Centers for Disease Control and Prevention, Atlanta, Georgia
| ABSTRACT |
|---|
|
|
|---|
OBJECTIVES. This study, based on responses to the Strengths and Difficulties Questionnaire, was designed to examine difficulties and impairments related to emotions, concentration, behavior, and social functioning among children with frequent or severe headaches (FSH).
METHODS. We used a cross-sectional study of 9264 children aged 417 years from the 2003 National Health Interview Survey, an ongoing, computer-assisted personal interview survey of the noninstitutionalized US population.
RESULTS. Approximately 6.7% of children experienced FSH during the previous 12 months. Overall, children with FSH were 3.2 times more likely than children without FSH to have a high level of difficulties and 2.7 times more likely to have a high level of impairment, suggesting potential mental health issues. More specifically, analyses revealed that children with FSH were significantly more likely than those without FSH to exhibit high levels of emotional, conduct, inattention-hyperactivity, and peer problems and were significantly more likely than children without FSH to be upset or distressed by their difficulties and to have their difficulties interfere with home life, friendships, classroom learning, and leisure activities.
CONCLUSION. Because children with FSH experience notable pain, mental health issues, and functional limitations, integrated care using a biopsychosocial approach is warranted.
Key Words: behavior impairment headache mental health
Abbreviations: QoLquality of life FSHfrequent or severe headache EXextended version SDQStrengths and Difficulties Questionnaire NHISNational Health Interview Survey CIconfidence interval ORunadjusted odds ratio AORadjusted odds ratio
Recent population-based research has been conducted examining the associations between headaches and psychological impairment in adult populations. Current research indicates that, as compared with adults without headaches, those with headaches have significantly poorer quality of life (QoL), often impairing family, work, and social functioning.13 In addition, severe headaches among adults are often comorbid with psychological conditions, such as depression, mania, and anxiety disorders.48 Whereas
6% of men and 18% of women in the United States have 1 or more migraine headaches per year,9 headaches are also common among children and adolescents. It is estimated that
10.6% of children aged 5 to 15 years10 and 28% of adolescents between the ages of 15 and 19 years11 experience migraine headaches. Surprisingly, however, few US population-based epidemiologic studies have examined the associations between frequent or severe headaches (FSHs) and emotional and behavioral difficulties in this younger population.
Predictors of headaches among children include sleeping difficulties or unusual tiredness, travel sickness, bedwetting, chronic or long-term disease, concentration difficulties, and having a mother with a history of headaches.12 Common personality characteristics and psychological factors have also been noted, including emotional rigidity, tendency to internalize feelings,13 depression,1418 anxiety,14,17,18 a high motivation to achieve,19,20 somatization,14,21,22 sensitivity,23 deliberation,23 pessimism,23 and hyperactivity.17,23
Another common psychological predictor of headaches among children, particularly tension-type headaches, is stress within the family2328 or at school.22,23,2528 Family stressors including marital problems, family bereavement, poor relationships with parents, illness in the family,29,30 and low socioeconomic status have been associated with childhood headaches.3,31 van der Wouden et al26 reported that
20% of all childhood headaches presented to Dutch general practice physicians were linked to psychosocial problems. A study by Metsahonkala et al32 indicated that bullying in school and problems with peer interactions are associated with headaches. Notably, Anttila et al33 found that the incidence of headaches among children aged 6 to 7 years was significantly higher during their first year of school than it had been 6 months before they started school.
Approximately 10% of children with migraines were found to have missed
1 day of school over a 2-week period, and nearly 1% missed 4 days.34 In addition to missing school, children with headaches have reported that pain significantly decreases the amount of leisure time they spend with peers and negatively affects their ability to engage in their usual activities, significantly impacting their QoL.3537
To our knowledge, this is the first study to examine the association between FSHs and emotional and behavioral difficulties among a nationally representative sample of US children based on responses to the extended version (EX) of the Strengths and Difficulties Questionnaire (SDQ). Given the potential clinical and scientific importance of this topic, the purpose of this study is threefold: (1) to determine whether caregivers of children with FSH were more likely to report that the child experienced emotional and behavioral difficulties than caregivers of children without FSH, (2) to determine whether these difficulties had a greater impact on the everyday lives of children with FSH compared with children without FSH, and (3) to compare the use of health care services between children with and without FSH.
| METHODS |
|---|
|
|
|---|
Our research focuses on the 2003 NHIS Sample Child Core and supplemental SDQ-EX data. Data for 9399 children aged 4 to 17 years were collected through interviews with a knowledgeable adult in the children's household. NHIS methods, including the weighting procedure, are described elsewhere.39,40 The FSH status of a child was assessed by asking an adult in the household the following "yes" or "no" question: "During the past 12 months, has [child's name] had frequent or severe headaches, including migraines?"
Information on gender, age group, race/ethnicity, parental structure, poverty level, and health insurance coverage were also included in our analyses. Age group was classified as 4 to 10 and 11 to 17 years. Race/ethnicity was defined as Hispanic, white non-Hispanic, black non-Hispanic, and other non-Hispanic. Parental structure was defined as follows: single parent household (mother and no father or father and no mother), both parents, and neither parent. Mothers and fathers include biological, adoptive, step, and foster relationships. A legal guardian was not considered a parent. Poverty status is categorized based on the ratio of the family's income in the previous calendar year to the appropriate poverty threshold (given the family's size and number of children) defined by the US Census Bureau for that year. Four categories of family income were defined: poverty (<100%), near poverty (100199%), low to middle income (200399%), and high income (
400%). Given the high nonresponse rates for total family income and person earning from employment in the previous calendar year, missing information for these variables were imputed.41 Insurance at the time of interview was classified as follows: Medicaid (children whose only insurance coverage is Medicaid, State Children's Health Insurance Programs, or other state-sponsored plans); private and other insurance (children with private coverage, those with other types of government-sponsored insurance, and those with a combination of public and private insurance); and children with no private or public insurance.42
To examine the associations between childhood FSH and emotional, behavioral, and social adjustment, logistic models were adjusted to account for the total number of comorbid conditions (0, 12, and
3). Three comorbidities, attention deficit disorder or attention deficit/hyperactivity disorder, congenital heart disease/other heart conditions, and asthma were referenced to ever being told by a doctor or other health provider that the child had the condition. The remaining comorbidities, hay fever, respiratory allergies, food or digestive allergies, frequent diarrhea or colitis, and
3 ear infections, were referenced to the past 12 months.
The SDQ-EX, developed and copyrighted by Goodman,43,44 consists of 2 sections. The first section, a brief behavioral screening questionnaire, contains 25 questions that ask an adult in the household about the child's symptoms and positive attributes during the past 6 months in 5 domains: emotional symptoms, conduct problems, hyperactivity-inattention, peer problems, and prosocial behavior.4345 This analysis concentrates on the first 4 domains. Each domain contains 5 questions or items. These items use a 3-point scale to indicate the degree to which each attribute applies to the child (not true = 0, somewhat true = 1, and certainly true = 2). If
3 of the 5 items in the domain contain a response, the remaining questions are prorated. Responses are then summed to produce individual domain scores that are further classified as low difficulties, medium difficulties, and high difficulties. If each of the 4 domains contains a score, the 4 scores are summed to generate a total difficulties score. The scoring algorithms published by Goodman45 on the SDQ Web site were used in these analyses. According to Goodman,45 bands should be developed so that
80% of children are in the low difficulties band, 10% are in the medium difficulties band, and 10% are in the high difficulties bands. Using this methodology, Bourdon et al46 recently developed the US low, medium, and high difficulties bands, which are used in this analysis (emotional symptoms [03, 4, and 510, respectively]; conduct [02, 3, and 410, respectively]; inattention-hyperactivity [05, 6, and 710, respectively]; peer problems [02, 3, and 410, respectively]; and total difficulties score [011, 1215, and 1640, respectively]).
The second section of the SDQ-EX, the impact supplement, contains 4 questions that ask a knowledgeable adult if they perceive the child to have emotional or behavioral difficulties and, if so, inquires further about chronicity, distress, and social impairment. The first 2 screening questions ask the adult (1) whether the child had difficulties with emotions, concentration, behavior, or being able to get along with others (no, yes-minor difficulties, yes-definite difficulties, and yes-severe difficulties) and (2) if so, how long the difficulties have been present (<1 month, 15 months, 612 months, >12 months). If the child had difficulties (minor, definite, or severe) that have been present for
1 month, additional inquiries are made: "Do the difficulties upset or distress the child?" and "Do the difficulties interfere with your child's everyday life in the following areas: Home life? Friendships? Classroom learning? Leisure activities?"43 These questions use a 3-point scale to indicate the degree to which each attribute applies to the child (not at all/a little = 0, a medium amount = 1, and a great deal = 2). If the child is reported not to have difficulties with emotions, concentration, behavior, or being able to get along with others, the questions regarding impact on home life, friendships, classroom learning, leisure activities, and distress are assigned the value of "not at all." The question responses are summed to produce an impact score that is categorized as low impairment, medium impairment, and high impairment (0, 1, and 210, respectively).
Studies of British children have shown that the SDQ discriminates between children with a psychiatric diagnosis and children from the general community.43,44 In addition, a US validity study conducted by Bourdon et al46 noted a significant relationship between a "definite" or "severe" response to the first item on the impact supplement and mental health service use.
Use of health care services was assessed using the following "yes" or "no" questions: "During the past 12 months, have you seen or talked to the following about [child's name] health? . . . A general doctor who treats a variety of illnesses (a doctor in general practice, pediatrician, family medicine or internal medicine)? . . . A mental health professional (psychiatrist, psychologist, psychiatric nurse or clinical social worker)?" Among those who spoke with a general doctor, the following question was asked: "Did you talk to the general doctor because of an emotional or behavioral problem that [child's name] may have?" Additional questions about special education were also asked: "Do any of the children under 18 in this family receive special education or early intervention services?" and "Does [child's name] receive these services because of an emotional or behavioral problem?"
SUDAAN 9.0.0 software (Research Triangle Institute, Research Triangle Park, NC) was used in the analyses to account for the complex sample design and to calculate prevalence estimates, 95% confidence intervals (CIs), unadjusted odds ratios (ORs), and adjusted odds ratios (AORs). For all of the analyses, P values < 0.05 were considered statistically significant. With the exception of Table 1, the models were adjusted by gender, age, race/ethnicity, parent structure, poverty status, type of health care insurance, and number of comorbid conditions. The model for Table 1 only includes sociodemographic characteristics (ie, excluded number of comorbid conditions). After excluding children with missing FSH information (n = 8), sociodemographic characteristics (n = 39), and comorbid conditions (n = 88), the final study consisted of 9264 children.
|
| RESULTS |
|---|
|
|
|---|
A detailed summary of the results of the SDQ behavioral screening questionnaire (first 25 questions), including the individual items, 4 domain scores, and overall prevalence of high difficulties by FSH status can be found in Table 2 and Fig 1. Overall, children with FSH were 3.2 times more likely to have a high level of difficulties compared with children without FSH. More specifically, children with FSH were significantly more likely than children without FSH to have a high level of emotional symptoms (AOR: 3.5), conduct problems (AOR: 2.5), hyperactivity-inattention (AOR: 2.6), and peer problems (AOR: 1.7).
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Notably, a study conducted by Langeveld et al35 indicated that QoL was improved in children who experienced decreased headache intensity and frequency. Unfortunately, however, pain among children and adolescents is one of the most misunderstood, underdiagnosed, and undertreated/untreated medical problems in children.51 Therefore, continued development of effective treatment approaches to alleviate pain and increase QoL and functioning among these children is necessary.
Conversely, psychological variables, such as depression, anger, and anxiety, influence headache prevalence.18,52 Therefore, periodic assessment of possible psychosocial issues and appropriate mental health referrals are also warranted.53
According to Pine et al,16 headaches are twice as common in adolescents who are depressed as in those who are not. Current evidence suggests that cognitive behavioral therapy,54 relaxation training,5456 and self-hypnosis56 are efficacious treatments for recurrent headache in children.
Our results confirm that children with headaches tend to have certain sociodemographic characteristics. For example, we found a significantly increased risk for FSH among older children57 and an inverse relationship between socioeconomic status and headache risk.15,31,58 Although we found no published studies suggesting an association between race/ethnicity and children's risk for headaches, a previous study among adults indicated that non-Hispanic whites are more likely to have migraine headaches than Hispanics.59 However, this study also indicated that white non-Hispanics are significantly more likely than black non-Hispanics to report migraines, a result we did not find.
There are several limitations to our study. First, we were unable to distinguish between types (eg, migraine, tension-type, and cluster), severity, or frequency of headaches experienced by children in our survey. Second, the data are based on parent reports; Goodman60 indicates that teachers and children themselves may report different levels of symptoms and problems. In addition, Goodman60 indicates that multi-informant (parent, teacher, and children) SDQs are optimal for predicting diagnostic status.60 Third, whereas there have been a number of SDQ validity studies conducted in various countries outside of the United States, there have been only 2 validation studies of the total difficulties score of the SDQ for US children.46,61 One of the validation studies, conducted by Dickey and Blumberg,61 suggests that US parents may construe conduct problems and peer problems differently than do European parents, potentially affecting the assessment of psychopathology for these children. Fourth, because our data are cross-sectional, we cannot infer causality between FSH, emotional and behavioral difficulties, and limitations in everyday functioning. Fifth, the questions about health care use only address contact with general doctors, mental health professionals, and special education services and do not address the level of treatment that the child may have received. In addition, there are other potential sources of health care not collected in this study. Finally, one of the questions in the emotional symptoms domain assesses somatization, one of which is headaches. This question was removed to calculate the level of emotional symptoms and the total difficulties score.
| CONCLUSIONS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
Address correspondence to Tara W. Strine, MPH, Division of Violence Prevention, National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, 4770 Buford Hwy NE, Mailstop K60, Atlanta, GA 30341. E-mail: tws2{at}cdc.gov
The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Centers for Disease Control and Prevention.
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Battistutta, R. Aliverti, M. Montico, R. Zin, and M. Carrozzi Chronic Tension-Type Headache in Adolescents. Clinical and Psychological Characteristics Analyzed Through Self- and Parent-Report Questionnaires J. Pediatr. Psychol., October 16, 2008; (2008) jsn102v1. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||