|Home | About | Journals | Submit | Contact Us | Français|
Little is known about how sociodemographic factors relate to children’s chronic pain. This paper describes the pain, health, and sociodemographic characteristics of a cohort of children presenting to an urban tertiary chronic pain clinic and documents the role of age, sex and minority status on pain-related characteristics. A multidisciplinary, tertiary clinic specializing in pediatric chronic pain. Two hundred and nineteen patients and their parents were given questionnaire packets to fill out prior to their intake appointment which included demographic information, clinical information, Child Health Questionnaire – Parent Report, Functional Disability Index – Parent Report, Child Somatization Index – Parent Report, and a Pain Intensity Scale. Additional clinical information was obtained from patients’ medical records via chart review. This clinical sample exhibited compromised functioning in a number of domains, including school attendance, bodily pain, and health compared to normative data. Patients also exhibited high levels of functional disability. Minority children evidenced decreased sleep, increased somatization, higher levels of functional disability, and increased pain intensity compared to Caucasians. Caucasians were more likely to endorse headaches than minorities, and girls were more likely than boys to present with fibromyalgia. Younger children reported better functioning than did teens. The results indicate that sociodemographic factors are significantly associated with several pain-related characteristics in children with chronic pain. Further research must address potential mechanisms of these relationships and applications for treatment.
Many children and adolescents experience chronic pain, often in the form of headache, stomachache and limb pain. Although most affected children are treated at primary care facilities a number continue to experience recurrent or persistent pain and associated disability, and receive treatment at tertiary clinics. Paralleling the adult literature, children affected by chronic pain utilize a high level of health care resources (1) and often experience co-morbid psychological, home, and social difficulties (2,3) Given the array of problems demonstrated by children disabled by chronic pain, there is a need to better understand this population.
The adult literature has demonstrated higher relative risk for chronic pain among certain groups, including women (4) as well as ethnic and racial minority individuals (5). Similar descriptive work is not as generally available for children and adolescents with chronic pain. Although it has been noted that girls are more likely to experience pain and pain-related disability than boys (6), a finding that is consistent with the adult literature, research examining racial or ethnic differences in children’s chronic pain is largely unexplored. From a developmental perspective, this lack of literature specific to children represents a considerable gap in the literature. Childhood and adolescence involves distinct physiological, emotional, cognitive and social challenges that shift over the course of maturation, often precluding the generalization of findings from adult populations. For example, it is known that the nervous system responds differently to pain in early compared to later life, and age-related differences in pain conditions and treatment efficacy indicate the need for developmentally specific research and knowledge regarding pediatric pain.
Amongst adults, the balance of evidence indicates that individuals from racial and ethnic minority backgrounds, including African American and Hispanics and possibly Asians, have increased pain sensitivity and compromised health outcomes compared to Caucasians. In clinical (5) and experimental studies (7,8), pain intensity and unpleasantness are rated higher by African-Americans than by Caucasians. African-American, Hispanic, and Asian outpatients with end-stage chronic and terminal illness were more likely to report pain than were Caucasians (9). Few previous studies have focused on ethnic or racial differences in pain and health status in a sample of pediatric chronic pain patients. One prior study, conducted by our research group, examined anxiety sensitivity and functional status in a subset of the current sample. This earlier work found that minority children reported fewer limitations in family activities and better perceived mental health than Caucasian children (10). However, the present study was concerned with pain and pain-related health outcomes in the clinic.
It was the intention of this paper to describe sociodemographic, pain and health characteristics of a cohort of children presenting to an urban tertiary pediatric chronic pain clinic; and to document the role of sociodemographic factors, including child sex, age and the impact of minority status, on pain and health outcomes in patients. This observational study was designed to serve as a preliminary step in understanding population needs and to highlight potential demographic groups at high-risk.
Participants included 219 pediatric chronic pain patients (70.8% female (n=155) with a mean age of 14.34 years (range = 7 years 9 months – 18 years 3 months, SD=2.49) presenting for treatment at a multidisciplinary, tertiary clinic specializing in pediatric chronic pain. The current sample was obtained from patients with initial clinic visits between February 2003 and October 2007. Presenting pain diagnoses were (percentages add to more than 100% due to multiple diagnoses): 44.7% headaches (n=98) (migraines; myofascial, vascular, tension, stress-related or any other type), 37% functional neurovisceral pain disorder (n=81) (functional bowel, uterine, or bladder disorder), 35.2% myofascial pain (n=77) (excluding headaches), 10% fibromyalgia (n=22), 7.3% complex regional pain syndrome (CRPS) (n=16), and .9% arthritis (n = 2). Average duration of pain symptoms was 45 months (SD=48.32). Almost a third of patients (29.7%, n=65) presented with multiple pain diagnoses.
Total child ethnic/racial composition was .9% (n=3)American Indian/Alaskan Native, 2.2% (n=5) Asian/Native Hawaiian/Pacific Islander, 3.6% (n=8) Black/African American, 13.3% (n=30) Latino/Hispanic, 64.9% (146) Caucasian, and 12.4% (n=28) Other. Further demographic information is presented in table 1.
For analyses comparing ethnic differences, participants were identified as either Caucasian or non-Caucasian (minority status).
Parent-reported marital status was 76.1% married, 11% divorced, 4.8% separated, 4.3% single living alone, 2.9% widowed, 1% single living with a companion. Parents were relatively educated; 3.4% had not completed high school, 5.9% had attained high school education, 34.7% completed some college, 22.4% completed college and 28.3% had a postgraduate qualification. Chi square tests categorizing education into three levels (high school or less, partial or completed college and postgraduate schooling) revealed significant differences between parents of Caucasian and minority children; parents of minority children reported significantly less education than parents of Caucasian children (Chi square = 15.95, p = .00).
The study received ethical approval by the University of California, Los Angeles (UCLA) Institutional Review Board (IRB). Written informed consent was completed by parents, and children provided written assent.
Prior to patients’ initial clinic visit and following verbal consent obtained by telephone, questionnaires were mailed to the home. Questionnaires assessed demographic and health information including measures of the child’s pain and functioning. Only questionnaires pertaining to this study’s aims were included in analyses and discussed below. Questionnaires were completed at home prior to clinic visit without assistance, and questionnaire packets and written informed consent were collected at the clinic intake appointment. All questionnaires were then reviewed by a research assistant and by the evaluating physician during the initial visit in order to ensure completion and clarify ambiguous responses.
A subset of 78 parents were administered additional measures, including the Functional Disability Inventory (FDI) (described below), as well as items about the child’s sleep, and current level of pain. These measures were included in the packet sent to clinic patients with intake dates after August 2005. This subset did not significantly differ from the overall group in age (t = −1.118, p > .05), sex (?2 = .140, p > .05), ethnicity (?2 = 2.553, p > .05), pain duration (t = −.848, p > .05), or parent education (?2 = 2.338, p > .05). The subset included 57 Caucasian subjects and 21 minority subjects. Composition of the minority subset included American Indian/Alaskan Native (n=2), Asian/Native Hawaiian/Pacific Islander (n=2), African American (n=2), Latino/Hispanic (n=12), and Other (n=3).
An author developed demographic information questionnaire was completed by parents, assessing information such as child age, sex, and race/ethnicity. Parents were asked about their level of education and their child’s current schooling situation, including whether the child is in regular full-time school. Responses were categorized according to levels (no high school, some high school, completed high school, some college, college degree and post-graduate degree), and were transformed to continuous variables for regression analyses. Parents also reported on the number of hours their child had slept the previous night.
Information regarding duration of pain symptoms, number and type of pain diagnoses, and psychological diagnoses were obtained from patients’ medical records via chart review. Additionally, information was obtained about school absenteeism and type of schooling (e.g. home or at school, independent study, etc.). These items were obtained from the child and parent during clinic intake interviews and recorded in the assessment portion of the initial evaluation report by the evaluating physician. A standard evaluation form was used to assess all new patients, so that the same group of questions was asked of each patient.
The CHQ PF-28 is designed to assess parents’ measurement of their child’s physical and psychosocial well-being. It has acceptable reliability and validity. The CHQ parent form is comprised of a number of sub-scales. For the purposes of assessing children’s pain and health status, we focused on the bodily pain and global health sub-scales. The bodily pain sub-scale asks about intensity and frequency of child general pain and discomfort and higher scored indicate better functioning, i.e. less pain. The global health sub-scale assesses the parent’s report of the child’s general health status, ranging from excellent to poor, with higher scored indicating better health.
The CSI contains 35 psychophysiological symptoms rated on 5-point scales: 0= not at all; 1= a little; 2= somewhat; 3= pretty much; and 4= very much. Parents indicated the extent to which their children experienced each symptom over the past two weeks. Symptoms include gastrointestinal (e.g. “nausea/upset stomach”), pseudoneurological (e.g. “seizures”), cardiovascular (e.g. “heart beats too fast”), and pain (e.g. “pain in arms or legs”) complaints. Total score can be computed by summing the score across all items. Higher scores indicate higher levels of somatic complaints. Good internal consistency and test-retest reliability have been reported (P=.001) (11).
The FDI is a 14-item questionnaire measuring children’s difficulty in physical and psychosocial functioning due to their physical health during the past two weeks. Parents rated the severity of their child’s functional disability in daily activities including walking, chores, social activities, eating and sleeping. Respondents rate the level of difficulty their child has performing each of the items and total scores are computed by summing the ratings for each item. Higher scores indicate greater disability. The FDI has been documented as valid, stable, and sensitive to change (12). This measure was completed by the subset of 78 patients.
Children were asked to rate their pain on a 10-cm Visual Analogue Scale (VAS) representing a continuum from no pain to worst pain possible. The VAS has been established and widely used as a reliable and valid measure of pain intensity with children. This method measured the amount of pain currently experienced by the child and was completed by the subset of 78 patients.
Bivariate analyses were initially conducted to examine relationships among the study variables prior to multivariate analyses. One sample t-tests for continuous data and chi-square tests for categorical data were used to examine mean differences between the present sample and national norms for the CHQ global health and pain subscales, as well as the child’s school attendance. Chi square tests and t-tests were also used to examine sociodemographic differences on: child pain diagnosis, pain duration, hours slept, CSI, CHQ pain, CHQ global health, FDI, and pain VAS. These measures were tested for differences based on race/ethnicity (Caucasians versus minority status individuals), sex (boys versus girls), and age (children; i.e., participants younger than 12 years, 11 months versus teens; i.e., participants aged 13 – 18 years).
Confirmatory multivariate analyses were conducted on the health and pain-related variables shown to differ based on race/ethnicity in the bivariate analyses. The use of multivariate analyses was intended to assess the relative impact of the child socio-demographic variables on pain and health status, and to determine the relative contribution of minority status to pain and health status after controlling for parent education, as well as child sex and age. Using sequential multiple regression analyses, the independent variables were entered in two steps: the first step consisted of parental education (categorized as a continuous variable), and the second step consisted of age (entered as a continuous variable), sex (boys coded as 0, girls coded as 1) and ethnicity (Caucasian coded as 0; minority status, coded as 1). This regression model was tested for each of the following dependent variables separately: CHQ global health, CSI scores, number of hours slept, FDI scores and VAS pain intensity.
Summary information for the pain clinic population is shown in table 1. Where possible, children’s functioning was compared to national norms. Less than half of children attended full time school, significantly less than the national rate of 91% (US Department of Education, 2003, 2005) (Chi square = 453.5, p = .00). The rate of homeschooling was 22%, ten times higher than the national rate of 2.2%. Children in the pain clinic also demonstrated significantly lower CHQ bodily pain scores (indicating more pain) (mean = 22.09, SD = 23.58) compared to the US norm of 81.7 (s.d =19) (t = −37.41, p <.05). Sample children also scored significantly poorer on the CHQ global health subscale (mean = 45.97, SD = 32.97) compared to the U.S. norm of 73 (s.d = 17.3) (t = −12.55, p <.05) (13). The mean FDI in the present sample was 21.84 (SD=14.04), significantly higher than a mean of 3.15 reported for healthy children (14). Given that a score of 10 and above is considered to be indicative of moderate to high disability (12), children in the present sample had high levels of disability compared to norms. Children in the present sample also scored higher on the CSI (mean = 28.43, SD = 19.80) than published norms from a school sample (mean = 19.02, SD = 17.06), with minority children (mean = 35.49, SD = 24.05) scoring significantly higher than Caucasian children (mean = 28.43, SD = 19.80) (15).
Table 1 also shows ethnic, sex and age differences for the sample. There were no significant differences between Caucasian and minority children with regard to age, sex, school program or number of doctors seen for pain. Boys and girls did not differ significantly on any of these variables. Children and teens did not differ except for type of school program attended (Chi square = 18.9, p< .01), with teens more likely to be home schooled than children.
As shown in table 1, there were a number of significant differences between groups on pain and functioning variables. Overall, minority children evidenced significantly compromised functioning compared to Caucasians. Thus, minority children had less parent-reported sleep the previous night, increased somatization, decreased global health, increased functional disability compared to Caucasian children, and increased child self-reported VAS pain intensity. However, Caucasian children were significantly more likely to experience headaches than minority children (Chi square = 4.14, p = .04).
In terms of sex differences, girls had significantly higher VAS pain intensity than did boys. Girls were also more likely to experience fibromyalgia than boys (Chi square = 4.76, p = .03). When examining age-group differences, teens displayed significantly more comprised functioning compared to younger children on most of the parent and self-report scales. Teens were also significantly more likely to experience myofascial pain compared to children (Chi square = 14.46, p = .00).
The results of the sequential linear regression analyses are shown in table 2.
For CHQ global health scores, age and ethnicity emerged as significant predictors after controlling for parental education, with minority status and older age associated with poorer health. Child sex was not a significant predictor of CHQ scores. For CSI scores and child sleep, neither sex nor age was significant; only ethnicity emerged as a significant predictor, with minority status associated with higher somatization and less hours of sleep. For FDI scores, only age was a significant predictor such that older children had more disability. For pain VAS, only sex was a significant predictor, with girls more likely to experience higher levels of pain.
The study provides a clinical profile of children presenting to an urban, tertiary chronic pain clinic. The cohort was largely female and had experienced pain for greater than three years. Children had significantly impaired health and high levels of pain. Based on national norms, a disproportionate number of children did not attend regular school and were home schooled. We identified minority status as a significant variable in a number of health-related characteristics.
Less than half of children were currently attending regular school. This is consistent with previous reports showing that chronic pain interferes with children’s academic lives.
Rates of school absence for chronic pain is at least as high or higher than absences due to other illnesses, with specific chronic pain conditions such as irritable bowel syndrome and chronic juvenile arthritis linked to low school performance (16). Studies examining children reporting to chronic pain clinics have found over 50% of children not attending school full time including many children missing school for months at a time (3).
Consistent with previous reports (17) girls experienced more pain than boys. Girls were more likely to present to the clinic with chronic pain (demonstrated by a higher number of female than male patients), had a greater incidence of fibromyalgia and reported greater VAS pain intensity, although boys had pain of longer duration than girls. Our findings are consistent with adult literature showing that females are more likely to experience clinical pain and to utilize health care facilities than are males (4, 18).
The clinical profile varied most consistently as a function of minority status and age. Consistent with previous clinical studies, adolescents were at relative risk compared to younger children (18). Univariate results also indicated that minority children had worse functioning than Caucasian children, including reduced hours of sleep and global health as well as increased somatization, disability and pain intensity. However, non-minority youth were more likely to report headaches than minority youth, consistent with previous reports examining prevalence of headache and other pain locations in white versus African American youth (19). Our findings are similar to other literature examining racial/ethnic differences in children’s pain. One study found that youth from a minority group reported more physical symptoms than Whites (20), while a community based study reported a high rate of physical symptoms amongst minority youth (21). However, this is the first time that ethnic differences in children’s chronic pain have been noted. The findings are also in accord with adult literature suggesting that minority status is a risk factor for compromised pain and health states (5,22,23).
Of note, we found no significant differences in pain intensity between minority and non-minority children in multivariate analyses when child sex, age and parent education were also considered. Previous research examining racial/ethnic differences in adult clinical pain have also reported differential results. Thus, minority status has been most consistently linked to dimensions of pain related to unpleasantness rather than intensity (5). We may have found stronger ethnic differences with questions more specifically targeting pain unpleasantness. Clearly, further research examining pain in children of different backgrounds is warranted. Further research is also needed to examine potential mechanisms of racial/ethnic disparities in health and pain outcomes. After controlling for parent education, ethnic differences remained in children’s health, somatization and sleep, however pain intensity no longer differed as a function of minority status. Thus, it is possible that at least for pain intensity, other factors such as child sex are more important. It is also possible that had we accounted for other socioeconomic status (SES) indicators, including income, health disparities would have reduced further (24). Cultural differences in the meaning and interpretation of pain may also play a role.
A number of limitations are evident in the design and execution of the study. An examination of mechanisms of impact is required before our findings related to health disparities in children’s chronic pain can be meaningfully interpreted. In addition, while the relationship between minority status and clinical outcomes in a pediatric chronic pain cohort is a novel finding, our results do not support a strong role for socio-demographic variables in child pain outcomes. Thus, the total variance accounted for by child demographic factors was between 7 and 12% (see table 2). These findings suggest a small, but significant role for child demographic variables in chronic pain.
A further limitation of the study is the inclusion of a single ethnic/racial minority dimension. In reality, African American children may experience different health and pain outcomes compared to Latinos, Asians and children from other minority backgrounds. In the present study we grouped these children together, for practical reasons due to the small numbers of children representing each background and due to the conceptual likelihood that children shared similarities in minority status distinct from Caucasians. For measures that were only administered to the subset of 78 participants, the low number of the minority group is another limitation. Another limitation was reliance upon a convenience cross-sectional design. We are unable to draw conclusions about how sociodemographic variables such as minority status, sex and age influence children’s health and pain over time. The findings also rely primarily on parent reports of children’s pain and health. While parent reports often provide meaningful and relevant information about children’s health-related functioning, only moderate levels of association between parent and child reports have been found (25).
Our findings that minority status predicted children’s health, somatization and sleep quantity even after controlling for disparities in parent education and including child sex and age in multivariate analyses suggests that important differences may exist in the presentation and ultimately, the treatment of pain experienced by children from diverse backgrounds. Growing levels of diversity in patient populations mean that substantive information regarding disparity issues relevant to clinical practice is required (19). Combining understanding of patient background with knowledge of specific problems in key populations will ensure optimal models of clinical care. Treatments can potentially be customized to better fit each patient through consideration of background variables such as race/ethnicity and sex.
Additional research is required to further identify sex and racial/ethnic disparities in pain. Only equipped with a thorough understanding of the issues presented by each sex and by distinct racial and ethnic groups is it possible to ensure that each patient receives the highest quality of treatment. Chronic pain problems are often poorly understood and mismanaged, resulting in a drain on both the healthcare system and the resources of families dealing with chronic pain. Using socio-demographic variables to predict and comprehend unique concerns to each patient helps ensure that every patient has equitable access to the highest levels of care.
This study was supported by 2R01DE012754 (PI: Lonnie K. Zeltzer) awarded by the National Institute of Dental and Craniofacial Research and by 1R01MH063779, awarded by the National Institute of Mental Health.