Journal Logo


The epidemiology of chronic pain in children and adolescents revisited: A systematic review

King, Saraa,*; Chambers, Christine T.b; Huguet, Annab; MacNevin, Rebecca C.b; McGrath, Patrick J.b; Parker, Louiseb; MacDonald, Amanda J.c

Author Information
doi: 10.1016/j.pain.2011.07.016
  • Free
  • Press Release


1. Introduction

Although precise definitions vary, the International Association for the Study of Pain typically defines chronic pain as pain lasting more than 3months [24]. Chronic pain can be either recurrent or persistent and is a serious developmental health concern that can interfere significantly with daily functioning [29,30]. Children experiencing persistent or recurrent chronic pain (hereafter referred to as chronic pain) may miss school, withdraw from social activities, and are at risk of developing internalizing symptoms in response to their pain [23,26,29]. Given these consequences, issues related to pediatric chronic pain (eg, prevalence, age of onset, epidemiology, etiology, assessment, and treatment) are important to researchers and clinicians to enable the development of effective strategies to ameliorate these problems.

Studies examining the prevalence of childhood pain have yielded inconsistent findings. For example, population-based studies of headache in children have found prevalence rates estimated anywhere from 23% to 51% [10,31,38]. Similarly, prevalence rates for abdominal pain have been shown to range between 10% and 27% [15]. Because prevalence rates vary considerably across studies, it is difficult to make general conclusions regarding the pervasiveness of different pains in children and adolescents. There are few comprehensive summaries of pain epidemiology in childhood and adolescence. The most recent comprehensive review was published in PAIN® 20years ago [15] and was the first and only paper to provide such a summary. Goodman and McGrath [15] also identified several methodological limitations of these studies that could account, at least in part, for discrepant findings with respect to pain prevalence rates. The use of inappropriate measures, disagreement regarding the definition of various types of pains, and the use of samples inappropriate for epidemiological investigations were noted as methodological concerns [15].

Since Goodman and McGrath’s pivotal review, significant improvements have been made to study design (eg, use of population-based recruitment, more stringent controls) and a number of epidemiological studies of pediatric pain have been published [31,38]. However, no comprehensive reviews of this epidemiological research have been published since 1991, making it difficult to describe prevalence rates of chronic pain in children and adolescents. The objectives of the current review were 4-fold: (1) to provide an updated examination of epidemiological studies of pediatric chronic pain; (2) to use both systematic and narrative methods to provide a more objective summary of the literature; (3) to describe individual factors related to pain prevalence (eg, age and sex differences and sociodemographic and psychological factors); and (4) to assess study quality as part of the current review to determine whether improvements have been made to methodology since the original review [15] was published. As noted by Goodman and McGrath [15], a more thorough understanding of the epidemiology of pediatric pain is beneficial for the treatment of pain in children and adolescents, and for understanding the antecedents and correlates of adult chronic pain.

2. Method

2.1. Inclusion/exclusion criteria

Key electronic database searches (EMBASE, Medline, CINAHL, and PsycINFO) were conducted by a Library Information Specialist using the following search terms: child, adolescent, chronic pain, pediatric pain, recurrent pain, prevalence, incidence, pain epidemiology, headache, abdominal pain, musculoskeletal pain, and back pain. Databases were searched for articles published in English or French between 1991 and 2009. No French articles were located for this time period; therefore, no non-English papers were included in the current review. Two independent reviewers (SK, RM) performed an initial screening using study titles and abstracts. The following details were recorded: number of studies found, number meeting inclusion criteria, number excluded, reason for exclusion, and whether the article needed to be obtained to determine eligibility. Articles that were included by one reviewer but excluded by another were to be decided based on consensus, with disagreements to be settled by a third reviewer (the third reviewer was not required). Once agreement had been reached on eligibility, the same 2 independent researchers reviewed each article in full to determine the significant findings in each study.

2.2. Data extraction

Data were extracted by the 2 independent reviewers (SK, RM) using customized forms specifying author, type of sample (population/community or school), number of participants, location of study, age range of participants, sex of participants, prevalence rate, and other findings of note (eg, socioeconomic differences in pain prevalence, increasing prevalence rates over time, associated psychosocial variables such as anxiety and depression). Studies meeting inclusion criteria were categorized according to the type of pain investigated, as follows: (1) headache only; (2) abdominal pain only; (3) back pain only; (4) musculoskeletal pain; (5) combined pain; and (6) other/general pain. The “other/general pain” category was included to capture recurrent pains other than those previously described (eg, recurrent earaches or throat pain) or studies in which children were asked whether they had experienced any kind of pain in a specified time period.

2.3. Assessment of study quality

A set of criteria for evaluating the quality of epidemiological studies of pain was compiled using existing sources [13,21,22,44], as well as other quality criteria for epidemiological studies (ie, Strengthening of the Reporting of Observational Studies in Epidemiology [STROBE]; [42]) as guidelines. The decision to develop a new set of criteria rather than use the existing STROBE criteria was due to the need to include items specific to the study of pain and to address the concerns raised by Goodman and McGrath [15]. The final criteria consisted of a total of 19 items; however, 2 of these items were not appropriate for those studies using self-administered questionnaires instead of interviews as the primary data collection method. Each item was scored as “1” if the criterion was met and “0” if the criterion was not met. Items scored as “unclear” were rated by a second reviewer. Therefore, quality scores for epidemiological studies using interview methodology could range from 0 to 19, whereas quality scores for studies using self-report measures could range from 0 to 17, with higher scores indicating higher study quality. Each study included in this review was coded on each of the quality criteria by one of the study coauthors (AH). Twenty-five percent of these studies were randomly selected to be coded by the lead author (SK) to establish percent agreement between coders. Coders were not blind to study authors or study findings.

3. Results

3.1. Search results

The database search yielded a total of 185 citation abstracts, with 21 duplicates, 4 book chapters, 15 editorials/commentaries, 24 review articles, and 1 dissertation. Based on the initial abstract review, 58 studies were retrieved for further review; of these abstracts, a total of 32 studies met our a priori inclusion criteria and were included in this review (Fig. 1). Given that the goal of the current review was to summarize epidemiological findings, a typical meta-analysis using effect sizes was not possible; therefore, findings were synthesized and described systematically. Given the discrepant findings with respect to prevalence rates across studies, the median prevalence rate was used to summarize findings, as this statistic is not influenced by outliers.

Fig. 1.:
Search strategy for included studies.

3.2. Headache studies

3.2.1. Prevalence

As shown in Table 1, 9 studies (n=24,230) specifically examined the prevalence of headache in children. Three studies (n=5516) examined headache in the context of larger studies of multiple pains. In total, 29,746 participants were included in studies of headache. Prevalence rates ranged widely for this type of pain and were largely dependent on the time period of reporting (ie, monthly, weekly, or daily prevalence rates). Three studies investigated prevalence rates over the past month [12,32] or 3months [10,14], with prevalence rates ranging from 26% to 69% in children aged 7 through 16years (median=47%). Weekly headache prevalence rates in children aged between 7 and 18years of age were reported in 8 studies [4,6,10,12,20,32,34,38]. Across these studies, prevalence of weekly headache ranged from 6% to 31% (median=23%). Finally, 2 studies presented prevalence rates for daily headache in children aged between 10 and 18years [6,34], with rates ranging from 1% to 9% (median=5.1%).

Table 1:
Studies of headache prevalence.

Four headache studies reported prevalence rates for migraine [1,6,11,36] and 3 reported prevalence rates for tension-type headaches [1,6,11]. Prevalence rates for migraine with and without aura ranged from 3% to 10% (median=8%), whereas prevalence rates for tension-type headache ranged much more widely, from 1% to 73% (median=25%).

3.2.2. Age differences

Age appears to be an important influence on headache prevalence among children. Four studies reported that headache prevalence increased with increasing age in children and adolescents [11,12,14,32], whereas one study reported that headaches were more prevalent in younger children [20]. Carlsson [12] reported a gradual increase in prevalence of headache “one a month or more” with increasing age, from 16% in grade 1 to 42% in grade 9. These authors also reported a marked increase in headache prevalence between grades 2 and 3, with prevalence rates increasing from 3% to 11%. In their study of headache in a sample of German children and adolescents, Fendrich et al. [14] also found that headache prevalence increased with age; however, this trend was observed in girls only; a similar trend was reported by Petersen et al. [32]. Perhaps the most worrisome finding with respect to headache occurrence in children and adolescents is that prevalence rates appear to be increasing over long periods of time, independent of increasing age of a particular cohort. In a study specifically designed to investigate changes in headache prevalence over time, Sillanpaa and Anttila [36] found that, over an 18-year period (ie, 1974–1992), the prevalence of headache over the last 6months increased from 14% to 52% in 2 samples of Finnish children matched on age and demographics.

3.2.3. Sex differences

Of the 10 studies that reported sex differences in headache prevalence rates [4,6,9–12,14,20,32,34,38], the majority reported that headaches are more prevalent in girls. Prevalence rates of headache were generally found to be higher in girls than in boys in both elementary and high school students, with more discrepant rates emerging in older children [4,10–12,32,34,38]. Given that these trends were observed in studies conducted in several countries, and that they occur across various samples and ages, it is likely that the finding of higher prevalence of headache in girls is not spurious.

3.2.4. Psychosocial factors

One study reported that increased headache prevalence is associated with low socioeconomic status (SES) and other family variables [11]. Headache prevalence was higher in children from low SES backgrounds, especially if mothers had a low level of education and if there was a positive family history of headache [11]. Low SES, daycare attendance as a youngster, and increased number of leisure activities were also identified as risk factors for headaches [36]. The relation between sociodemographic factors and headache was less clear in other studies examining these variables [12,20,36]. Generally, these studies found that SES tended to interact with factors such as age and sex to influence headache prevalence in children. In addition to SES, other factors associated with increased headache prevalence include ethnicity [34], depression [34,38], anxiety [38], and low self-esteem [34]. Relations between pain and sex, age, and other psychosocial/sociodemographic factors (as described below) across studies are presented in Table 2.

Table 2:
Summary of prevalence rates by pain type.

3.3. Abdominal pain studies

3.3.1. Prevalence

As shown in Table 3, 5 studies (n=21,420) specifically examined the prevalence of abdominal pain in children, and a further 4 examined abdominal pain in the context of wider investigations of multiple pains (n=6106), making the total sample size 27,526 children and adolescents. Of these 9 studies, 5 investigated prevalence rates of recurrent abdominal pain (RAP) as defined by Apley and Naish [2]. According to Apley and Naish [2], RAP can be defined as at least 3 episodes of abdominal pain, severe enough to limit the child’s functioning, over at least 3months. Across the studies examining RAP [7,8,27,28,33], prevalence rates varied fairly widely in children aged between 2 and 17years, ranging from 41% in a small sample of Malaysian 11- and 12-year-olds [7], 33.2% in a sample of German preschoolers [28], to 4% in a sample of 2-year-old children in the UK [33]. The median prevalence rate of RAP across these studies was 12%. Four studies examined prevalence of abdominal pain in various samples of children and adolescents [10,18,32,38]. Kristjansdottir [18] measured overall lifetime prevalence of stomach pain in Icelandic 11–12- and 15–16-year-olds and found that 53% of the sample had experienced “frequent” stomach pain [18]. Four studies examined weekly abdominal pain prevalence [10,18,32,38], with rates ranging from 8% to 22% (median=19%).

Table 3:
Studies of abdominal pain prevalence.

3.3.2. Age

With respect to age differences in prevalence of RAP, Ramchandani et al. [33] reported that the prevalence of RAP increases with age in early childhood. Using data from a large longitudinal cohort study in the UK, these authors found that prevalence rates of RAP increased from 4% in 2-year-old children to 7% in 3-year-old children, to 12% in 6-year-old children (median=7%). With respect to age-related differences in stomach pain, Kristjansdottir [18] reported that the prevalence of stomach pain was significantly higher in younger (23%) as opposed to older children (14%) in 11–12- and 15–16-year-old children from Iceland. Similar findings were reported by Petersen et al. [32], who noted that prevalence rates for monthly stomach ache increased from 34% to 55% over grades 0 through 6. In contrast to these findings, Boey and colleagues [8] did not find age-related differences in the prevalence of RAP in a sample of Malaysian 11–16-year-olds. These authors also noted that their prevalence rate of 10% was similar to that reported by Apley and Naish [2]. Similarly, in another study of RAP in Southeast Asian children and adolescents, Oh and colleagues [27] reported that prevalence rates of RAP did not increase with increasing age, although these authors noted a prevalence rate of 23% in their sample – more than twice that of Boey et al. [8].

3.3.3. Sex

As with headache, sex differences in the prevalence of RAP and stomach pain were observed across these studies. Based on findings from a Canadian sample, Stanford et al. [38] reported that girls tend to have more pain complaints, including stomach ache, than boys at all ages. Kristjansdottir [18] reported that the prevalence rate of stomach pain was higher in girls overall and that this sex difference was observed in both younger and older children. Specifically, younger girls had a “weekly” prevalence rate of 27% compared to 19% in younger boys. Older girls had a “weekly” prevalence rate of 18% as opposed to 10% in older boys. Older girls reported higher prevalence rates of “monthly” stomach pain as opposed to younger girls; 49% and 36%, respectively). Ramchandani and colleagues [33] reported a clear sex difference in RAP, with more girls than boys reporting symptoms, as did Brun Sunblad et al. [10]; 10% of girls in their sample reported weekly abdominal pain, as compared to 5% of boys. Oh and colleagues [27] reported similar findings, in that RAP was more common in girls than in boys (odds ratio=1.7, 95% confidence interval 1.3–2.1). On the other hand, Boey et al. [8] reported that prevalence of RAP in Malaysian children and adolescents did not differ by sex.

3.3.4. Psychosocial factors

Across studies, findings were mixed with respect to the sociodemographic factors associated with abdominal pain in children and adolescents. In one study that specifically examined sociodemographic factors related to abdominal pain [18], it was found that SES interacted with age such that abdominal pain was more prevalent in younger children from lower and middle SES backgrounds. One other study found abdominal pain to be more prevalent in rural children [8], whereas other authors reported higher prevalence rates of abdominal pain in children from higher social classes [33]. Other factors associated with increased abdominal pain prevalence are anxiety in children and their mothers [33], anxiety and depression in children [38], feelings of sadness in younger girls [10], and school stress [27].

3.4. Back pain studies

3.4.1. Prevalence

As shown in Table 4, 2 studies (n=3619) specifically examined the prevalence of back pain in children and adolescents. Two studies examined back pain in the context of a larger study investigating multiple pains (n=3643), making the total number of participants 7262. As with studies of headache prevalence, reported rates of low back pain varied and were dependent on the time period of reporting. Two studies reported 1-month prevalence rates of back pain ranging from 18% to 24% (median=21%) in samples of English and Swedish children [32,43]. Finally, 3 studies reported prevalence rates of weekly or “at least weekly” back pain ranging from 9% to 25% (median=21%) [17,32,38].

Table 4:
Studies of back pain prevalence.

3.4.2. Age

All studies of back pain in children and adolescents included in the current review reported increasing prevalence rates with age in this population. Kristjansdottir [17] further noted that older children in rural areas tended to report more back pain than children in more urban areas (28% and 23%, respectively).

3.4.3. Sex

Sex differences were not as pronounced in studies of back pain as in headache and RAP/stomach pain studies, as only one of the above-mentioned studies noted sex differences with respect to back pain. Watson et al. [43] reported that the 1-month prevalence rate of back pain in a sample of adolescents between the ages of 11 and 14years was significantly higher in girls (29%) than boys (19%). Stanford and colleagues [38] reported that, over the course of adolescence, girls were more likely to experience higher frequencies of back pain than boys. The remaining back pain studies reported no sex differences in back pain prevalence rates.

3.4.4. Psychosocial factors

One of the studies examining back pain included measurement of psychosocial/emotional functioning. Stanford et al. [38] reported that, when predictors were entered into a regression model separately, sex and parent- and youth-reported anxiety/depression were predictive of start and end points of back pain trajectories; however, anxiety/depression was no longer significant when variables were entered simultaneously. Kristjansdottir [17] measured sociodemographic variables in a study of back pain in Icelandic children and found that social class was not related to self reports of back pain, although there was an interaction between age and living condition with respect to back pain.

3.5. Musculoskeletal/limb pain studies

3.5.1. Prevalence

As shown in Table 5, 4 studies (total n=3842) specifically examined the prevalence of musculoskeletal and/or limb pain in children and adolescents. One study examined musculoskeletal/limb pain in the context of a larger study investigating multiple pains (n=1908), making the total number of participants 5750. Again, prevalence rates varied depending on time period of reporting, and calculation of median prevalence rates was made difficult due to the wide variety of musculoskeletal pain symptoms reported in each study. Additionally, the relation between musculoskeletal/limb pain and athletic participation further confounds findings in this area, as many participants reported that their pain was the result of a sports injury. In a Norwegian study that asked participants between the ages of 10 and 15years whether they “usually” feel pain [37], results indicated that 25% of respondents experienced musculoskeletal pain several days a week. Brun Sunblad and colleagues [10] found that weekly musculoskeletal pain was experienced by only 9% of their sample of children aged 9, 12, or 15years. In another study examining prevalence of knee pain in children aged either 9–10years or 14–15years [39], the overall prevalence rate was 12%; however, 56% of reported pain was linked to athletic participation. Mikkelsson and colleagues [25] examined prevalence rates of nonspecific musculoskeletal pain in children between the ages of 9 and 11years over a 3-month period and found that 32% of these children reported pain once a week, whereas 39% experienced pain once a month. In their study of Norwegian children, Brun Sunblad and colleagues [10] found that, over a 3-month period, 29% of respondents reported experiencing musculoskeletal pain “now and again.” Finally, in a Brazilian study examining prevalence rates of musculoskeletal pain in 10–18-year-olds [45], 40% of respondents indicated that they had experienced musculoskeletal pain at some point during the last 6months.

Table 5:
Studies of musculoskeletal and limb pain prevalence.

3.5.2. Age

Based on results of these studies, the prevalence rate of musculoskeletal pain in children and adolescents increased with age. Vahasarja [39] noted that adolescents had significantly more knee pain than children (19% and 4%, respectively); these results were echoed by Zapata et al. [45], who also reported increasing rates of musculoskeletal pain with increasing age in their sample. Rhee et al. [35] report increasing musculoskeletal pain with increasing age; however, these authors note that this trend is quadratic, with a peak in musculoskeletal pain at age 16. Smedbraten et al. [37] report not only that pain prevalence rates increase with age, but that the consequences of musculoskeletal pain (ie, difficulty concentrating, reliance on medications) also increase with increasing age. Finally, Brun Sunblad and colleagues [10] found that the increase in musculoskeletal pain symptoms with increasing age was evident only in girls.

3.5.3. Sex

With respect to sex differences in musculoskeletal pain, the majority of the studies reported that musculoskeletal pain is more common in girls than in boys. For example, Smedbraten et al. [37], in their comprehensive study of multiple musculoskeletal pain symptoms, found that girls generally reported pain in more body parts than boys and also reported more frequent pain as compared to boys. Girls reported significantly more pain in the head, neck, shoulders, back, and abdomen as compared to boys, whereas boys reported significantly more knee pain than girls. Zapata et al. [45] echoed these results, and reported more overall musculoskeletal pain in girls as compared to boys in their sample (34% vs. 11%). Specifically, girls reported significantly more upper limb pain than boys (11% vs. 7%), more soft-tissue stress injury than boys (27% vs. 9%), more benign joint hypermobility syndrome than boys (14% vs. 5%), and a greater incidence of myofascial syndrome compared to boys (7% vs. 2%). Mikkelsson et al. [25] also reported that girls reported more pain in the chest and upper back compared to boys. In contrast to these findings, Rhee et al. [35] found that boys experienced increased rates of musculoskeletal pain as compared to girls in their study of American adolescents.

3.5.4. Psychosocial factors

Only one of the studies examining musculoskeletal pain examined psychological variables related to pain; therefore, it is difficult to draw firm conclusions as to the impact of these variables on this type of pain. With respect to psychological functioning in children and adolescents experiencing musculoskeletal pain, Brun Sunblad and colleagues [10] found that feeling sad was associated with pain in girls, but no psychological variables were associated with musculoskeletal pain in boys.

3.6. Combined pain studies

3.6.1. Prevalence

As shown in Table 6, 3 studies (total n=30,843) specifically examined the prevalence of multiple pains in children and adolescents; that is, these studies combined pain types in order to obtain prevalence rates for these combinations and did not examine single pains [3,19,35]. Two of the studies (total n=3043) noted previously also assessed multiple pains in addition to singular pains [28,38], resulting in a total of 33,886 participants. Given the wide range of pains examined in these studies, prevalence rates were, again, somewhat variable and were dependent on the time period over which children were asked to respond. In their large-scale study of multiple recurrent pains in an American sample of children and adolescents in grades 7 through 12, Rhee and colleagues [35] found that the 12-month prevalence rate of combined headache/abdominal pain/musculoskeletal pain was 4%. Prevalence of monthly pain combinations in an Icelandic sample was reported by Kristjansdottir [19], who found that 16% of respondents reported 2 pains on a monthly basis. Weekly prevalence rates for multiple pains were reported in 3 studies [3,19,38], with rates ranging from 4% to 11% (median=6%). One study reported prevalence rates for parent- and child-reported combined headache and abdominal pain in a sample of German children [28]; the parent-reported prevalence rate for these combined pains was 46%, whereas children reported a prevalence rate of 49%.

Table 6:
Studies of combined pain prevalence.

3.6.2. Age

Age differences in multiple pain prevalence were not reported by Bakoula et al. [3]; however, Kristjansdottir [19] reported that 21% of 11–12-year-olds and 14% of 15–16-year-olds have some combination of 1, 2, or 3 pains. These prevalence rates represent overall findings, however, and rates were varied across age groups depending on the specific pain combination in question (ie, number of symptoms and site of pain). Rhee and colleagues [35] found no significant linear relation between age and total number of symptoms.

3.6.3. Sex

With respect to sex differences in pain combinations, findings were generally consistent with studies of single pains, in that girls tended to report more symptoms than boys. Kristjansdottir [19] found that girls reported higher overall pain scores as compared to boys. In terms of specific pain combinations, 17% of boys and 16% of the girls had some combination of 1, 2, or 3 weekly pains. Again, rates varied depending on specific combinations. Findings from the other study that sought to specifically examine pain combinations in children and adolescents [3] indicated that, in a sample of Greek children, girls reported more recurrent pain than boys (9% vs. 6%). These findings were similar to Rhee et al. [35], who also found that girls were more likely to report multiple pain symptoms than boys.

3.6.4. Psychosocial factors

Although several of the studies of multiple pains examined psychosocial variables in relation to pain, all except one examined the associations between psychosocial variables and single pains. One study [3] examined the relation between psychosocial factors and multiple pain and found that complaints of recurrent chronic pain was significantly correlated with several variables: chronic health problems, frequent change of residence, poor performance at school, frequent television watching, and fewer interactions with other children. These authors did not find significant associations between pain and family structure or socioeconomic status.

3.7. Other/general pain studies

3.7.1. Prevalence

As shown in Table 7, 5 studies (n=7051) specifically examined general (ie, nonspecific) pain in children and adolescents. As in all other pain categories described thus far, reported prevalence rates for general pain varied and were dependent on the time period of reporting. Three studies [5,16,41] included point prevalence rates of general pain (ie, pain at time of interview/survey). Point prevalence rates ranged from 5% (chronic pain at time of survey) to 27% (median=18%). van Dijk and colleagues [41] also examined lifetime prevalence rates for various types of pain (ie, recurring pain, 2 or more recurring pains, long-lasting pain, and chronic pain condition); these rates ranged from 6% (chronic pain condition) to 57% (recurring pain) (median=38%). Pain over the past 3months was reported in 2 studies [16,31] and ranged from 54% to 88% (median=71%). Additionally, Perquin and colleagues [31] reported that chronic pain accounted for 25% of pain in their sample, whereas 24% of pain was described as nonchronic. Finally, pain over the last month was reported in one study [40] and was reported as 60%.

Table 7:
Studies of other/general pain prevalence.

3.7.2. Age

Age differences in pain were reported in 3 of the 5 studies. In their study of Spanish school children, Barajas and colleagues [5] reported that pain steadily decreased with increasing age in their sample. Prevalence in younger children was reported to be 33%, whereas prevalence in older children was reported to be 24%; this difference, however, was not statistically significant. Additionally, an age×sex interaction was noted, with higher prevalence rates for girls aged 6–7years of age (36%) and boys 8–10years of age (30%). Perquin and colleagues [31] reported increasing rates of chronic pain with age in both sexes, noting that 1/3 of adolescents between 12 and 15years of age reported chronic pain. Finally, Huguet and Miro [16] also found an increase in chronic pain with increasing age; especially in girls compared to boys.

3.7.3. Sex

As with other studies of specific types of pain, general pain was reported to be more prevalent in girls than in boys. Perquin and colleagues [31] found that girls reported chronic pain more frequently than boys, with a considerable increase in chronic pain in girls between the ages of 12 and 14years. These results were echoed by Huguet and Miro [16], who found that, in their sample of Spanish children and adolescents, the prevalence of chronic pain was higher in girls as opposed to boys, regardless of age. Some interesting trends were also noted with respect to sex differences in pain: chronic lower limb pain was more common in boys than girls (57% vs. 19.7%), whereas chronic pain at multiple sites was more common in girls than in boys (51% vs. 22%).

3.7.4. Psychosocial factors

Two of the 5 studies of general pain [16,40] examined psychosocial variables and the impact of pain in children and adolescents. Huguet and Miro [16] reported that, for children and adolescents with chronic pain, quality of life (as assessed by the Pediatric Quality of Life Inventory (PedsQL; PedsMetrics 1998–2011)) was significantly lower than for children who did not report chronic pain. Fifty-three percent of children with chronic pain reported visiting a doctor as a result of their pain, as compared to 30% of children who did not report chronic pain [16]. Children with chronic pain were also more likely to use medication for their pain compared to children without chronic pain (55% vs. 39%). Finally, Huguet and Miro [16] noted that the prevalence rate of chronic pain in their Spanish sample was not associated with socioeconomic level or with parental marital status. van Dijk and colleagues [40] also examined psychological functioning to some extent in their study of pain in Canadian children and adolescents. Findings from this study indicated that pain was associated with poor self-rated health, psychological symptoms (ie, feeling low, irritable or bad temper, feeling nervous, and difficulties falling asleep).

3.8. Study quality

With respect to study quality, percent agreement between coders was 96%. As shown in Table 8, the quality of the studies included in the present systematic review was generally low-moderate. Specifically, the mean number of criteria met by interview studies was 4.5/19 (SD=2.9) and the mean for studies using self-report was 9.9/17 (SD=7.7). The range for median number of quality criteria met by each type of study was not large, ranging from 7 to 9 (Table 2). The most frequently met criteria were the following: random recruitment, clearly stating the recruitment process, high response rate (ie, >80%), and providing information regarding the questions used to assess chronic or recurrent pain. Areas of weakness included using biased (ie, convenience) samples, omitting sample size calculations, omitting analyses of nonresponders, and failing to present demographic information such as age, sex, and SES.

Table 8:
Assessment of study quality.

4. Discussion

The current review systematically examined epidemiological studies of pain in children and adolescents to update the literature and evaluate progress made since Goodman and McGrath published their seminal paper in 1991. Additionally, we developed a set of criteria to assess the quality of studies included in this review. To our knowledge, this is the first review of epidemiological studies of pediatric pain to employ a systematic approach and to assess study quality. An extensive search of the current literature identified 41 studies of pain epidemiology in nonclinical samples of children and adolescents published since 1991, in contrast to the 34 studies (published from 1951 to 1991) identified by Goodman and McGrath [15]. The greater number of published studies of pain epidemiology in children and adolescents since 1991 suggests that this is an area of interest to researchers in the field and that pediatric pain is a health concern worthy of study. Identifying prevalence rates of pediatric pain has important implications for investigating developmental trajectories of pain and may lead to a more thorough understanding of the development of adult chronic pain conditions.

This review found that chronic and recurrent pain is prevalent in children and adolescents, with girls generally experiencing more pain than boys and prevalence rates increasing with age. Psychosocial variables impacting pain prevalence included anxiety, depression, low self-esteem, other chronic health problems, and low SES. Overall, headache was found to be the most commonly studied single pain type in youth, with an estimated median prevalence rate of 23%. Other types of pain (ie, abdominal pain, back pain, musculoskeletal pain, and pain combinations) were less frequently studied than headache, and prevalence rates were variable due to differences in reporting; however, overall results indicated that these pain types are highly prevalent in children and adolescents, with median prevalence rates ranging from 11% to 38%. These rates are of great concern, but what is even more concerning is that research suggests that prevalence rates of childhood pain have increased over the last several decades [4,36].

Findings such as this argue that researchers and clinicians alike should be aware of the problem of pain in children and of the long-term consequences of chronic pain in childhood. Indeed, studies of musculoskeletal pain found prevalence rates similar to adult rates, suggesting that at least some types of pain may become chronic in childhood and may be predictive of long-term pain complaints and pain-related disability. The majority of studies included in the current review were conducted using somewhat restricted age ranges and were not longitudinal in nature. As a result of these shortcomings, there are several gaps in our understanding of age differences in pain prevalence rates and the developmental trajectories of pain across the lifespan. In order to fully understand pain prevalence in childhood, as well as the risk factors that predict pain in adulthood, it is essential that pain research adopt a more developmental focus; that is, using wider age ranges and/or conducting more longitudinal studies of pain prevalence will increase understanding of age differences in pain prevalence. This knowledge will, in turn, lead to a greater understanding of pain prevalence throughout the lifespan.

Findings from this review also indicated that most types of pain are more prevalent in girls than in boys, but the factors that influence this sex difference in pain prevalence are not entirely clear, suggesting that more research investigating the correlates of pain in both girls and boys is necessary. Similarly, pain prevalence rates tend to increase with increasing age (with the exception of abdominal pain, which tends to be more prevalent in younger children). Targeting younger children for interventions may be a key factor in determining whether pain difficulties become chronic or not; therefore more longitudinal research is needed to determine whether early intervention can serve as a protective factor in young children.

Although some of the studies included in the current review included assessment of psychosocial and sociodemographic factors (eg, SES, mental health factors, parental factors) influencing the development and maintenance of recurrent and chronic pain in children and adolescents, there is, overall, a dearth of knowledge in this area. Of the studies that examined these factors, it was found that SES, parental education, and mental health status, type of residence, and time spent watching television contributed to complaints of pain in children and adolescents. However, the contribution of these variables to the development of long-term pain complaints has not been fully examined, despite being acknowledged as important by several authors. By focusing more epidemiological research on the psychosocial and environmental factors that influence the development of pain complaints, it may be possible to identify those children most in need of medical or psychological interventions, with the goal of preventing the onset of chronic pain.

Perhaps the most interesting finding from the current study is that, despite suggestions by Goodman and McGrath [15] regarding operational definitions of pain and accurate reporting of pain intensity, frequency, and duration, there was little consistency across studies with respect to these factors, as indicated by the quality rating component of this review. Specifically, most studies did not provide operational definitions of the type of pain being investigated, making it difficult to accurately ascertain whether acute, recurrent, or chronic pain was being studied. For this reason, the current review included all studies that examined recurrent or chronic pain, based on information contained in the abstract or methodology sections. Clearly, providing accurate operational definitions of pain will allow for more specific study of the nature of various types of pain in future studies. Similarly, there was great variability in prevalence rates across studies, mostly due to time periods over which pain was reported. Comparisons among studies will facilitate accurate reporting of pain prevalence rates in children and adolescents across sex, age, and culture.

The lack of consistency between studies in terms of pain definitions and reporting highlights the need to apply a set of standardized quality criteria when examining the epidemiology of pain in children and adolescents. As part of the current study, a set of criteria were developed with which to assess study quality; findings indicated that few studies met the majority of these criteria. Design flaws such as omission of sample size calculations, using biased samples, failing to adequately operationalize the outcome variable (ie, pain) and omitting demographic information, as well as extremely varied reporting periods for pain make it difficult to accurately assess prevalence rates in children and adolescents. As suggested by Goodman and McGrath [15] and reiterated here, future epidemiological studies of chronic pain in children and adolescents should adhere to a set of quality criteria to allow for more direct comparison among studies, as well as advancing understanding of prevalence rates of pediatric pain across regions and cultures. More attention to study quality would facilitate combination of studies for the purposes of obtaining overall prevalence rates and may also lead to improved developmental models of chronic and recurrent pain.

Although this review provides updated prevalence rates and identified key psychosocial variables associated with pediatric chronic pain, as well as provided suggestions for improving the quality of epidemiological studies of childhood pain, some limitations should be acknowledged. First, the estimates of overall pain prevalence provided in this review are based on median prevalence rates. These rates were designed to provide a general estimate of prevalence rates; however, they should be interpreted with caution and the reader is referred back to the original study to obtain more precise prevalence rates. Second, median prevalence rates were used as part of the current review, as the median is not affected by outliers in the data; however, median prevalence rate may still be affected by sample size, in that larger samples would likely result in higher rates of reported pain, thereby creating bias. Given that there is no clear standard in the literature for reporting prevalence rates in epidemiological studies, the current study prioritized sensitivity to outliers over sample size.

In conclusion, results of this review indicate that persistent and recurrent chronic pain is overwhelmingly prevalent in children and adolescents and should be recognized as a major health concern in this population. The current review identified several demographic and psychosocial factors associated with high prevalent rates of specific pain types; it is suggested that researchers carefully consider these factors when designing and conducting epidemiological studies of pain in youth, as they often play a significant role in the onset and maintenance of chronic pain conditions. By shifting focus to factors associated with chronic and recurrent pain, it may be possible to identify the most salient risk factors for the development of chronic pain in children and adolescents, thereby leading to early and intensive intervention for the most at-risk group. Continued focus in this area could have a dramatic impact on pain prevalence rates and may result in improved quality of life, decreased pain-related disability, and improved outcomes for children and adolescents challenged by chronic pain.

Conflict of interest statement

No conflicts of interest are declared.


During the preparation of this manuscript, Sara King was a Canadian Institutes of Health Research Training fellow in the Canadian Child Health Clinician Scientist Program, in partnership with SickKids Foundation, Child & Family Research Institute BC, Women & Children’s Health Research Institute (Alberta) and Manitoba Institute of Child Health. Christine Chambers, Patrick McGrath, and Louise Parker hold Canada Research Chairs. Sara King is now in the Faculty of Education at Mount Saint Vincent University, Halifax, NS, Canada.


[1]. Abu-Arefeh I, Russell G. Prevalence of headache and migraine in schoolchildren. BMJ. 1994;309:765-769.
[2]. Apley J, Naish N. Children with recurrent abdominal pains: a field study of 1000 children. Arch Dis Child. 1958;33:165-170.
[3]. Bakoula C, Kapi A, Veltsista A, Kavadias G, Kolaitis G. Prevalence of recurrent complaints of pain among Greek schoolchildren and associated factors: a population-based study. Acta Paediatr. 2006;95:947-951.
[4]. Bandell-Hoekstra IE, Abu-Saad HH, Passchier J, Frederiks CM, Feron FJ, Knipschild P. Prevalence and characteristics of headache in Dutch schoolchildren. Eur J Pain. 2001;5:145-153.
[5]. Barajas C, Bosch F, Banos J. A pilot survey of pain prevalence in schoolchildren. Pain Clin. 2001;13:95-102.
[6]. Barea LM, Tannhauser M, Rotta NT. An epidemiologic study of headache among children and adolescents of southern Brazil. Cephalalgia. 1996;16:545-549. discussion 523.
[7]. Boey CC, Yap SB. An epidemiological survey of recurrent abdominal pain in a rural Malay school. J Paediatr Child Health. 1999;35:303-305.
[8]. Boey CC, Yap S, Goh KL. The prevalence of recurrent abdominal pain in 11- to 16-year-old Malaysian schoolchildren. J Paediatr Child Health. 2000;36:114-116.
[9]. Brattberg G. The incidence of back pain and headache among Swedish school children. Qual Life Res. 1994;3(Suppl. 1):S27-S31.
[10]. Brun Sunblad GM, Saartok T, Engstrom LT. Prevalence and co-occurrence of self-rated pain and perceived health in school-children: age and gender differences. Eur J Pain. 2007;11:171-180.
[11]. Bugdayci R, Ozge A, Sasmaz T, Kurt AO, Kaleagasi H, Karakelle A, Tezcan H, Siva A. Prevalence and factors affecting headache in Turkish schoolchildren. Pediatr Int. 2005;47:316-322.
[12]. Carlsson J. Prevalence of headache in schoolchildren: relation to family and school factors. Acta Paediatr. 1996;85:692-696.
[13]. Elwood JM., 2007. Critical appraisal of epidemiological studies and clinical trials, Oxford University Press, New York.
[14]. Fendrich K, Vennemann M, Pfaffenrath V, Evers S, May A, Berger K, Hoffmann W. Headache prevalence among adolescents – the German DMKG headache study. Cephalalgia. 2007;27:347-354.
[15]. Goodman JE, McGrath PJ. The epidemiology of pain in children and adolescents: a review. Pain. 1991;46:247-264.
[16]. Huguet A, Miro J. The severity of chronic pediatric pain: an epidemiological study. J Pain. 2008;9:226-236.
[17]. Kristjansdottir G. Prevalence of self-reported back pain in school children: a study of sociodemographic differences. Eur J Pediatr. 1996;155:984-986.
[18]. Kristjansdottir G. Sociodemographic differences in the prevalence of self-reported stomach pain in school children. Eur J Pediatr. 1996;155:981-983.
[19]. Kristjansdottir G. Prevalence of pain combinations and overall pain: a study of headache, stomach pain and back pain among school-children. Scand J Soc Med. 1997;25:58-63.
[20]. Kristjansdottir G, Wahlberg V. Sociodemographic differences in the prevalence of self-reported headache in Icelandic school-children. Headache. 1993;33:376-380.
[21]. Loney PL, Chambers LW, Bennett KJ, Roberts JG, Stratford PW. Critical appraisal of the health research literature: Prevalence or incidence of a health problem. Chronic Dis Canada. 1998;19:170-176.
[22]. Loney PL, Stratford PW. The prevalence of low back pain in adults: a methodological review of the literature. Phys Ther. 1999;79:384-396.
[23]. McGrath PJ, Dunn-Gier J, Cunningham SJ, Brunette R, D’Astous J, Humphreys P, Latter J, Keene D. Psychological guidelines for helping children cope with chronic benign intractable pain. Clin J Pain. 1986;1:229-233.
[24]. Merskey H, Bogduk N., 1994. International Association for the Study of Pain. Classification of chronic pain: descriptions of chronic pain syndromes and definitions of pain terms, IASP Press, Seattle.
[25]. Mikkelsson M, Salminen JJ, Kautiainen H. Non-specific musculoskeletal pain in preadolescents: prevalence and 1-year persistence. Pain. 1997;73:29-35.
[26]. Oddson BE, Clancy CA, McGrath PJ. The role of pain in reduced quality of life and depressive symptomology in children with spina bifida. Clin J Pain. 2006;22:784-789.
[27]. Oh MC, Aw MM, Chan YH, Tan LZ, Quak SH. Epidemiology of recurrent abdominal pain among Singaporean adolescents. Ann Acad Med Singapore. 2004;33:S10-S11.
[28]. Ostkirchen GG, Andler F, Hammer F, Pohler KD, Snyder-Schendel E, Werner NK, Markett S, Horacek U, Jockel KH, Diener HC. Prevalences of primary headache symptoms at school-entry: a population-based epidemiological survey of preschool children in Germany. J Headache Pain. 2006;7:331-340.
[29]. Palermo TM. Impact of recurrent and chronic pain on child and family daily functioning: a critical review of the literature. J Dev Behav Pediatr. 2000;21:58-69.
[30]. Palermo TM, Chambers CT. Parent and family factors in pediatric chronic pain and disability: an integrative approach. Pain. 2005;119:1-4.
[31]. Perquin CW, Hazebroek-Kampschreur AA, Hunfeld JA, Bohnen AM, van Suijlekom-Smit LW, Passchier J, van der Wouden JC. Pain in children and adolescents: a common experience. Pain. 2000;87:51-58.
[32]. Petersen S, Bergstrom E, Brulin C. High prevalence of tiredness and pain in young schoolchildren. Scand J Public Health. 2003;31:367-374.
[33]. Ramchandani PG, Hotopf M, Sandhu B, Stein A. ALSPAC Study Team. The epidemiology of recurrent abdominal pain from 2 to 6 years of age: results of a large, population-based study. Pediatrics. 2005;116:46-50.
[34]. Rhee H. Prevalence and predictors of headaches in US adolescents. Headache. 2000;40:528-538.
[35]. Rhee H, Miles MS, Halpern CT, Holditch-Davis D. Prevalence of recurrent physical symptoms in US adolescents. Pediatr Nurs. 2005;31:314-350.
[36]. Sillanpaa M, Anttila P. Increasing prevalence of headache in 7-year-old schoolchildren. Headache. 1996;36:466-470.
[37]. Smedbraten BK, Natvig B, Rutle O, Bruusgaard D. Self-reported bodily pain in schoolchildren. Scand J Rheumatol. 1998;27:273-276.
[38]. Stanford EA, Chambers CT, Biesanz JC, Chen E. The frequency, trajectories and predictors of adolescent recurrent pain: a population-based approach. Pain. 2008;138:11-21.
[39]. Vahasarja V. Prevalence of chronic knee pain in children and adolescents in northern Finland. Acta Paediatr. 1995;84:803-805.
[40]. van Dijk A, McGrath PA, Pickett W, Van Den Kerkhof EG. Pain and self-reported health in Canadian children. Pain Res Manag. 2008;13:407-411.
[41]. van Dijk A, McGrath PA, Pickett W, VanDenKerkhof EG. Pain prevalence in nine- to 13-year-old schoolchildren. Pain Res Manag. 2006;11:234-240.
[42]. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening of the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. Ann Intern Med. 2007;147:573-577.
[43]. Watson KD, Papageorgiou AC, Jones GT, Taylor S, Symmons DP, Silman AJ, Macfarlane GJ. Low back pain in schoolchildren: occurrence and characteristics. Pain. 2002;97:87-92.
[44]. Zaccai JH. How to assess epidemiological studies. Postgrad Med J. 2004;80:140-147.
[45]. Zapata AL, Moraes AJ, Leone C, Doria-Filho U, Silva CA. Pain and musculoskeletal pain syndromes in adolescents. J Adolesc Health. 2006;38:769-771.

Pain; Children; Epidemiology; Systematic review; Quality

© 2011 Lippincott Williams & Wilkins, Inc.