headache. In
addition, if a patient indicated that the headache “never
goes away” (question V [6]), he or she was identified as
having a chronic daily headache.13,14
To determine test-retest reliability of this method for
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
arriving at an IHS headache diagnosis, a subgroup (n ⳱
10) of these PWCP (N⳱154) were readministered the
questionnaire 48 hours after completing the first ques-
tionnaire. The reliability of whether patients were placed
in the same headache diagnostic group was then evalu-
ated using the ␬ statistic. The chance-corrected measure
is considered to be the standard index of dichotomous
agreement.15
In an attempt to delineate the relation between the IHS
headache diagnostic groups, the percentage of PWCP
within each diagnostic group fulfilling the criteria for
other diagnostic groups was determined (i.e., overlap).
The baseline demographic and clinical characteristics for
the seven headache diagnostic groups were then calcu-
lated. Means and SDs were calculated for the continuous
variables and frequencies, and proportions were calcu-
lated for the categoric variables.
We wanted to characterize the “no diagnosis” group
(those PWCP not fulfilling criteria for any diagnostic
group) in terms of the other six diagnostic groups. As
such, for the no diagnosis group, the percentage of pa-
tients not meeting each of the major diagnostic criteria in
the diagnostic groups was calculated. To determine the
most frequent headache precipitant within each IHS di-
agnostic group, a frequency distribution for precipitants
was developed. The IHS headache diagnostic groups
were compared by ␹2 for the presence of neck-associated
symptoms. Each diagnostic group was compared with
the remaining patients.
Finally, to better evaluate the diagnostic value of neck-
associated symptoms, these questionnaire items were
evaluated by discriminant analysis to determine their di-
agnostic value in relation to each diagnostic group. Three
variable sets were used for the discriminant analysis. Set
1 (clues to onset, four questions) comprised questions on
severe headache beginning at the neck or at a trigger
point and traveling up, neck pain getting worse, neck
pain traveling up to the skull, and numbness in arms and
legs. Set 2 (headache brought on, four questions) com-
prised questions on neck position, driving, arms posi-
tioned overhead, and lifting overhead. Set 3 (associated
neck symptoms, three questions) comprised questions on
neck pain, tender point at the back of the neck, and feel
of severe headache in the neck.
RESULTS
During the time of this study (August 1994–Sep-
tember 1997), 1,466 PWCP (pain duration longer than 6
months) were screened for the presence of headache. Of
these, 247 (16.8%) complained of headache either as a
primary complaint or in association with chronic neck
 IHS HEADACHE DIAGNOSTIC PATTERNS
pain or chronic LBP. Eighty-one of these PWCP did not
have severe headaches (did not go beyond question 3)
that interfered with function. One was thought to have
paroxysmal hemicrania. Eleven PWCP fulfilled the func-
tional definition for medication overuse9 and were using
these medications specifically for their headaches. As
such, they were given the diagnosis of analgesic/
ergotamine rebound headache. Thus, 154 PWCP had
headaches that interfered with function, and these pa-
tients were broken down to the following groups: mi-
graine with aura, migraine without aura, tension head-
ache, cluster headache, cervicogenic headache, chronic
daily headache (nonanalgesic/ergotamine rebound head-
ache), and no diagnosis.
Questionnaire test-retest reliability ␬ values for arriv-
ing at IHS headache diagnoses were as follows: tension
headache, 1.00; migraine with aura, 0.746; migraine
without aura, 0.870; cervicogenic headache, 1.00; cluster
headache, 1.00; and chronic daily headache, 1.00. These
results indicate a high or perfect agreement between the
initial and retest questionnaires for each of the diagnostic
categories. Of the 154 PWCP, 86 (55.8%) indicated that
the onset of the headache was related to an injury for
which they were seeking treatment. One hundred twenty-
nine of the 154 PWCP (83.7%) had neck pain. Of the
PWCP with neck pain, 67.5% related the beginning of
their neck pain to a head injury, neck injury, or back
injury.
Table 1 presents the diagnostic breakdown of the total
PWCP group (N ⳱ 154). Migraine headache was the
most common diagnosis, representing 90.3% of the
group. Cervicogenic headache represented the second
most common diagnostic group (33.8% of the total
group). Of the 154 patients with headache, only 86
(55.8%) had one headache diagnosis and did not overlap.
If the no diagnosis group was not included, only 64
(41.6%) patients had a headache diagnosis that did not
TABLE 1. Percentages of headache patients with chronic pain (N = 154) who fulfilled International Headache Society
diagnostic criteria and proposed cervicogenic headache criteria without overlap to other diagnostic groups and with overlap to
other diagnostic groups
Headache diagnostic Percentage of total group
category (N ⳱ 154)
Migraine with aura 85 (55.1%)
Migraine without aura 87 (56.5%)
Total migraine 139 (90.3%)
Tension headache 25 (16.2%)
Cluster headache 4 (2.6%)
Cervicogenic headache 52 (33.8%)
Chronic daily headache 12 (7.8%)
No diagnosis 25 (16.2%)
*These patients overlapped with the diagnosis of cervicogenic headache, which is not currently an official International Headache Society diagnosis.
81
overlap with other diagnostic groups. The cervicogenic
headache diagnosis had the greatest number of patients
who overlapped (94.2%), and the migraine diagnosis had
the second greatest number of patients who overlapped
(68.3%).
Table 2 presents specific overlap percentages of each
diagnostic group versus the other diagnostic groups. As
noted, the greatest percentage of overlap for each diag-
nostic category was as follows: for migraine with aura, it
was migraine without aura; for migraine without aura, it
was migraine with aura; for tension headache, it was
cervicogenic headache; for cluster headache, it was cer-
vicogenic headache; for cervicogenic headache, it was
migraine without aura; for chronic daily headache, it was
migraine without aura; for migraine both with and with-
out aura, it was cervicogenic headache; for both tension
headache and migraine, it was migraine with aura; and
for no diagnosis, it was cervicogenic headache.
Some of the demographic and headache-specific vari-
ables of the various IHS headache diagnostic groups are
presented in Table 3. The mean age of the diagnostic
groups in Table 3 varied from a high of 57.44 ± 16.86
years for the no diagnosis group to a low of 42.30 ±
12.28 years for the migraine without aura group. In gen-
eral, the no diagnosis, cluster headache, and chronic
daily headache groups were older, and the tension head-
ache, cervicogenic headache, and migraine groups were
younger. Headache duration varied from a high of 5.5 ±
1.91 years for the cluster headache group to a low of 3.52
± 2.20 years for the tension headache group. On a five-
point scale, chronic daily headache patients complained
of the severest headache, followed by the migraine with
and without aura group. The no diagnosis group seemed
to have the lowest severity of headache. As expected, the
chronic daily headache group reported the greatest fre-
quency of headache on a seven-point scale, followed
by the tension headache/migraine group and the
Percentage of total group
(N ⳱ 154)
Without overlap With overlap
19 (12.3%) 66 (42.9%)
25 (16.2%) 62 (40.3%)
44 (28.6%) 95 (61.7%)
15 (9.7%) 10 (6.5%)
2 (1.3%) 2 (1.3%)
3 (1.9%) 49 (31.8%)
0 (0%) 12 (7.8%)
22 (14.3%) 3* (1.9%)
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
82 FISHBAIN ET AL.

TABLE 2. Number and percentage of various International Headache Society headache criteria diagnostic groups who overlap
with other International Headache Society headache diagnostic groups
Chronic Migraine
Migraine Migraine Tension Cluster Cervicogenic daily with and
with aura without aura headache headache headache headache without aura
(N ⳱ 85) (N ⳱ 87) (N ⳱ 25) (N ⳱ 4) (N ⳱ 52) (N ⳱ 12) (N ⳱ 60)
Migraine with aura — 60 8 2 37 8 —
(69.0%) (32.0%) (50.0%) (71.2%) (66.7%)
Migraine without aura 60 — 4 0 38 9 —
(70.6%) (16.0%) (0%) (73.1%) (75.0%)
Tension headache 8 4 — 1 9 4 2
(9.4%) (4.6%) (25.0%) (17.3%) (33.3%) (3.3%)
Cluster headache 2 0 1 — 4 0 0
(2.4%) (0%) (4.0%) (7.7%) (0.0%) (0%)
Cervicogenic headache 37 38 9 4 — 4 30
(43.5%) (43.7%) (36.0%) (100.0%) (33.3%) (50.0%)
Chronic daily headache 8 9 4 0 4 — 7
(9.4%) (10.3%) (16.0%) (0%) (7.7%) (11.7%)

cervicogenic headache group. Of the various IHS diag-
nostic groups, the chronic daily headache patients fol-
lowed by the migraine with/without aura groups had the
greatest number of patients reporting that they woke up
with a severe headache. In reference to headache awak-
ening the PWCP, the chronic daily headache and mi-
graine with/without aura groups also had the greatest
percentages of patients reporting this symptom.
The male sex distribution within each headache diag-
nostic group was as follows: migraine with aura, 32%;
migraine without aura, 52%; migraine overlap, 25%; to-
tal migraine, 33%; tension headache, 47%; cervicogenic
headache, 23%; and no diagnosis, 23%. As can be seen,
total migraine had a male-to-female ratio of approxi-
mately 1:2. Cervicogenic headache had a male-to-female
ratio of approximately 1:3.
The following percentages of each headache diagnos-
tic group related the beginning of their headache to an
injury: migraine with aura, 67.5%; migraine without
aura, 67.1%; tension headache, 75%; cluster headache,
50%; cervicogenic headache, 72%; chronic daily
headache, 75%; total migraine, 69.6%; tension/ migraine
overlap, 80%; and no diagnosis, 36%. Overall, a large
percentage (55.5%) of patients attributed the beginning
of their headache to an injury. The tension headache/
migraine overlap, tension headache, chronic daily head-
ache, and cervicogenic headache groups had the highest
percentages of patients relating their headache to an in-
jury. The no diagnosis group had the lowest percentage
of patients relating their headache to an injury (36%).
Table 4 presents the percentage of patients in the no
diagnosis group who did not meet some IHS headache
diagnostic criteria within the major headache diagnostic
groups. As can be seen, the criteria most frequently not
met were the following: for migraine without aura, du-
ration; for migraine with aura, aura items; for tension
headache, no nausea or photophobia/phonophobia; for
cervicogenic headache, unilateral head pain; for cluster
headache, duration; and for chronic daily headache,
none.

TABLE 3. Demographic and headache specific variables for each International Headache Society headache diagnostic group
Migraine with
Migraine with Migraine Tension Cluster Cervicogenic Chronic daily and without
aura without aura headache headache headache headache aura
(N ⳱ 85) (N ⳱ 87) (N ⳱ 25) (N ⳱ 4) (N ⳱ 52) (N ⳱ 12) (N ⳱ 60)
Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean (SD) Mean
Age (years) 43.49 (12.98) 42.30 (12.28) 48.32 (15.60) 52.75 (16.32) 45.40 (13.60) 51.50 (15.48) 42.22
How often get headache 5.57 (1.09) 5.43 (1.15) 5.76 (1.05) 5.50 (1.00) 5.73 (0.97) 6.83 (0.39) 5.49
How severe is pain 3.62 (1.06) 3.79 (1.05) 3.40 (1.04) 3.00 (0.00) 3.65 (1.03) 4.33 (0.89) 3.82
How long have severe headache 4.10 (2.09) 4.08 (2.14) 3.52 (2.20) 5.50 (1.91) 4.10 (2.08) 4.50 (2.43) 4.19

N (%) N (%) N (%) N (%) N (%) N (%) N (%)

Awakens with severe headache 62 (75.6) 66 (79.5) 14 (60.9) 1 (25.0) 39 (78.0) 11 (100.0) 46 (80)
Severe headache awakens you 67 (79.8) 67 (78.8) 14 (60.9) 0 (0.0) 39 (75.0) 12 (100.0) 52 (88)

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

 IHS HEADACHE DIAGNOSTIC PATTERNS 83

TABLE 4. Proportion of the no diagnosis group who did the migraine without aura, migraine both with and with-
not meet International Headache Society headache out aura, and no diagnosis groups.
diagnostic criteria within the major headache
diagnostic groups
Table 6 presents the IHS headache diagnostic groups
compared by ␹2 for the presence of neck-associated
Diagnostic criteria within Proportion symptoms. As can be seen, the migraine with aura, mi-
Diagnostic group each diagnostic group meeting criteria
graine both with and without aura, and cervicogenic
Migraine without aura B Duration 33
C Specific characteristics 70 headache groups seemed to have a significantly greater
D Nausea or photo/phonophobia 78 number of neck-associated symptoms when compared
Migraine with aura B Aura items 63
Tension headache A Pain characteristics 93 with the remaining patients.
B No nausea or photo/phono 37 Percentages of patients in each diagnostic group who
C Pericranial muscle disorder 52
Cervicogenic headache 1 Neck involvement 56 complained that they had a tender point at the back of
3 Unilateral head pain 11 their head or neck was as follows: migraine with aura,
Cluster headache B Frequency 70
B Duration 11 78.9%; migraine without aura, 56%; migraine overlap,
C Signs 33 69.1%; tension headache overlap, 100%; tension head-
Chronic daily headache Chronic and severe 0
ache only, 100%; no diagnosis, 51.8%; and cervicogenic
headache, 96.1%. Overall, 115 patients (74.6%) com-
plained that they had a tender point at the back of their
Table 5 presents the reported headache precipitant fre- head or neck. The tension headache, tension headache
quency distribution by IHS headache diagnostic groups. overlap, cervicogenic headache, and migraine with aura
According to this table, the most frequent headache pre- groups had the greatest frequency of head or neck tender
cipitant was mental stress for all IHS diagnostic groups points.
except cluster and cervicogenic headaches. For these, the Table 7 presents the discriminant analysis for sets of
most frequent precipitant was neck position. The second neck symptoms as predictors of each diagnostic group.
most frequent precipitant for all diagnostic groups except As can be seen, severe headache beginning at the neck or
cluster and cervicogenic headaches was neck position. at a trigger point was the most common predictor across
Activity/exercise was the third most common precipitant IHS diagnostic groups for the first set of neck symptoms.
in all IHS diagnostic groups except the no diagnosis For the second set of neck symptoms, neck position and
group, where it was fourth. Moving the neck ranged from arms positioned overhead were the most common pre-
being the most frequent precipitant in the cluster head- dictors across IHS diagnostic groups. For the third set of
ache group to being the fifth most frequent precipitant in neck symptoms, having a tender point in the neck and

TABLE 5. Distribution of headache precipitants by International Headache Society headache diagnostic groups
Migraine Migraine Tension Cluster Cervico-Genic Chronic daily Migraine with
All with aura without aura headache headache headache headache and without aura
(N ⳱ 54) (N ⳱ 85) (N ⳱ 87) (N ⳱ 25) (N ⳱ 4) (N ⳱ 52) (N ⳱ 12) (N ⳱ 60)
Mental stress 68.8 80.0 75.9 60.0 50.0 65.4 75.0 83.3
Neck position 53.2 74.1 57.5 56.0 75.0 67.3 66.7 76.7
Anger 43.5 56.5 54.0 24.0 25.0 48.1 41.7 61.7
Activity 42.9 56.5 56.3 24.0 25.0 55.8 58.3 65.0
Moving neck 41.6 56.5 43.7 48.0 75.0 55.8 66.7 56.7
Weather 38.3 48.2 47.1 28.0 0.0 46.2 58.3 51.7
Bright light 35.7 54.1 48.3 20.0 25.0 51.9 50.0 61.7
Fatigue 32.5 42.4 40.2 12.0 25.0 38.5 41.7 46.7
Hunger 31.8 38.8 39.1 24.0 25.0 40.4 25.0 41.7
Arms over head 27.3 37.6 36.8 16.0 25.0 42.3 58.3 46.7
Eye strain 27.3 40.0 32.2 36.0 50.0 36.5 41.7 43.3
Personal problems 26.6 38.8 32.2 16.0 50.0 32.7 33.3 38.3
Driving 23.4 31.8 27.6 12.0 50.0 32.7 16.7 31.7
Lifting overhead 22.7 30.6 29.9 16.0 25.0 32.7 50.0 36.7
Sickness 22.1 27.1 26.4 4.0 25.0 21.2 25.0 30.0
Overheated 20.8 32.9 27.6 12.0 0.0 30.8 33.3 35.0
Chewing 18.8 29.4 27.6 8.0 0.0 28.8 33.3 36.7
Certain food 15.6 17.6 18.4 4.0 0.0 19.2 25.0 18.3
Monosodium glutamate 14.3 18.8 18.4 4.0 25.0 17.3 25.0 20.0
Menstruation 14.3 21.2 21.8 8.0 0.0 26.9 16.7 26.7
Allergies 13.6 20.0 17.2 0.0 0.0 21.2 25.0 21.7
Some drugs 12.3 15.3 14.9 4.0 0.0 11.5 16.7 18.3
Season of year 11.0 14.1 13.8 8.0 0.0 13.5 16.7 15.0
Alcohol 7.8 8.2 9.2 4.0 0.0 11.5 8.3 8.3

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

84 FISHBAIN ET AL.

TABLE 6. International Headache Society headache diagnostic groups compared for presence of neck associated symptoms
by ␹2
Migraine Migraine Tension Cluster Cervicogenic Chronic daily Migraine with Tension headache
with aura without aura headache headache headache headache and without aura and migraine
(N ⳱ 85) (N ⳱ 87) (N ⳱ 25) (N ⳱ 4) (N ⳱ 52) (N ⳱ 12) (N ⳱ 60) (N ⳱ 10)
Severe headache begins in
neck or trigger point p < 0.002 p < 0.007 p < 0.000
Clues to onset: neck pain
gets worse p < 0.000 p < 0.016 p < 0.000
Clues to onset: neck pain
travels up p < 0.000 p < 0.007 p < 0.009
When full-blown: feel in
neck p < 0.000 p < 0.050 p < 0.000 p < 0.041 p < 0.000
Brought on by neck position p < 0.000 p < 0.012 p < 0.000
Brought on by driving p < 0.006 p < 0.051
Brought on by arms over
head p < 0.001 p < 0.002 p < 0.002 p < 0.015 p < 0.000
Brought on by lifting over
head p < 0.009 p < 0.015 p < 0.035 p < 0.024 p < 0.000
Have neck pain p < 0.012 p < 0.020
Have tender point in neck,
head p < 0.012 p < 0.003 p < 0.000
Brought on by moving neck p < 0.000 p < 0.010 p < 0.002
Clues on onset: numbness in
arms p < 0.000 p < 0.010 p < 0.020 p < 0.001

Only probability values less than .05 or .05 presented.

feeling severe headache in the neck were the most com-
mon predictors across IHS diagnostic groups. In general,
the three sets of neck symptoms seemed to most
accurately classify migraine with aura and cervicogenic
headache.
DISCUSSION
As indicated in Tables 1 and 2, most of our headache
diagnostic groups overlapped with other diagnostic
groups (i.e., had more than one headache diagnosis). In
fact, there were more patients who overlapped within
each diagnostic group than those who did not. This was
especially so for the cervicogenic headache diagnostic
group, where 94.2% overlapped. These findings are simi-
lar to other reports in the literature with other headache
populations. For example, in a headache clinic popula-
tion, Sarin et al.16 reported that most patients required
two or more IHS diagnoses and that only one quarter of
patients with migraine had it as the only diagnosis. In our
study, the findings were similar: only 28.6% of patients
had some form of migraine as the only diagnosis. Mi-
graine and tension symptoms have also been reported to
overlap,17,18 as was the case with our patients. In refer-
ence to migraine as a group, Centonze et al.19 reported a
significant overlap between migraine with and without
aura. We have also found this with our patient group.
These studies generally support the results of our study.
Because physicians usually record one headache diagno-
sis,18 pain physicians should begin to attempt to make
multiple IHS diagnoses in their headache patients.

TABLE 7. Discriminant analyses for sets of neck symptoms as predictors of each International Headache Society
Set
1: Clues to onset
2: Brought on by
3: Neck symptoms
diagnostic group
Item that predicts
migraine with aura
[4] [2] [3] Correctly
classified 69.18%
[5] [7] Correctly
classified 73.38%
[11] [10] Correctly
classified 67.57%
Item that predicts
Item that predicts Item that predicts cervicogenic
migraine without aura tension headache headache
[4] Correctly [1] Correctly [1] [4] Correctly
classified 50.65% classified 41.78% classified 54.11%
[7] Correctly [6] [5] [7] Correctly [7] [5] Correctly
classified 57.79% classified 72.73% classified 67.53%
[10] [11] Correctly [10] [11] [9] [11] [10] Correctly
classified 62.16% Correctly classified classified 67.57%
73.97%

Set 1: [1] severe headache begins at neck or trigger point; [2] clue to onset: neck pain gets worse; [3] clue to onset: neck pain travels to skull; [4]
numbness in arms and legs.
Set 2: [5] brought on by neck position; [6] brought on by driving; [7] brought on by arms overhead; [8] brought on by lifting overhead.
Set 3: [9] have neck pain; [10] have tender point in neck; [11] feel severe headache in neck.

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

 IHS HEADACHE DIAGNOSTIC PATTERNS
The overlap results of this study have some implica-
tions for the headache nosology problem. In 1992, Wil-
son20 proposed an expansion of the benign recurring
headache model, where headaches are placed along a
continuum.21 In this model, migraine with aura is placed
at one end and headache with and without pericranial
muscle disorders is placed at the other end. Wilson20
expanded this model to include cervicogenic headache.
Our overlap results along with our headache precipitant
results (see Table 5) support this continuum model. In
addition, these results show that cervicogenic headache
type symptoms can be found in migraine patients, indi-
cating that perhaps these symptoms also present in head-
ache populations along a continuum.
Recently, the validity of the current IHS classification
system for separating migraine from tension headache
has been questioned.22 It has been suggested that these
two types of headache are part of a spectrum/continuum
rather than two distinct entities.22 There is some experi-
mental evidence for this view.23 Our overlap results also
partially support this view. A large percentage (40%) of
the tension headache patients overlapped with a migraine
diagnosis. From another standpoint, it has been claimed
that all tension headaches are really cervicogenic head-
aches.24 This, however, is not supported by our results
(see Table 2). Only nine cervicogenic headache patients
overlapped with the tension headache diagnostic group.
This was 36% of the tension headache group. Thus,
PWCP with the cervicogenic headache diagnosis overlap
with the tension headache group, but not all tension
headaches are cervicogenic headaches.
Our headache precipitant results (see Table 5) indicate
that the most common precipitant reported by all IHS
headache groups except for cluster and cervicogenic
headaches was mental stress. This is not an unusual find-
ing and has been reported by a number of researchers.
Mental stress has been reported to be the most common
headache precipitant for migraine25–27 and tension26–28
headaches. What is unusual is our finding that neck po-
sition is the most common trigger reported by cluster
headache, cervicogenic headache, and tension headache/
migraine patients. Neck position is also the second most
frequent trigger for migraine with aura, migraine without
aura, tension headache, chronic daily headache, migraine
both with and without aura, no diagnosis, and all head-
ache patients. In addition, neck movements are noted to
be common precipitants in this headache patient sample.
These findings for neck precipitants have not been re-
ported previously for other than cervicogenic headache.
For example, in a study of 300 headache outpatients,
Leone et al.29 reported that in only 2 patients, or 1% of
the migraine group, was the pain triggered by head or
85
neck movements. This major difference between our re-
sults and those of Leone et al.29 may indicate that our
headache group is different from other previously de-
scribed groups or that these neck-associated symptoms
have not been characterized previously in these headache
groups. Our headache group also seems to be exquisitely
sensitive to activity/exercise as a precipitant. Approxi-
mately 42.9% of our total headache group, 65% of the
migraine group, and 24% of the tension headache group
identified activity/exercise as a precipitant. These results
seem to be the reverse of previous reports, where the
figures for migraine have ranged from 1525 to 44.9%26
and the figure for tension headache is 67.3%.26 These
results would indicate that these patients are dissimilar to
those previously described in the literature. In addition,
these results would indicate that pain facilities should
activate these patients cautiously.
Cervicogenic headache has been defined as a unilat-
eral headache30,31 beginning at the neck, spreading to the
front of the head, and characterized by protracted pain
episodes.32,33 In addition, this type of headache is
claimed to have a marked female preponderance, to oc-
cur after whiplash trauma, and to be associated with a
reduction of range of neck movement and with ipsilateral
shoulder/arm pain.32 Further, attacks can be precipitated
mechanically by neck movements or by physicians
pressing on circumscribed neck points.32 Patients with
cervicogenic headache have been found to have lower
pressure-pain threshold measurements in the neck versus
patients with migraine and tension headaches.34 In addi-
tion, cervicogenic headache patients have been found to
have a greater number of trigger points on the symptom-
atic side of the neck versus the other side.35 Also, cer-
vicogenic headache patients seem to share some mi-
graine characteristics: they seem to demonstrate photo-
phobia and phonophobia.36–38 This type of headache is
claimed to be unresponsive to ergots.39,40 In terms of
prevalence, cervicogenic headache has been reported to
be found in 17.8% of the general population41 and in
13.8%42 of headache populations. In our sample of pain
facility PWCP with headache, using Sjaastad et al.’s10
most recent criteria, we found that cervicogenic head-
ache was seen in a high percentage (33.7%) of the total
headache population. Yet, only 5.8% of the PWCP with
a diagnosis of cervicogenic headache did not overlap
with other diagnostic groups (see Table 1). Over 70% of
the cervicogenic group fulfilled diagnostic criteria for
migraine, with significantly fewer (17.3%) fulfilling di-
agnostic criteria for tension headache. These results mir-
ror some other studies30,43 originating from headache
facilities, which have found that some cervicogenic pa-
tients fulfill diagnostic criteria for migraine. Sjaastad and
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
86 FISHBAIN ET AL.
Bovim38 have also pointed out that typical migraine find-
ings do occur in cervicogenic headache. Yet, our per-
centages of cervicogenic patients fulfilling diagnostic
criteria for migraine and other types of headache seem to
be much higher than those found in other studies,30,43
where the percentages were 20% and 10%, respectively.
These discrepancies could be related to the fact that we
did not use the full criteria for cervicogenic headache.
Sjaastad et al.10 have stated that criteria II (confirmatory
evidence by diagnostic anesthetic blockade) is an obliga-
tory criteria for scientific work. Because we do not per-
form blocks on our cervicogenic patients, our methods
did not fulfill the proposed criteria. As such, this ap-
proach could be the reason for the differences between
our results and those of others. In support of this state-
ment, there was a high overlap between the diagnoses of
cluster and cervicogenic headache, suggesting the possi-
bility of a C2 neuralgia, as this can mimic cluster head-
ache symptoms and signs. This raises the question of the
validity of deriving a diagnosis of cervicogenic head-
ache, or any headache type that it can mimic, without the
use of anesthetic blockade. Alternatively, as is pointed
out by other authors,29,40 a more precise definition of the
clinical criteria for cervicogenic headache versus mi-
graine may be required and is the reason for our results.
Chronic daily headache has been defined as frequent,
almost continuous, or continuous headache usually seen
at headache centers.14 This type of headache is not a
category that appears in the IHS headache classification,
where these patients can only be classified as having a
chronic tension type headache.14 Actually, chronic daily
headache is a syndrome consisting of a group of disor-
ders that can be subclassified as primary and secondary
types.13,44 The primary types include transformed mi-
graine and chronic tension type headache. The secondary
types include posttraumatic headache, cervical spine dis-
orders, and analgesic rebounding headache.13 In this
study, we eliminated one type of secondary chronic daily
headache (analgesic/ergotamine rebound) from our
analyses. Because of this, and because our chronic daily
headache group was rigorously defined (i.e., headache
always present rather than present daily), we believe that
our remaining chronic daily headache group represents
this concept well. The selectivity of our process for this
diagnosis, however, could be the reason for the limited
(7.8%) number of patients of the total headache group
identified as having chronic daily headache. In addition,
it is to be remembered that most of our PWCP had either
LBP or neck pain rather than a primary complaint of
headache. Thus, it would be expected that our headache
population should contain significantly fewer chronic
daily headache patients than a population of patients with
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
the primary complaint of headache, as may be seen at a
headache treatment facility or clinic. Of our chronic daily
headache group, 66.7% fulfilled diagnostic criteria for
migraine with aura, 75% for migraine without aura,
33.3% for tension headache, and 33.3% for cervicogenic
headache (see Table 2). Thus, these patients could have
been suffering from transformed migraine, chronic ten-
sion type headache, and cervicogenic headache prob-
lems. These overlapping diagnostic results are supported
by Srikiatkhachorn and Phanthumchinda,45 who found
that 30% of their chronic daily headache patients could
be diagnosed as suffering from migraine and 36.7% from
chronic tension type headache. The finding of patients
suffering from chronic daily headache who seem to fit
within other diagnostic headache groups in a sample of
pain facility patients is then compatible with the findings
of previous reports in other populations.
The lifetime prevalence of migraine in the general
population has been reported to have a 1:3 male-to-
female ratio.46,47 For our headache group, our migraine
male-to-female ratio was approximately 1:2. This figure
approximates that of the general population. Cervico-
genic headache is also claimed to have a marked female
preponderance.32 In our headache group, our cervico-
genic male-to-female ratio was approximately 1:3. Thus,
these results are also supported by the previous literature.
Large percentages of our migraine, tension headache,
cervicogenic headache, and chronic daily headache pa-
tients reported that they woke up with a headache (see
Table 3). Morning or nocturnal headaches have been
reported to be frequent indicators of a sleep distur-
bance,48 and it has been recommended that their pres-
ence might justify polysomnography. These data would
indicate that pain physicians should consider sleep dis-
orders as a possible factor in headaches seen at pain
facilities. Our data (see Table 3) also indicate that a large
percentage of migraine, tension headache, cervicogenic
headache, and chronic daily headache patients report that
their headache awakens them. This is a new finding that
we believe has not been reported previously. Because
headache patients have been reported to have frequent
sleep problems, it is possible that these sleep problems
could be secondary to the headache awakening the
patient.
Within each IHS diagnostic group, a large percentage
of our patients related the beginning of their headache to
a head, neck, or back injury. These percentages ranged
from 50% upward. There have been previous reports of
minor trauma to the head and/or neck precipitating re-
current episodes of classic migraine in patients with no
prior history of headaches.49,50 What is unusual about
our results is the large percentages of patients who make
these claims in each diagnostic group. We are not aware
 IHS HEADACHE DIAGNOSTIC PATTERNS
of any other studies that have documented this problem
in a similar fashion in a pain facility headache group.
The most interesting aspect of our results relates to the
presence of neck-associated symptoms in a large per-
centage of all our headache diagnostic groups (see
Tables 6 and 7). Our results demonstrate that both mi-
graine and cervicogenic headache patients frequently
have neck-associated symptoms (see Table 6). In addi-
tion, our results show that, in general, over 50% of each
headache diagnostic group complains of a tender point in
the neck and that this is most frequently reported by
migraine and/or cervicogenic headache patients. Finally,
the discriminant analysis data indicate that certain neck
headache characteristics predict headache. These are se-
vere headache beginning at the neck or trigger point,
headache brought on by neck position, and having a ten-
der point in the neck. These data support and are sup-
ported by previous studies. First, it is to be noted that it
has been reported that neck disorders are associated with
headaches.51 It has also been reported that neck pain
often accompanies headache in muscle contraction and
migraine headache. 2,52,53 A number of studies54–56 have
also demonstrated that headache patients (migraine, ten-
sion headache) are more likely to have neck tender points
or trigger points than controls. Finally, there is also one
report57 of the close association of upper limb pain in
close temporal relation with headache. This report is akin
to our finding of “clue to development of headache,
numbness in arms or legs” as a predictor of headache
within some diagnostic groups. Thus, our results under-
score these previous findings and demonstrate a large
number of neck-associated symptoms across IHS head-
ache diagnostic groups in a sample of pain facility head-
ache patients. What is the possible physiologic explana-
tion for these findings? It seems that the association of
neck pain with migraine can be explained physiologi-
cally by the interaction of the upper cervical dorsal roots
with the trigeminal nucleus caudalis in the dorsal horn of
the upper cervical spinal cord.58 This anatomic associa-
tion allows trigeminal nerve activation in migraine head-
ache to activate the dorsal horn of the upper spinal cord,
referring pain to the upper neck. Conversely, noxious
activation of the upper cervical nerve roots activates the
trigeminal nucleus caudalis, referring pain to the head
and face.58 Our results would indicate that this, in turn,
could trigger a migraine attack.
CONCLUSIONS
The results of this study lead to a number of general
observations and conclusions. First, a significant per-
centage of consecutive pain facility PWCP complain of
debilitating headache. Second, over 50% of these PWCP
87
attribute their headache to a head, neck, or low back
injury. Third, in over 90% of these PWCP, headache
characteristics fulfill IHS diagnostic criteria for mi-
graine. Fourth, approximately 45% of the PWCP can be
assigned more than one IHS headache diagnosis (i.e., a
significant number of PWCP overlap with more than one
headache diagnosis). Fifth, over 90% of the PWCP given
a cervicogenic headache diagnosis overlap with other
headache diagnoses, with the most common diagnosis
overlapped being migraine. Sixth, most of these head-
ache PWCP complain of neck-associated headache
symptoms which seem to have an impact on their head-
aches. These observations lead to the following general
conclusions. First, the frequency of complaints of debili-
tating headache in this pain facility PWCP sample would
indicate that this complaint should be routinely looked
for in PWCP complaining of neck pain or LBP. Second,
future research should address the issue of the etiology of
these headaches as it relates to a history of injury to the
head, neck, and lower back. Third, our overlap results
would indicate that pain physicians should consider the
possibility of a migraine treatment regimen in diagnoses
such as tension, cervicogenic, and chronic daily head-
ache, especially when treatments aimed at these head-
aches have failed. This statement does not preclude the
fact that each headache patient requires a headache his-
tory, physical examination, workup (if necessary), and
derivation of IHS headache diagnosis based on the re-
sults of this process. Fourth, the overlap data in reference
to migraine versus cervicogenic headache would indicate
that more precise diagnostic criteria are required to sepa-
rate migraine from cervicogenic headache. It is to be
noted, however, that there are expert headache clinicians
who believe that there are adequate criteria currently
available to separate cervicogenic from migraine head-
ache when expertly applied. Fifth, the demonstration of
the importance of neck-associated symptoms in most of
these patients, including those with migraine headache,
should be investigated further in future studies and es-
pecially in general headache populations. Because these
conclusions are based on data derived from a specialized
pain facility population, it is likely that they may not be
generalizable to other distinct pain populations such as a
general headache population found at a headache clinic.
Acknowledgments: This study was partially supported by
grant H133A00032 from the National Institute of Disability
and Disability Research. The authors thank Ms. Sandy Vassi-
latos for manuscript typing.
REFERENCES
1. Fishbain DA, Goldberg M, Meagher BR, et al. Male and female
chronic pain patients categorized by DSM-III psychiatric diagnos-
tic criteria. Pain 1986;26:181–97.
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
88 FISHBAIN ET AL.
2. Lebbink J, Spierings EL, Messinger HB. A questionnaire survey of
muscular symptoms in chronic headache. An age-and sex-
controlled study. Clin J Pain 1991;7:95–101.
3. Blau JN, MacGregor EA. Migraine and the neck. Headache 1994;
24:88–90.
4. Hasvold T, Johnsen R, Forde OH. Non-migrainous headache, neck
or shoulder pain, and migraine—differences in association with
background factors in a city population. Scand J Prim Health Care
1996;14:92–9.
5. Ahern DK, Lezzi A, Hursey K, et al. The co-occurrence of chronic
low back pain and headache: prevalence, neuroticism, and treat-
ment outcome. Clin J Pain 1988;4:27–31.
6. Egan KJ, Betrus P. Psychological functioning in five pain syn-
dromes: tension headache, backache, migraine, and temporoman-
dibular and gastrointestinal pain syndromes. Clin J Pain 1987;2:
233–8.
7. Duckro PN, Schultz KT, Chibnall JT. Migraine as a sequela to
chronic low back pain. Headache 1994;4:279–81.
8. Headache Classification Committee of International Headache So-
ciety. Classification and diagnostic criteria for headache disorders,
cranial neuralgias, and facial pain. Cephalalgia 1988;8:1–96.
9. Solomon S. Diagnosis of primary headache disorders. Validity of
the International Headache Society criteria in clinical practice. Adv
Headache 1997;15:15–26.
10. Sjaastad O, Fredriksen A, Pfaffenrath V. Cervicogenic headache:
diagnostic criteria. Headache 1998;38:442–5.
11. Standard of Care Committee for Headache Diagnosis and Treat-
ment. Standards of care for treating headache in primary care prac-
tice. Cleve Clin J Med 1997;64:373–83.
12. Benoliel R, Sharav Y. Paroxysmal hemicrania. Case studies and
review of the literature [see comments]. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 1998;85:285–92.
13. Silberstein SD. Tension-type headaches. Headache 1994;34(Sup-
pl):S2–7.
14. Nappi G, Granella F, Sandrini G, et al. Chronic daily headache.
How should it be included in the IHS classification? Headache
1999;39:197–203.
15. Shrout PE, Spitzer RL, Fleiss JL. Quantification of agreement in
psychiatric diagnosis revisited. Arch Gen Psychiatry 1996;44:
172–7.
16. Sanin LC, Mathew NT, Bellmeyer LR, et al. The International
Headache Society (IHS) headache classification as applied to a
headache clinic population. Cephalalgia 1994;14:443–6.
17. Diehr P, Wood R, Wolcott B, et al. On the relationships among
headache symptoms. J Chron Dis 1982;35:321–31.
18. Stang PE, Von Korff M. The diagnosis of headache in primary
care: factors in the agreement of clinical and standardized diag-
noses. Headache 1994;34:138–42.
19. Centonze V, Polito BM, Valerio A, et al. Migraine with and with-
out aura in the same patient: expression of a single clinical entity?
Cephalalgia 1997;17:585–7.
20. Wilson PR. Cervicogenic headache. Am Pain Soc J 1992;1:
259–64.
21. Marcus DA. The cervical spine in benign recurring headache. Am
Pain Soc J 1992;1:268–71.
22. Marcus DA. Migraine and tension-type headaches: the question-
able validity of current classification systems. Clin J Pain 1992;
8:28–36.
23. Schade AJ. Quantitative assessment of the tension-type headache
and migraine severity continuum. Headache 1997;27:646–53.
24. Bogduk N, Corrigan B, Kelly P, et al. Cervical headache. Med J
Aust 1985;143:202, 206–7.
25. Robbins L. Precipitating factors in migraine: a retrospective review
of 494 patients. Headache 1994;34:214–6.
26. Scharff L, Turk DC, Marcus DA. Triggers of headache episodes
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
and coping responses of headache diagnostic groups. Headache
1995;25:397–403.
27. Rasmussen BK. Migraine and tension-type headache in a general
population: precipitating factors, female hormones, sleep pattern
and relation to lifestyle. Pain 1993;53:65–72.
28. Ulrich V, Russell MB, Jensen R, et al. A comparison of tension-
type headache in migraineurs and in non-migraineurs: a popula-
tion-based study. Pain 1996;67:501–6.
29. Leone M, D’Amico D, Moschiano F, et al. Possible identification
of cervicogenic headache among patients with migraine: an analy-
sis of 374 headaches. Headache 1995;12:461–4.
30. Sjaastad O, Bovim G, Stovner LJ. Laterality of pain and other
migraine criteria in common migraine. A comparison with cervi-
cogenic headache. Funct Neurol 1992;7:289–94.
31. D’Amico D, Leone M, Bussone G. Side-locked unilaterality and
pain localization in long-lasting headaches: migraine, tension-type
headache, and cervicogenic headache. Headache 1994;34:526–30.
32. Sjaastad O. The headache of challenge in our time: cervicogenic
headache. Funct Neurol 1990;5:155–8.
33. Sjaastad O. Cervicogenic headache. Clin Neurol Neurosurg
1992;94(Suppl):S147–9.
34. Bovim G. Cervicogenic headache, migraine, and tension-type
headache. Pressure-pain threshold measurements. Pain 1992;51:
169–73.
35. Jaeger B. Are “cervicogenic” headaches due to myofascial pain
and cervical spine dysfunction. Cephalalgia 1989;9:157–64.
36. Vanagaite Vingen J, Stovner LJ. Photophobia and phonophobia in
tension type cervicogenic headache. Cephalalgia 1998;18:313–8.
37. Sjaastad O, Fredriksen TA, Pfaffenrath V. Cervicogenic headache:
diagnostic criteria. Headache 1990;30:725–6.
38. Sjaastad O, Bovim G. Cervicogenic headache: the differentiation
from common migraine. An overview. Funct Neurol 1991;6:
93–100.
39. Bovim G, Sjaastad O. Cervicogenic headache: responses to nitro-
glycerin, oxygen, ergotamine and morphine. Headache 1993;33:
249–52.
40. Pollmann W, Keidel M, Pfaffenrath SB. Headache and the cervical
spine: a critical review. Cephalalgia 1997;17:801–16.
41. Nilsson N. The prevalence of cervicogenic headache in a random
population sample of 20–59 year olds. Spine 1995;20:1884–8.
42. Pfaffenrath V, Kaube H. Diagnostics of cervicogenic headache.
Funct Neurol 1990;5:159–64.
43. Leone M, D’Amico D, Grazzi L, et al. Cervicogenic headache: a
critical review of the current diagnostic criteria. Pain 1998;78:1–5.
44. Sandrini G, Manzoni GC, Zanferrari C, et al. An epidemiological
approach to the nosography of chronic daily headache. Cephalal-
gia 1993;13(Suppl 12):72–7.
45. Srikiatkhachorn A, Phanthumchinda K. Prevalence and clinical
features of chronic daily headache in a headache clinic. Headache
1997;37:277–80.
46. Rasmusen BK, Jensen R, Schroll M, et al. Epidemiology of head-
ache in a general population: a prevalence study. J Clin Epidemiol
1991;44:1147–57.
47. Stewart WF, Chechter A, Rasmussen BK. Migraine prevalence. A
review of population-based studies. Neurology 1994;44(Suppl
4):17–23.
48. Paiva T, Batista A, Martins P, et al. The relationship between
headaches and sleep disturbances. Headache 1995;35:590–6.
49. Weiss HD, Stern BJ, Goldberg J. Post-traumatic migraine: chronic
migraine precipitated by minor head or neck trauma. Headache
1991;31:451–6.
50. Solomon S. Posttraumatic migraine. Headache 1998;38:772–8.
51. Leclerc A, Niedhammer I, Landre MF, et al. One-year predictive
factors for various aspects of neck disorders. Spine 1999;24:
1455–62.
52. Vernon H, Steiman I, Hagino C. Cervicogenic dysfunction in
muscle contraction headache and migraine: a descriptive study. J
Manipulative Physiol Ther 1992;15:418–29.
 IHS HEADACHE DIAGNOSTIC PATTERNS
53. Featherstone HJ. Migraine and muscle contraction headaches: a
continuum. Headache 1985;25:194–8.
54. Marcus DA, Scharff L, Mercer S, et al. Musculoskeletal abnor-
malities in chronic headache: a controlled comparison of headache
diagnostic groups. Headache 1999;39:21–7.
55. Boquet J, Boismare F, Payenneville G, et al. Lateralization of
headache: possible role of an upper cervical trigger point. Cepha-
lalgia 1989;9:15–24.
89
56. Burg HE. Chronic recurrent headache and regional myofascial
pain: the common denominator. Presented at the 10th Annual
Meeting of the American Pain Society, New Orleans, LA, Novem-
ber 7–10, 1991.
57. Raudino F. Limb pain and headache. Headache 1994;34:169–
71.
58. Weiller C, May A, Limmroth J, et al. Brain stem activation in
spontaneous human migraine attacks. Nat Med 1995;1:658–60.
The Clinical Journal of Pain, Vol. 17, No. 1, 2001
90 FISHBAIN ET AL.

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

IHS HEADACHE DIAGNOSTIC PATTERNS 91

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

92 FISHBAIN ET AL.

The Clinical Journal of Pain, Vol. 17, No. 1, 2001

IHS HEADACHE DIAGNOSTIC PATTERNS 93

The Clinical Journal of Pain, Vol. 17, No. 1, 2001