Pain Pract. 2010 (May); 10 (3): 214—221 ~ FULL TEXT
Yaw A. Nyame MHSA,
Andrew P. Ambrosy BS,
Miguel Saps MD,
Papa N. Adams BS,
Gati N. Dhroove MD,
Santhanam Suresh MD
Department of Pediatric Anesthesiology Children's Memorial Hospital,
Northwestern University's Feinberg School of Medicine,
Chicago, Illinois 60614, USA.
OBJECTIVES: The aim of this study was to longitudinally evaluate the epidemiological characteristics of headaches in a school-based, community setting and to determine the impact of headache symptoms on the health of children.
METHODS: After institutional review board approval, a prospective cohort study was conducted at two Chicago public schools for a period of 6 months. Members of the research team surveyed both schools weekly for headache and other pain symptoms. The students rated each pain symptom on a 5-point scale from 0 ("not at all") to 4 ("a whole lot"). Demographic information was collected at the time of enrollment, and all participants were asked to complete age-appropriate and validated pediatric surveys to assess the severity of concurrent somatic complaints, anxiety symptoms, functional limitations, and quality of life issues.
RESULTS: Of the participating children, 89.5% reported at least one headache during the study period. Females experienced more frequent headaches compared with males (P < 0.05). Children reporting headaches had a significantly increased risk of experiencing other troubling somatic symptoms (P < 0.05). Headache severity showed a moderate correlation with increased feelings of anxiety, functional disability, and a diminished quality of life (P < 0.05).
CONCLUSIONS: School-aged children commonly experience headaches. Children experiencing headaches are more likely to report other somatic symptoms, feelings of anxiety, functional limitations, and quality of life impairments.
Key Words: headaches, prevalence, children, anxiety,functional disability, quality of life
From the FULL TEXT Article:
Studies conducted worldwide have found headaches to
be one of the primary complaints among children. [1–4]
In the U.S., it is estimated that headaches are responsible
for approximately 2.1 million emergency department
visits each year for patients under the age of
18.  Research has demonstrated that the prevalence
and severity of headaches varies among children by
gender,  age, [7, 8] and socioeconomic status. [9, 10] In girls,
the frequency and severity of headaches is greater following
pubertal changes than boys. Conversely, boys
typically experience more frequent headaches than
girls before the onset of puberty. [7, 11–13] Prior studies
have not provided conclusive evidence to support
ethnic disparities in headache prevalence or severity.
However, there are data showing that several other
socioeconomic variables, such as parental income,
marital status, and education level, may exert an effect
on headache prevalence.  Moreover, headache severity
is exacerbated by anxiety and/or stress [15–17] and chronic
pain problems. [18, 19]
Understanding the instigating factors that increase
headache frequency and severity is important due to the
negative consequences that headache pain can have on
children. Frequent headaches in children are associated
with well-documented behavioral problems such as
aggression, impulsivity, and inattention.  Additionally,
there is strong evidence in the existing literature supporting
the notion that chronic headaches pose the
greatest physical, psychological, and social threat to
children. [14, 17, 21, 22]
This study explored the dynamic attributes of headaches
through a prospective cohort study of school-aged
children. The research team aimed to determine the
epidemiological characteristics of headaches in the
cohort and to evaluate the impact of headaches on concurrent
somatic symptoms, the state and trait of anxiety,
functional limitations, and quality of life.
After institutional review board (IRB) approval, a prospective
cohort study was conducted at two ethnically
and socioeconomically diverse Chicago Public Schools
(WLNA and LMAS). All 3rd to 8th grade students and
their parents/guardians received an invitation to participate
in this prospective cohort study. Consent and
assent (in children over 12) forms were obtained by one
of the investigators. The study objectives were not
revealed to the children or their parents, and there were
no exclusion criteria for enrollment.
Between January and June of 2006, members of the
research team visited each school in the middle of the
week to administer a brief, confidential survey. The
study questionnaire asked participants if they had experienced
headaches, chest pain, stomach pain, or limb
pain during the previous week. The students rated their
pain on a 5-point scale, from 0 (“not at all”) to 4 (“a
whole lot”) based on the Children’s Somatization Inventory
(CSI).  It was explained to the students that “not at
all” represented no pain and “a whole lot” was equivalent
to maximum pain.
Based on the weekly questionnaires, the prevalence and
severity of headaches was determined. The proportions
of children experiencing headaches for 4, 8, and 12
consecutive weeks were calculated as a proxy for headache
chronicity. To minimize the effect of school absenteeism,
consecutive weeks were defined as time periods
in which study participants completed 75% or more of
the total number of surveys administered.
For each student enrolled in the study, the total
number of headaches and a pain score were determined
for the entire study period. The pain score was simply
the average of all headache scores reported during the
study period. The Pearson correlation coefficient was
calculated to define the relationship between the pain
score and the number of headaches reported for the
entire duration of the study.
Demographic data, including age, gender, and ethnicity,
were recorded for all willing study participants. The
effects of age, sex, and ethnicity on headache frequency
and severity were determined using an unpaired, twotailed
Student’s t-test. Site-related differences between
the two schools participating in the study were also
analyzed using an unpaired, two-tailed Student’s t-test.
This and all subsequent statistical analysis was completed
using JMP—Version 7, SAS system (Cary, NC,
Measures of Concurrent Pain Symptoms,
Psychological Impact, and Quality of Life
All study subjects completed the CSI,  the State-Trait
Anxiety Inventory for Children (STAIC),  the Pediatric
Functional Disability Inventory (PFDI), and the Pediatric
Quality of Life Inventory (PEDSQL)  version 4 at
the time of enrollment.
CSI is designed to assess the severity of somatic
symptoms (ie, “faintness or dizziness,” “sore muscles,”
etc.) that do not require an organic disease etiology.
Study participants rated each of 31 symptoms on a scale
ranging from 0 (“not at all”) to 4 (“a whole lot”). The
number of reported somatic symptoms irrespective of
the severity (range 0 to 31) and a total score taking into
account both the number and the severity of symptoms
(range 0 to 124) were calculated. The Pearson correlation coefficients between both objective measures and
headache pain were determined. Positive correlations
signified a direct relationship between headache pain
scores and concurrent somatic symptoms.
State-Trait Anxiety Inventory for Children.
survey consists of two scales: the S-anxiety scale and the
T-anxiety scale, each of which is comprised of 20 statements
designed to measure the Diagnostic and Statistical
Manual of Mental Disorders defined symptoms of anxiety. 
The S-anxiety scale is designed to measure transitory
anxiety states that fluctuate over time, while the
T-anxiety scale is intended to measure stable differences
in anxiety that are fixed over time. For both scales, each
statement was rated from 1 to 4, with higher scores
indicating higher levels of anxiety. The responses were
summed for total values for both scales for each participant.
The Pearson correlation coefficient between the
S-anxiety and T-anxiety scales and headache pain scores
was calculated. Positive values indicated a direct correlation
between headache severity and anxiety.
Pediatric Functional Disability Inventory.
assesses health-related difficulties in physical and psychosocial
functioning in children. The survey consists of
15 potential activity limitations rated on a scale of 0 to
4. Total scores were calculated by summing the rating
for each item. The total score had a potential range of 0
to 60 with higher values indicating greater functional
disability. The Pearson correlation coefficient between
the total scores and headache pain scores was determined.
Positive correlations corresponded to a direct
relationship between headache severity and functional
Pediatric Quality of Life Inventory.
The PEDSQL is a
modular survey that measures physical, emotional,
social, and school functioning. A total of 23 items are
rated on a scale of 0 to 4 and converted to a reverse
score of 0 to 100 where 0 = 100, 1 = 75, 2 = 50, 3 = 25,
and 4 = 0. On the 100-point scale, higher numbers correspond
to a greater quality of life. The total scale score
was computed by averaging the rating for each survey
item. The relationship between the average PEDSQL
scale values and headache pain scores was assessed
using the Pearson correlation coefficient. For consistency
and comprehensibility, the inverse of the correlation
coefficient was reported. Using this convention, a positive correlation denoted a direct relationship
between headache pain score and decreases in quality of
Four hundred and ninety-five students were eligible to
participate in the study. Two hundred and thirty-seven
children, representing 48% of the eligible subjects, were
enrolled in the study. An average of 208 students (88%)
completed the survey each week, and all enrolled participants
completed the study.
The average age of the students was 11.75 years
(SD ± 1.59 years, range of 8 to 15 years of age). The
gender composition of the cohort was 131 females and
105 males. One student did not provide any demographic
information. The self-reported ethnicities of the
cohort were 33% Black, 22% Hispanic, 21% Caucasian,
8% Asian, 13% other, and 3% not recorded.
89.5% of children reported experiencing at least one
headache during the study period. The overall weekly
prevalence of headaches among the students was
38.2%. Fifty-four percent of the participants experienced
headaches for 4 consecutive weeks, 30.4% for 8
consecutive weeks of headaches, and 22.8% for 12 consecutive
weeks (Table 1).
In general, children reported low headache pain
scores. The mean of headache scores for the cohort was
0.6 (SD ± 0.73, range of 0 to 4). Statistical analysis
showed a strong correlation (R = 0.80) between the
headache pain score and the total number of headaches
reported (Figure 1).
Demographic Differences in Headache Epidemiology
There was not a significant difference between the two
schools (WLNA and LMAS) in the total number of
headaches or headache pain scores reported. Furthermore,
with one exception, there were no age or
ethnicity-related differences in the total number of headaches or the pain scores reported by study participants. There was a significant difference between Blacks and
Hispanics (P < 0.05), with Blacks reporting 37% fewer
headaches than Hispanics. Interestingly, female study
participants reported more headaches than males
(P < 0.05). The girls also reported greater pain scores
than boys; however, the difference was not statistically
Concurrent Somatic Symptoms
The average number of somatic symptoms reported was
8.97 (SD ± 6.77, range 0 to 31) for the study population
(Table 2). There was a statistically significant correlation
(R = 0.44, 95% CI: 0.31 to 0.54) between headache
pain scores and the number of concurrent somatic
symptoms reported. There were no statistically significant gender differences in the correlations. Similarly, the
average total score for the CSI, reflecting number and
severity, was 13.73 (SD ± 15.31, range 0 to 104) for the
cohort. Total score and headache pain score were also
directly related (R = 0.50, 95% CI: 0.38 to 0.60) and
the magnitude of the correlation was comparable for
boys and girls.
The data revealed a weak association between the headache
pain score and a state of anxiety as measured by
the S-anxiety scale (R = 0.23, P <0.05). Interestingly,
there was a stronger association between headache pain
scores and trait anxiety as calculated by the T-anxiety
scale (R = 0.45, P < 0.05).
Quality of Life and Headaches
The PFDI mean sum total for the cohort was 7.00
(SD ± 7.61, range 0 to 41) (Tables 3 and 4). There was
a moderate correlation (R = 0.53, 95% CI: 0.42 to
0.63) between headache pain scores and the presence of functional disability, which was consistent across girls and boys. Study participants also reported an average
total score of 85.17 (SD ± 13.37, range 26.09 to 100)
for the PEDSQL (Tables 5 and 6). Similarly, there was a
moderate correlation (R = 0.48, 95% CI: 0.36 to 0.59)
between headache pain scores and decreases in quality
of life. This finding was also independent of gender.
The primary objective of this study was to characterize
the epidemiology and the social impact of headaches on
school-aged children through a large, prospective,
community-based cohort study. Eighty-nine percent of
the students surveyed reported experiencing at least one
headache during the 6-month study period. Current
research protocols for children and adults define chronic
headaches as 5 or more days with headache symptoms
each week.  It was not possible to quantify headache
chronicity in this study based on this criterion because
the data were not collected on a daily basis. However, it
is important to note that 30% of study participants
experienced headaches for 8 or more consecutive weeks.
These findings are consistent with prior research and
indicate that headaches are one of the principle chronic
pain symptoms experienced by children. [28, 29]
The data also showed that girls reported experiencing
headaches more frequently than their male peers. Previously
conducted studies have found that headaches tend
to occur more frequently in boys than in girls prior to
puberty, while the opposite is true during and following
pubertal changes. Unfortunately, the research team did
not collect data on the pubertal status of study participants
and therefore could not definitively assess the
relationship between puberty and gender-related differences
in headache frequency. However, based on the
average age of study subjects (11.75 years, range 8 to 15
years, SD ± 1.59 years), it is probable that this relationship
existed in our study population as well, explaining
the fact that girls reported a greater total number of
headaches for the entire duration of the study.
In contrast to gender, the research team did not find
any age-related differences in the total number of headaches
or the average headache pain score. The absence
of an age effect is not surprising because the age distribution
of study participants was narrow. Similarly, with one exception, the research team did not find significant
differences between ethnic groups in headache epidemiology.
In contrast, a previous study investigating racial
and ethnic differences in the physical pain symptoms
reported by adolescents found the prevalence of recurrent
headaches to be highest in Caucasians and lowest in
Asians.  However, the impact of race and ethnicity on
headache epidemiology is difficult to assess and is likely
complicated by confounding socioeconomic factors
such as family income and parental education level.
It is noteworthy that this study found that experiencing
headache symptoms moderately correlated with
experiencing additional undesirable somatic symptoms
as measured by the CSI. This finding has important
implications and demonstrates that headache symptoms
rarely occur in isolation and are more often found
to be part of a constellation of distressing somatic
symptoms. Interestingly, there was a moderate and statistically
significant association between anxiety and
headache pain in the cohort. This finding is consistent
with the outcome of a previous study measuring the
impact of headache pain on anxiety using the Spielberg
STAIC questionnaire.  A final important outcome of
this study is the finding that headaches showed a
modest correlation with alterations in quality of life. It
is possible that this correlation would have been stronger
if the PFDI and PEDSQL had been administered
each week for the entire duration of the study. Nonetheless,
this confirms earlier studies utilizing the PFDI
and PEDSQL surveys, which have found frequent and
severe headaches to be associated with significantly
increased functional disability and decreased quality of
life. [31–33] It is interesting to note that although girls
reported experiencing a statistically significant greater
total number of headaches, this study found that the
functional limitations and quality of life impairments
resulting from severe headaches is a relatively genderindependent
There are several methodological strengths of this
study including its prospective design and community
setting. In addition, the data were collected from two
socioeconomically diverse Chicago public schools,
which are exemplary of the social, racial, and economic
variability of other inner-city populations. There were
no statistically significant differences between the two
study sites in the total number of headaches or the
average headache pain score reported. This observation
further substantiates the notion that the findings in this
cohort are somewhat representative of other populations
of inner-city school children in the U.S.
The research group recognized the possibility for
recall bias and put several measures in place to increase
the accuracy and reliability of the answers provided by
the students. First, the students were only asked to
retain information for the previous week. Second,
instructions were provided to the participants before
and during the completion of each weekly questionnaire.
Third, members of the research team were available
on-site to answer questions while students filled out
the surveys. The research personnel were also able to
check the questionnaires as students completed them,
allowing members of the research team to immediately
address any incomplete surveys. These precautions were
integral to obtaining a high percentage (88%) of completed
data throughout the study.
In addition, approximately 50% of eligible students
enrolled in the study. Although incomplete participation
presented the possibility of selection bias, concealing the
objectives of the study from the students and their
parents minimized this potential concern. Prior to study
enrollment, families were informed that the research
team was interested in collecting epidemiologic data on
general childhood symptoms, including headache and
other pain symptoms. Nonetheless, the research team
recognizes that families with sicker children or healthconscious
parents could have enrolled their children in
the study with greater frequency. Similarly, it is also
possible that sicker children may have missed the opportunity
to enroll in the study due to school absenteeism.
The research team attempted to compensate for this
potential bias by allowing children who missed the first
week of the study to enroll in the study during subsequent
A potential limitation of the data is the absence of
diagnoses, such as the International Classification of
Headache Disorders published by the International
Headache Society. Collecting accurate and reliable
headache diagnoses directly from study participants
and/or their parents would have necessitated assuming
an adequate level of health literacy in the cohort. In fact,
a recent critical review of the literature focusing on the
impact of headaches on quality of life in children found
that less than half of the 33 studies reviewed included
headache diagnoses.  Furthermore, the majority of the
studies offering diagnostic categories did not provide
defined diagnostic criteria. It is our opinion that assigning
formal headache diagnoses is at best not desirable
and at worst not reliable. In a community setting, children
and adolescents are affected by the burden of headache
and somatic symptoms. In contrast, formal diagnoses are best made in a clinical setting where
they take on meaning with respect to prognosis and
In conclusion, this study demonstrates that headaches
are a common complaint among children. A significant
subset of the cohort reported experiencing
headaches in successive weeks, suggesting that headache
symptoms were a chronic problem for many study participants.
Finally, this study found headaches to be associated
with concurrent somatic symptoms, feelings of
anxiety, functional restrictions, and a discernibly diminished
quality of life. Consequently, early headache recognition
and treatment may represent an important
means for improving the physical and psychosocial
health of children and adolescents.
This research was funded in part by a FAER-MSARF
grant, American Society of Anesthesiology, 2008.
Fendrich K, Vennemann M, Pfaffenrath V, et al.
Headache prevalence among adolescents—the German DMKG headache study.
Gordon KE, Dooley JM, Wood EP.
Self-reported headache frequency and features associated with
frequent headaches in Canadian young adolescents.
Headaches in children and adolescents.
Am Fam Physician. 2002;65:625–632.
Petersen S, Brulin C, Bergström E.
Recurrent pain symptoms in young schoolchildren are often multiple.
Goldstein JN, Camargo CA, Pelletier AJ, Edlow JA.
Headache in United States emergency departments: demographics,
work-up and frequency of pathological diagnoses.
Ghandour RM, Overpeck MD, Huang ZJ, Kogan MD, Scheidt PC.
Headache, stomachache, backache, and morning fatigue among adolescent girls
in the United States: associations with behavioral, sociodemographic,
and environmental factors.
Arch Pediatr Adolesc Med. 2004;158:797–
Virtanen R, Aromaa M, Rautava P, et al.
Changing headache from preschool age to puberty. A controlled study.
Virtanen R, Aromaa M, Rautava P, et al.
Changes in headache prevalence between pre-school and pre-pubertal ages.
Racial/ethnic differences in adolescents’ physical symptoms.
J Pediatr Nurs. 2005;20:153–162.
Sasmaz T, Bugdayci R, Ozge A, Karakelle A, Kurt O, Kaleagasi H.
Are parents aware of their schoolchildren’s headaches?
Eur J Public Health. 2004;14:366–368.
Aromaa M, Sillanpaa M, Rautava P, Helenius H.
Pain experience of children with headache and their families: a controlled study.
Aromaa M, Sillanpaa ML, Rautava P, Helenius H.
Childhood headache at school entry: a controlled clinical study.
Sillanpaa M, Aro H.
Headache in teenagers: comorbidity and prognosis.
Funct Neurol. 2000;15(suppl 3):116–
Queiroz LP, Barea LM, Blank N.
An epidemiological study of headache in Florianopolis, Brazil.
Greene JW, Walker LS.
Psychosomatic problems and stress in adolescence.
Pediatr Clin North Am. 1997;44:1557–
Kröner-Herwig B, Morris L, Heinrich M.
Biopsychosocial correlates of headache: what predicts pediatric headache occurrence?
Childhood headache, stress in adolescence, and primary headache in young adulthood:
a longitudinal cohort study.
Galli F, D’Antuono G, Tarantino S, et al.
Headache and recurrent abdominal pain: a controlled study by the means
of the Child Behaviour Checklist (CBCL).
Laimi K, Metsähonkala L, Anttila P, et al.
Outcome of headache frequency in adolescence.
Virtanen R, Aromaa M, Koskenvuo M, et al.
Externalizing problem behaviors and headache: a follow-up study
of adolescent Finnish twins.
Bugdayci R, Ozge A, Sasmaz T, et al.
Prevalence and factors affecting headache in Turkish schoolchildren.
Pediatr Int. 2005;47:316–322.
Fearon P, Hotopf M.
Relation between headache in childhood and physical and psychiatric symptoms
in adulthood: national birth cohort study.
Garber J, Walker LS, Zeman J.
Somatization symptoms in a community sample of children and adolescents:
further validation of the children’s somatization inventory.
Psychol Assess. 1991;3:588–595.
Manual for the State-Trait Anxiety Inventory for Children.
Consulting Psychologist Press; 1973.
Claar RL, Walker LS.
Functional assessment of pediatric pain patients: psychometric properties
of the functional disability inventory.
Varni JW, Seid M, Kurtin PS.
PedsQL 4.0: reliability and validity of the Pediatric Quality of
Life Inventory version 4.0 generic core scales in healthy and patient populations.
Med Care. 2001;39:800–812.
Gladstein J, Holden EW, Winner P, Linder S.
Chronic daily headache in children and adolescents: current status
and recommendations for the future.
Migraine in school children. A study of the incidence and short-term prognosis,
nd a clinical, psychological and electroencephalographic comparison between
children with migraine and matched controls.
Acta Paediatr Suppl. 1962;136:1–151.
Prevalence of migraine and other headache in Finnish children starting school.
Andrasik F, Kabela E, Quinn S,
Attanasio V, Blanchard EB, Rosenblum EL.
Psychological functioning of childrenwho have recurrent migraine.
Fichtel A, Larsson B.
Psychosocial impact of headache and comorbidity with other pains among
Swedish school adolescents.
Powers SW, Patton SR, Hommel KA, Hershey AD.
Quality of life in childhood migraines: clinical impact and comparison
to other chronic illnesses.
Pediatrics. 2003; 112:e1–e5.
Powers SW, Patton SR, Hommel KA, Hershey AD.
Quality of life in paediatric migraine: characterization of agerelated effects
using PedsQL 4.0.
Kernick D, Campbell J.
Measuring the impact of headache in children: a critical review of the literature.
Return to the HEADACHE Page