Pediatric chronic pain and caregiver burden in a national survey

1 Introduction

Pediatric chronic pain is characterized as pain which persists or recurs over 3 or more months, and is associated with adverse outcomes such as decreased functional status and high rates of emergency care utilization [1]. Previous studies have differed in the estimated incidence of of pediatric chronic pain, varying from 3% to 46% [2], [3], [4], [5], [6]. Our group’s recent analysis of a nationally representative survey suggests that 6% of children experience functional difficulties due to chronic pain, with a higher prevalence among adolescents and children from socioeconomically disadvantaged families [1]. In addition to physical impairment, pediatric chronic pain has been associated with negative emotional, social and academic outcomes [7], [8].

Due to these consequences, parents of children with chronic pain may experience additional caregiver burdens as they manage their child’s chronic pain and associated comorbidities. For example, parents are responsible for transporting their children to appointments as well as attending most appointments and paying for health care costs that are not covered by insurance [9]. Furthermore, parents are responsible for educating themselves about their child’s condition and available treatment options. These responsibilities can contribute to parents’ reduced quality of life, emotional difficulties, out-of-pocket cost, and reduced participation in the labor force [10], [11]. Prior studies have reported that parents of children with chronic pain have significant caregiver burdens [1], [9]. Several studies have sought to determine which caregivers are especially at risk. Factors that may predict increased caregiver burden include lack of control over the child’s pain, type and level of pain, increased dependence on caregivers, limited access to interventional resources, and the child’s age and gender [10], [12], [13], [14], [15]. However, prior research in this area has generally been limited by the inclusion of families seen at a single institution, whereas nationally representative and generalizable data on caregiver burden among children with chronic pain remain scarce. Identifying families for whom a child’s chronic pain imposes significant caregiver burden may help target these families for interventions that alleviate caregiver distress [13], [14]. Addressing caregiver burden may also improve parents’ efficacy at managing their children’s chronic pain.

To better understand caregiver burden in families where a child has chronic pain, we performed a secondary analysis of the 2016 National Survey of Children’s Health (NSCH). Our primary aim was to provide population estimates of caregiver burden by measuring time use, reduced labor force participation, and out-of-pocket cost among parents of adolescents with chronic pain. Our secondary aim was to identify child and caregiver factors associated with increased caregiver burden.

2 Methods

The study was deemed exempt from review by the Institutional Review Board at Nationwide Children’s Hospital. We used de-identified, publicly-available data from the 2016 NSCH, a nationally representative cross-sectional paper and web survey directed by the US Health Resources and Services Administration [16]. The survey screened households to identify eligible children ages 0–17 years and randomly selected one child per household about whom an adult caregiver completed a detailed questionnaire. Survey weights were provided to allow generalization to the relevant national population; in this case, the subpopulation of children with chronic pain. We have previously used the 2016 NSCH data to describe the prevalence of chronic pain and differences in health care utilization and costs between children with and without chronic pain [1]. As chronic pain was most commonly reported in adolescents, we limited the present study to adolescents ages 12–17 years who had chronic pain (defined as repeated or chronic physical pain, including headaches or other back or body pain during the past 12 months). We included children with complete data on caregiver burden outcomes and study covariates, as described further below. To assure comparability of reported caregiver burden across respondents, we excluded children whose caregiver completing the survey was not one of the child’s parents.

We examined three dimensions of caregiver burden, according to outcomes emphasized in prior research and the availability of appropriate data in the NSCH. These included: (1) total time spent providing or coordinating health care for the child in an average week (including time spent by other family members); (2) reduced labor force participation (whether the parent or other family members stopped working or cut down on hours worked because of their child’s health); and (3) out-of-pocket costs associated with the child’s health care for during the past 12 months. Time use and reduced labor force participation were examined as binary measures. The time use measure was categorized as spending at least 1 h per week on either providing or coordinating health care, vs. spending <1 h per week on both types of activities. Costs were analyzed as an ordinal measure, categorized as $0, $1–$249, $250–$499, $500–$999, $1,000–$5,000, and >$5,000. For each measure of caregiver burden, we sought to determine if it was associated with the adolescent’s health status and health care needs; demographic characteristics; and socioeconomic status (SES).

Adolescents’ demographic characteristics included age, gender, and race/ethnicity. Family demographic characteristics included parent age, gender, and marital status; language spoken in the home; and the state of residence. SES measures included the adolescent’s health insurance coverage, family income in the last calendar year [percent of the Federal poverty line (FPL)], parents’ highest level of educational attainment, and whether the household experienced food insecurity, defined as sometimes or often being unable to afford enough food to eat. Adolescent health characteristics included general health (from 1=excellent to 5=poor); special health care needs (SHCN) status, defined as requiring health services beyond those required by children generally [17]; a history of emergency department visits in the past 12 months; and need for coordination of care between multiple providers [18]. Adolescents were classified as requiring care coordination if the parent reported that someone helped them coordinate care among different doctors, or that they felt they could use extra help coordinating their child’s care. We also controlled for comorbidities present at the time of the survey. Specific physical conditions included allergies and asthma; while specific mental health conditions included anxiety, depression, behavioral or conduct problems, developmental delay, speech disorder, learning disability, autism, and attention deficit/hyperactivity disorder (ADHD). Non-specific conditions and conditions present in <5% of the sample were grouped together as other physical conditions or other mental health conditions [1].

3 Results

The 2016 NSCH sample included 2,262 adolescents age 12–17 with chronic pain. We excluded 250 surveys completed by someone other than a parent, 64 surveys missing data on study outcomes, and 237 surveys missing data on study covariates. The remaining responses represented 1,711 adolescents. In 258 cases, parents reported spending at least 1 h/week on their child’s health care (weighted percentage: 15%), in 214 cases, parents reported reduced labor force participation (14%), and in 802 cases, parents reported spending ≥$500 on their child’s health care in the past 12 months (36%; Table 1). Weighted estimates based on the NSCH sample suggested that adolescents with chronic pain were 61% female, 11% African American, and 27% Hispanic. Twenty-seven percent were living in households below the poverty line. Adolescents with chronic pain generally had high needs for health care, as evidenced by approximately half (48%) meeting the definition for SHCN, 33% having visited the ED in the past 12 months, and a majority (56%) requiring coordination of care between multiple physicians or services. The most common physical comorbidities were allergies (42%) and asthma (20%). The most common mental health conditions were anxiety (27%), depression (18%), and ADHD (18%).

The multivariable models of study outcomes are shown in Tables 2–4. Parents of adolescents with greater health care needs reported greater time spent on their child’s health care. For example, caregivers were more likely to spend at least 1 h/week on their child’s health care if they reported their child’s health status as fair or poor [odds ratio (OR)=7.5; 95% CI: 2.5, 23.0], if their child had SHCN (OR=5.7; 95% CI: 2.3, 14.2), or if their child required care coordination (OR=2.2; 95% CI: 1.2, 4.0). Adolescents’ fair or poor health status was also associated with a greater likelihood of family members reducing labor force participation to care for the child. However, measures of adolescents’ health status and medical complexity were not consistently associated with reporting higher out-of-pocket costs. For example, reported out-of-pocket costs tended to be higher for children who have recently visited the ED (comparing one visit to none), but were surprisingly lower among adolescents who needed coordination of care between multiple providers.

Considering demographic characteristics, these were not associated with time spent on health care; although younger parents, parents who were not married, and parents who spoke a language other than English were more likely to report reduced labor force participation. Parents of girls with chronic pain were less likely to report reduced labor force participation as compared to parents of boys (OR=0.4, 95% CI: 0.2, 0.8), and parents of African American adolescents with chronic pain reported lower out-of-pocket costs as compared to parents of Caucasian adolescents. Socioeconomic differences were inconsistent between outcomes of caregiver burden or measures of SES. For example, spending at least 1 h/week on the adolescent’s health care was more common among parents with a graduate or professional degree (OR=2.7; 95% CI: 1.2, 6.0), but also among parents reporting food scarcity in the household (OR=3.2; 95% CI: 1.4, 7.5). By several measures of SES, parents in advantaged families were more likely to report reducing their labor participation (e.g. if they had a graduate or professional degree, or if they had a household income of 200–399% as compared to <100% FPL). Parents with higher income and private insurance also reported higher out-of-pocket costs associated with the child’s health care.

4 Discussion

Caregiver burden is a recognized challenge for families with children who have health problems or SHCN [19], [20], [21]. Caring for children with chronic pain can be particularly stressful because pediatric chronic pain is frequently associated with diagnostic uncertainty and a lack of care options [22]. Our analysis of data from a national survey indicates that on measures of time, cost, and out-of-pocket expenses, parents of adolescents with chronic pain incur significant caregiver burden. Furthermore, greater burdens are found among parents of adolescents with worse overall health status and greater health care-related needs. Conversely, SES and demographic differences in caregiver burden are inconsistent within this population, reveal no clear disparity in caregiver burden for economically disadvantaged families, and do not offer clear targets for intervention. Given the high levels of caregiver burden and high needs for care coordination or special health care services among adolescents in this study, coordination of chronic pain treatment with other services may provide an opportunity to better support the families of children with chronic pain.

In a previous study, parents described the pain their child experiences as unbearable, and mentioned consequent restrictions on their social life, including loss of personal time, missed work, and significant economic burden related to physician contacts, medication use, and hospitalizations [23]. Another study found that 65% of employed parents of children with chronic pain reported work absenteeism and 22% reported high or very high financial burden during the past 6 months [12]. These results may have been representative of children seen by specialized pain clinics, whereas our analysis suggests caregiver burden is also common among parents of adolescents with chronic pain recruited to participate in a population-based survey. In this sample, caregiver burden was highest among parents of children with significant morbidity or health care needs. While this association may have been partially due to other comorbidities or disabilities among adolescents with chronic pain, it may also be related to under-treatment of chronic pain and greater resulting impairment in adolescents’ functioning. Therefore, measuring improvement in caregiver burden may be an important outcome in future studies evaluating interventions to treat adolescent chronic pain. Furthermore, high levels of caregiver burden seen among adolescents with chronic pain suggest that long-term outcomes for both patients and families are likely to depend on the quality of care received for chronic pain at this stage of the life course.

Prior studies have reported specific demographic and socioeconomic disparities in the extent of caregiver burden among parents of children with chronic pain. In one previous study, caregiver burden was greater for parents of older children (school-age and teenage) [10]. Our study of adolescents did not find a consistent correlation between the child’s age or gender and increased caregiver burden. On the labor force participation measure, our analysis showed that parents of a girl with chronic pain were in fact less likely to reduce work participation than parents of a boy with chronic pain. Considering SES, one study found that families with lower SES experienced decreased caregiver burden in terms of disturbance in family interaction as compared to those above poverty line [21]. However other previous studies have reported that parents with lower SES experience increased caregiver burden [19], [20]. We found no consistent association between socioeconomic disadvantage and caregiver burden; on the contrary, parents with higher income or educational attainment reported lower labor force participation and higher out-of-pocket costs, related to providing health care for adolescents with chronic pain. It is important to note that the NSCH measure of labor force participation may be biased towards voluntary reduction in work hours or job resignation, whereas involuntary unemployment is plausibly more common among parents of lower SES who have a child with chronic pain.

Specific interventions may reduce caregiver burden for parents of children with chronic pain. Referral to a multidisciplinary chronic pain clinic can improve management of chronic pain in children whose pain is not adequately managed by primary or specialty care providers [24], [25]. Several evaluations of multidisciplinary pain clinics have reported favorable effects on reducing caregiver burden [26]. Additional considerations include providing families with more detailed preparatory information, coping strategies, and psychosocial intervention to address pediatric chronic pain and associated caregiver burden [13]. In another study, Internet-delivered cognitive-behavior therapy for families of children with chronic pain led to reductions in parental depression, anxiety, self-blame about their adolescent’s pain, and in their maladaptive behavioral responses to their adolescent’s pain [14]. Additionally, addressing the need for care coordination among parents of children with chronic pain may also be beneficial. Particularly, care coordination for adolescents with chronic pain could include coordination of specialized pain treatment with other services used by these children, given that over half of adolescents in this population require coordination of care between multiple providers. Our results regarding factors associated with caregiver burden suggest that such initiatives could be most beneficial for adolescents with chronic pain who have worse overall health and greater health care needs.

The conclusions from our study are limited by some aspects of the data set and analytic approach. We were unable to account for all factors that may predict increased caregiver burden in pediatric chronic pain, including parents’ perceived control over the child’s pain, the type and severity of pain, and limited access to interventional resources [10], [12], [13], [14], [15]. Additionally, we have used a relatively restrictive definition of chronic pain, which applies to only 6% of children in the United States [1]. When considering previous studies that have used more expansive definitions of chronic pain, our results may not be representative of adolescents whose pain was less severe or did not cause functional difficulty [2]. A further limitation in our analysis relates to the wording of questions about caregiver burden, which included time use and costs incurred by anyone in the family caring for the adolescent with chronic pain, and were not specifically limited to the parent answering the survey. In our statistical analysis, we applied survey weights to obtain nationally representative estimates, but it should be noted that due to the design of the NSCH, these estimates are meant to be representative of the population of adolescents with chronic pain, and might not be representative of the total population of adults who care for these adolescents [19]. Future studies on the epidemiology of pediatric chronic pain may combine the NSCH’s population-based design with more detailed measurements of pain characteristics and access to pain treatment, to identify families in which caregiver burden associated with pediatric chronic pain could be successfully alleviated through improved pain management and increased access to health care resources.

In summary, we analyzed nationally-representative survey data to describe high levels of health care needs and caregiver burden among adolescents with chronic pain in the United States. Adolescent health status and health care needs were associated with increased caregiver burden as reported by their parents; whereas no consistent association was found between caregiver burden and demographic factors or socioeconomic disadvantage. Recognizing families in which pediatric chronic pain may impose significant caregiver burden could aid the health care team in addressing parents’ needs in alleviating distress, maintaining their own quality of life, and providing effective care to a child with chronic pain.