ORIGINAL ARTICLE
Musculoskeletal Pain in Obese Compared With Healthy-Weight Children Margarita D. Tsiros, PhD,* Jonathan D. Buckley, PhD,* Peter R. C. Howe, PhD,* Jeff Walkley, PhD,w Andrew P. Hills, PhD,z and Alison M. Coates, PhD*
Objectives: To investigate whether obesity is associated with musculoskeletal pain in children. Materials and Methods: Obese (n = 107) and healthy-weight (n = 132) 10- to 13-year-old children (132 males, 107 females) participated in an observational case-control study. Children self-reported pain location (excluding abdominal pain), pain intensity (current and prior week), and pain prevalence (overall and lower limb) using the Pediatric Pain Questionnaire. Body composition was assessed (dual-energy x-ray absorptiometry) and children wore an accelerometer for 8 days. Results: After adjustment for accelerometry (weekly average counts per hour) and socioeconomic status, obese children had more intense pain (worst pain, P = 0.006), pain in more locations (Pr0.005), and a higher prevalence of lower limb pain (60% vs. 52% respectively, P = 0.012) than healthy-weight children. Significant relationships were observed between body mass index and total pain locations (Pr0.004, unadjusted and adjusted) and worst pain intensity (Pr0.009, adjusted for socioeconomic status/accelerometry). There were no significant relationships between percent body fat and pain variables (unadjusted/adjusted analyses, P = 0.262 to 1.0). Discussion: Obesity in children was associated with increased overall and lower limb musculoskeletal pain, for which body mass index was a stronger predictor than adiposity. Clinicians treating obese children should screen for pain and prescribe exercise programs that take their symptoms into account. Key Words: body mass index, child, pain, adiposity, obese
(Clin J Pain 2014;30:583–588)
P
ain in childhood is a frequent occurrence with around 15% to 25% of children and adolescents experiencing recurrent/chronic symptoms: specifically abdominal-related,
Received for publication January 8, 2013; revised October 7, 2013; accepted August 2, 2013. From the *Nutritional Physiology Research Centre, School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, SA; wSchool of Health Sciences, RMIT University, Melbourne, Vic; and zMater Mothers’ Hospital, Mater Research and Centre for Musculoskeletal Research, Griffith Health Institute, Griffith University, Nathan, QLD, Australia. Present address: Peter R. C. Howe, PhD, Clinical Nutrition Research Centre, University of Newcastle, Callaghan, NSW, 2308, Australia. The authors declare no conflict of interest. This study was funded by a seeding grant from the Physiotherapy Research Foundation and internal funds (ATN Centre for Metabolic Fitness). M.D.T. was supported by an Australian Post Graduate Award and the Nutritional Physiology Research Centre, Adelaide, SA, Australia. Supported by the Australian Technology Network Centre for Metabolic Fitness and the Physiotherapy Research Foundation, Camberwell, Vic, Australia. Reprints: Margarita D. Tsiros, PhD, Nutritional Physiology Research Centre, School of Health Sciences, University of South Australia, P.O. Box 2471, Adelaide, SA 5001, Australia (e-mail: margarita. [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
musculoskeletal-related, or headache-related pain.1 Obesity has been identified as a possible contributor to musculoskeletal pain in children because of increased loading and/or biomechanical alterations imposed by additional fat, possibly compounded by periods of rapid growth that are characteristic of childhood.2–4 Children with obesity experience lower health-related quality of life (HRQoL) and greater disability than their healthy-weight peers5,6 and obesity-related pain may be a contributing factor.7 A detailed review of pain in obese children and adolescents, concluded that most research has focused on low back pain (LBP),3 with prospective studies finding no link between body mass index (BMI) and LBP incidence.8–11 However, it was unclear in these studies whether obese children were adequately represented.8–11 In addition, many such studies assessed lifetime pain prevalence which has questionable recall reliability in children, especially in relation to chronic pain.12 Although few studies have specifically investigated overall musculoskeletal pain in obese children, findings tend to suggest a higher pain prevalence in obese compared with nonobese children.13–17 Most notable was a recent population-based survey of adolescents which found that those with obesity were more likely to report musculoskeletal pain (odds ratio of 1.33) with greater symptom severity than nonobese children17; although generalizability to all children is limited given the sample included only 17-year-olds. Other limitations of remaining research include a lack of validated pain tools,13–15 relying on physical examinations or interviews,13–15,18 not including a control group,16,18 or recruiting participants from an orthopedic clinic leading to a possible recruitment bias.16 Lower limb pain prevalence has also been positively associated with obesity in children13,18,19; however, studies have been limited to treatment-seeking samples and may not be generalizable to obese children in the community. Moreover, most research examining overall musculoskeletal pain has been limited in that it only assessed whether pain was present13–16,19–21 and on few occasions has the frequency of pain been captured.13,19 Three studies have assessed pain intensity which is particularly relevant as it has been linked with functional limitations in general pediatric samples.1,22,23 However, these studies have either focused on a limited age range that may not be generalizable to children of other ages,17 included nonmusculoskeletal pain (eg, abdominal pain)21 or have focused on severely obese treatment-seeking youths18 who may represent an extreme. Early adolescence has the potential to be a particularly high-risk period for the development of obesity-related pain given the many tumultuous changes occurring during this time (eg, hormonal changes, rapid growth, changes in physical activity behaviors, emotional/psychosocial changes). From an intervention perspective, early adolescence is also a better time to intervene to prevent obesity-related pain given www.clinicalpain.com |
583
Clin J Pain
Tsiros et al
that parents still tend to have some influence over their children’s lives. However, scant research has specifically focused on early adolescence. Taylor et al14 found that obese children (mean age 12.8 y) were 4.04 times more likely than nonobese children to have musculoskeletal pain, whereas Bell et al15 found that the odds of experiencing musculoskeletal pain increased >2-fold per unit increase in BMI z score (6 to 13 y olds). In contrast, studies limited to examining low back pain concluded that such pain is unrelated to BMI in 10- to 14-year-olds.7–9,24–26 Physical activity is another important consideration as it could confound any relationship between weight status and pain. Yet few studies (examining LBP only) have addressed the potential influence of physical activity.10,24,25,27,28 Although some have reported LBP to be associated with limited activity or certain types of sport/leisure,10,24,27,28 they have not considered overall physical activity levels nor used objective measures of physical activity, instead relying on self-report measures which are subject to recall bias.29 Furthermore, studies have not sought to examine whether physical activity levels could confound relationships between weight status and pain. There are many possible ways that physical activity could confound the obesity-pain relationship. For example, although speculative, physical activity could increase children’s tolerance to uncomfortable sensations, thereby causing them to have higher pain thresholds, reporting less intense pain. Alternatively, particular types of physical activity may stress or strain certain parts of the body resulting in pain. The majority of studies (21 of 22 cited in our recent review3) have not used internationally accepted criteria for defining weight status using BMI such as International Obesity Task Force criteria,30 thereby reducing the generalizability of their findings. Moreover, BMI may not provide an accurate indication of adiposity,31 which can be quantified by dual-energy x-ray absorptiometry (DXA). In essence, prior research has many methodological limitations and few studies have examined the impact of weight status on musculoskeletal pain with the exception of LBP. Therefore, the present study aimed to examine if children with obesity have increased musculoskeletal pain compared with their healthy-weight peers and to explore the relationship between adiposity and musculoskeletal pain. This is the first study in children to use a validated pain assessment tool, control for objectively assessed physical activity levels, and characterize body composition using internationally accepted criteria and DXA.
MATERIALS AND METHODS Obese and healthy-weight 10- to 13-year-olds, as defined by International Obesity Task Force criteria,30 were recruited from 3 Australian states as part of a larger observational case-control study.5 Recruitment methods included media advertisements/releases, flyers, school newsletters, and a pediatric clinic. Exclusions for the larger study comprised children who were underweight, overweight,30,32 intellectually/neurologically impaired, engaged in a weight-loss program in the preceding 3 months, had a past medical history of ear infections/balance issues, acute injury (prior 6 mo) necessitating medical management (eg, sprain or fracture), or a pathologic cause for their obese status (such as hypothyroidism). Ethical approval was provided by the University of South Australia, RMIT University, Queensland University of Technology and Flinders Medical Centre.
584 | www.clinicalpain.com
Volume 30, Number 7, July 2014
Pain Children completed age-appropriate versions of the Pediatric Pain Questionnaire (PPQ)33 under interviewer supervision. The PPQ is a reliable and valid pain measure in children33,34 which captures qualitative descriptors of pain/ discomfort, while also quantifying pain locations (using a body chart) and pain intensity using 100 mm horizontal visual analogue scales to report current pain and recall worst pain over the previous week. The PPQ assesses general pain—as the current study was concerned with musculoskeletal pain, abdominal/stomach pains were excluded from the analysis, whereas all other pain locations were included (ie, head/neck, trunk, upper and lower limb pains). Variables of interest included pain intensity (current and worst over past week), total number of current pain locations (ie, pain anywhere on the body with the exception of abdominal/stomach pains), and total number of current lower limb pain locations (ie, pain reported in the hips/ thighs/knee/shin/calf/ankle/foot/toe regions).
Body Composition Anthropometric and body composition measures were recorded as described in detail previously.5 Briefly, participants wearing only a hospital gown and underwear had body mass and height recorded to calculate BMI and BMI z score.4 Percent body fat was estimated from a whole-body DXA scan.
Physical Activity Participants wore a uniaxial accelerometer (Actigraph; MTI Health Services, Fort Walton Beach, FL, version 2.2 model 7164) on their right hip full-time for 8 days during the school-term, removing it only for water-based activities. Actigraphs sampled at 1-minute epochs and minimum data eligibility criteria included at least 4 weekdays and both weekend days with at least 10 “eligible” hours (defined as >3000 counts/h) of wear-time each day. Average counts per hour per day were then calculated to allow for a differing number of hours awake per day between participants. Weekday and weekend-day average counts per hour were then calculated and an overall average taken to provide an indication of activity over a given week (called weekly average counts per hour). Note that weekdays and weekend days were weighted equally as children spend approximately half of the total days of a year in school over a year (4 terms10 wk 5 d = 200 d, minus about 10 d of public holidays, and teacher-free days, and about 10 d of sickness = 180 d/y). For descriptive purposes, data were also reduced to average minutes of moderate to vigorous physical activity (MVPA) per day. Actigraphs have been reported to have strong interinstrument reliability and have been widely validated in children and adolescents.35
Background Information Information regarding average annual household income, maternal education, and child medical conditions/ medications were collected through a parent-completed questionnaire. The data on household income and maternal education were used as measures of socioeconomic status (SES). Children self-assessed their pubertal development using the Tanner scales.36
Statistical Analysis SPSS version 17 for WINDOWS (International Business Machines Corporation, New York) and STATA r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
Musculoskeletal Pain in Obese Children
TABLE 1. Pain Data
Obese N = 107
Healthy-Weight N = 132
Unadjusted Sig (P)
Adjusted Sig (Inc, Mat Ed) (P)
Adjusted Sig (Inc, Mat Ed, Accel) (P)
2.0 (8.0, 0.0-8.0) 26.0 (48.0, 5.0-53.0)
2.0 (5.9, 2.0-6.9) 19.0 (32.3, 4.5-36.8)
0.584 0.039
0.932 0.006*
0.790 0.006*
2.0 (2.0, 1.0-3.0) 1.0 (2.0, 0.0-2.0)
1.0 (1.5, 0.5-2.0) 1.0 (1.0, 0.0-1.0)
0.005* 0.037
0.006* 0.015
0.005* 0.010*
Outcomes Current pain (mm) Worst pain (last week) (mm) Total pain locations Lower limb pain locations
Data were not normally distributed and are presented as median (interquartile range, 25th to 75th percentile). *Significant difference between groups using an exact Mann-Whitney U test and a sequential Bonferroni significance level (Pr0.013). Monte Carlo P-values are reported. Obese and healthy-weight were defined using International Obesity Task Force criteria.30 Notably, the Mann-Whitney U test is a comparison of mean ranks rather than medians, thus the 2 groups can have the same medians, but still be statistically significantly different as was observed with lower limb pain locations Accel indicates accelerometry weekly average counts per hour; inc, household income; mat ed, maternal education; sig, significance.
(StataCorp LP, College Station, TX) were used for analyses. Abdominal/stomach pains were excluded from the analysis of pain location data for 4 participants in the obese group. Missing data were imputed by expectation maximization algorithms after applying Little test37 showing that data were missing completely at random (P = 0.61). Between-group differences were examined using the Mann-Whitney U test for non-normally distributed data. When data could not be transformed, variables were ranked before using analysis of covariance to adjust for potential confounders that significantly differed between obese and healthy-weight; namely, accelerometry, maternal education, and household income. Relationships between percent body fat or BMI and pain outcomes were examined using quantile regression—as such only unstandardized regression coefficients were available for report. Adjustments were made for the same confounding variables. Quantile regression uses the median rather than the mean, is robust against outliers and is also appropriate to use when the residuals are not normally distributed (as in the current study), thereby violating the assumptions of other regression techniques such as ordinary least squares in relation to hypothesis testing.38 To account for type I error, a post hoc sequential Bonferroni procedure was used.39 Data were not split by gender as sex was not found to be a significant effect modifier of the relationships between pain outcomes and weight status (P-values ranging from 0.062 to 0.712). Given the homogeneous age range of the sample, data were not stratified by age.
RESULTS Full-population demographics have been previously reported.5 In summary, 107 obese and 132 healthy-weight
children participated. Mean group ages were similar (11.8 ± 0.1 and 12.0 ± 0.1 y, P = 0.38) with a reasonably even distribution of sexes between groups (obese: 51 females, 56 males; healthy-weight: 56 females, 76 males, P = 0.42) and no differences in pubertal development (86% being Tanner stage 4 or less, P = 0.13) or height (1.55 ± 0.01 and 1.54 ± 0.01 m, P = 0.33). Notably, more obese children were from families of lower income (43% earning r$60 k vs. only 19% in the healthy-weight group) and lower maternal education (27% with a University degree or higher vs. 49% for healthyweight). Obese children had higher percent fat (45.4 ± 0.5 vs. 21.7 ± 0.6%, P < 0.01), BMI (29.6 ± 0.4 vs. 18.2 ± 0.2 kg/ m2, P < 0.01), and BMI z score (2.16 ± 0.02 vs. 0.03 ± 0.05, P < 0.01) and were also less physically active than their healthy-weight peers (27 037 ± 769 vs. 32 376 ± 1 091 counts/ h, P < 0.01 and 52.1 ± 2.6 vs. 73.8 ± 3.2 min/d of MVPA). The obese group reported pain in more body locations irrespective of accelerometry and/or SES (Pr0.005 for total number of pain locations, Table 1). Obese children reported experiencing significantly more intense pain than healthyweight children in the preceding week (worst pain adjusted for accelerometry and/or SES factors, P = 0.006, Table 1). Furthermore, lower limb pain was more prevalent in obese compared with healthy-weight children (60% vs. 52% respectively, P = 0.012), with obese children also reporting a greater number of lower limb pain locations when controlling for accelerometry and SES factors (P = 0.010, Table 1). There were significant relationships between BMI and the total number of pain locations (adjusted and unadjusted analyses, Pr0.004) and worst pain over the prior week (when adjusted for accelerometry and SES factors, P = 0.009) (Table 2). In contrast, there were no significant relationships between
TABLE 2. Relationships Between BMI and Pain Outcomes in Obese and Healthy-Weight Children
Unadjusted
Adjusted (Inc, Mat Ed)
Adjusted (Inc, Mat Ed, Accel)
Pain Outcomes
Unst. r
P
Unst. r
P
Unst. r
P
Current pain (mm) Worst pain (last week) (mm) Total pain locations Lower limb pain locations
0.00 0.977 *0.05 0.00
1.00 0.04 *0.001 1.00
0.02 1.24 *0.06 0.00
0.667 0.022 0.002 1.00
0.02 *1.38 *0.07 0.00
0.617 *0.009 *0.004 1.00
Quantile regression was used as the residuals were not normally distributed, hence only unstandardized regression coefficients were available for report. *Significant relationship between BMI and the pain outcome using a sequential Bonferroni significance level (Pr0.013). Accel indicates accelerometry weekly average counts per hour; BMI, body mass index; Inc, household income; mat ed, maternal education; Unst. r, unstandardized regression coefficient.
r
2013 Lippincott Williams & Wilkins
www.clinicalpain.com |
585
Clin J Pain
Tsiros et al
percent body fat and pain variables in unadjusted or adjusted analyses (P-values ranging from 0.262 to 1.0) (data not shown).
DISCUSSION Previous research examining pain prevalence and frequency suggests that children with obesity experience more musculoskeletal pain than their healthy-weight counterparts.14–16,19,28 The current study extends on these findings, indicating that obese children experience musculoskeletal pain in more body locations than healthyweight children. More intense overall musculoskeletal pain, combined with a higher prevalence of lower limb pain also featured in obese compared with healthy-weight children after correcting for differences in SES and/or accelerometry, which is consistent with prior research in an older sample.17 Although de Sa Pinto et al13 also reported a higher prevalence of lower limb pain in obese children, they did not use a validated pain assessment tool. As expected, children with higher BMIs tended to report more pain in terms of total body locations and more intense pain over the prior week. Considering the 11.4 kg/ m2 mean BMI differential between the obese and healthyweight group, the observed coefficients would translate into the obese group experiencing 15.7 mm greater intensity of worst pain over the prior week and having 0.8 more pain locations in their body than the healthy-weight group. Given that a change of B10 mm is generally considered “meaningful” on a 100-mm visual analogue scale,40 the relationship between BMI and worst pain intensity over the prior week has particular clinical relevance. In contrast, there was no correlational relationship between percent body fat and any musculoskeletal pain variables. These discrepant findings suggest that the total mass (used to calculate BMI) rather than the adiposity of the child may be of greater importance with regards to musculoskeletal pain. This premise fits with prior research reporting associations between increased weight/BMI and lower limb pain prevalence in a treatment-seeking sample of severely obese children.19 Obesity is characterized by excessive fat, but there are also compensatory increases in fat-free mass and skeletal mass41 which contribute to the total mass a child is required to move during activities of daily living. It is conceivable that children with obesity (as defined by BMI) may exert a greater load on their joints and tissues and that this excessively prolonged loading could lead to biomechanical deviations and musculoskeletal pain.42 Certainly, evidence of increased joint moments/forces has been reported in children with obesity during locomotion when compared with their healthy-weight peers.43 Although not the primary focus of this work, it would appear that SES and overall physical activity may play some role in the musculoskeletal pain experience, as for a number of outcomes, pain was significantly greater (or increased with higher BMI) only after controlling for lower physical activity and SES in the obese group. Hence, this raises the question of whether obese children would experience more pain if they were more active (thereby placing greater load/stress on their joints). Previous research links higher levels of leisure-time physical activity with an increased frequency of musculoskeletal pain (and vice versa) in general pediatric populations aged 10 to 16 years (N = 698), suggesting that sport may induce stress/acute injuries in the growing child44; a finding further supported
586 | www.clinicalpain.com
Volume 30, Number 7, July 2014
by studies of LBP in children.24,27,28 Alternatively, others have found that low leisure activity may be associated with LBP.10,27 It is also possible that the presence of pain may impact a child’s willingness and ability to participate in physical activity, as suggested by Wilson et al21 who found that pain intensity was a stronger predictor of difficulties with active pursuits than BMI. Although the current study extends on prior research by examining overall physical activity levels (measured objectively), treating it as a confounding factor, there is clearly much work that still needs to be done to dissect multivariate relationships between obesity, musculoskeletal pain, and physical activity, which was beyond the scope of this study. Regarding SES, it could be that children from higher SES backgrounds have a lower musculoskeletal pain threshold (or vice versa), thereby confounding obese/healthy-weight differences, although this explanation is speculative. Alternatively, SES factors could also influence pain indirectly by physical activity, given that a higher maternal education/ household income has been shown to promote higher MVPA levels in children (N = 17 766, 11 to 21 y olds).45 This could be due to a greater awareness of the health benefits of physical activity in higher educated families, or participation in user-paid sport/recreation. There is a paucity of literature regarding the impact of obesity on musculoskeletal pain in children.13–16,20 This study makes a novel contribution, offering insight into the intensity of musculoskeletal pain and how widespread it is in the lower limbs and throughout the body. It should be noted that pain from the abdominal region was excluded from the analysis of number of pain locations (impacting 4 obese participants) however, it is possible that if a participant experienced abdominal pain that this may have influenced their rating of pain intensity and hence these data should be interpreted with caution. Particular strengths of this study included the use of internationally accepted definitions of weight status and DXA to characterize adiposity, considerations for objectively measured physical activity, and the use of a validated pain assessment tool (PPQ). Despite this, the PPQ does not offer an exhaustive analysis of the pain experience; for example, the chronicity/recurrence of symptoms, whether the amount of time a child has been obese has any bearing on symptoms, and whether pain thresholds and coping strategies differ between obese and healthy-weight children remains to be explored. Because of the limited age range of children in this study (10 to 13 y old), the generalizability of these findings should be noted. While there were no differences in pubertal development between the healthy-weight and obese groups, these findings may not translate to older or younger children as we cannot rule out that the data could have been influenced by the developmental status of the participants. In addition, it is possible that differences in pain due to weight status may change for a variety of reasons as children develop. The findings of this study have a number of potential clinical implications. Pain is an unpleasant, noxious experience which is always characterized by a certain amount of suffering.22 Not surprisingly, a child’s well-being (otherwise known as HRQoL) is compromised by pain.7 Considering that childhood obesity is already linked with many adverse health outcomes including marked reductions in HRQoL,6 the coexistence of pain and obesity have the potential to further compound impairments in well-being; a premise confirmed by Hainsworth et al’s7 finding that cooccurring r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
chronic pain and obesity increased the risk of impaired physical HRQoL nearly 3-fold compared with either condition in isolation. The presence of pain in children with obesity may also pose a barrier to their participation in physical activity,21 whereas certain types of activity could have the potential to induce/aggravate symptoms.24,27,2,8,44 In conclusion, this work suggests that even in children aged 10 to 13 years, obesity may be associated with increased overall and lower limb musculoskeletal pain, although overall mass appears to be of greater importance than adiposity. Physical activity and SES also appear to play a role, especially with regard to musculoskeletal pain intensity and lower limb pain, but further research is needed to dissect these effects. Clinicians treating children with obesity should screen for musculoskeletal pain and, if needed, refer for management (eg, medical review, physiotherapy), while taking care to design exercise programs that will not aggravate existing symptoms. Low-impact activity such as walking, swimming, or aqua exercise may be better suited to children with obesity experiencing pain, although further research is needed to confirm this. For some children, specialist prescription (eg, by a physiotherapist or exercise physiologist) of appropriate exercises for their musculoskeletal pain condition may also be required.
ACKNOWLEDGMENTS The Quality of Life study described in this paper was carried out using the PedsQL developed by Dr James W. Varni. The authors thank Professor Adrian Esterman, PhD (School of Nursing and Midwifery, Sansom Institute for Health Research, University of South Australia, Adelaide, SA, Australia) (advice on statistical analysis) and to Dr Paul Grimshaw, PhD (School of Mechanical Engineering, University of Adelaide, SA, Australia), Professor Timothy Olds, PhD (Health and Use of Time Group, School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, SA, Australia) and Dr Anthony Shield, PhD (School of Exercise and Nutrition Sciences, Institute of Health Biomedical Innovation, Queensland University of Technology, Brisbane, QLD, Australia) (general advice on study design). Technical assistance was provided with thanks to: Ms Keren Kneebone, B HSci (Human Movement) (School of Exercise and Nutrition Sciences, Deakin University, Burwood, Vic, Australia) Dr Rachel Wood, PhD (Centre of Research Excellence in Translating Nutritional Science to Good Health, University of Adelaide, Adelaide, SA, Australia) Dr Sarah Shultz, PhD (School of Sport and Exercise, Massey University, Palmerston North, New Zealand) Dr Masa Kagawa, PhD (Institute of Nutrition Sciences, Kagawa Nutrition University, Saitama, Japan) Ms Lara Taylor, B App Sci (Psych Hon) (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Mr Richard Mallows, PhD, (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Mr Kaine Grigg, B App Sci (Psych Hon) (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Ms Fiona Spargo, M Sci Hons, (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) and Ms Kate Greenway, PhD (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia). r
2013 Lippincott Williams & Wilkins
Musculoskeletal Pain in Obese Children
REFERENCES 1. Roth-Isigkeit A, Thyen U, Stoven H, et al. Pain among children and adolescents: restrictions in daily living and triggering factors. Pediatrics. 2005;115:e152–e162. 2. Hestbaek L, Leboeuf-Yde C, Kyvik KO. Are lifestyle-factors in adolescence predictors for adult low back pain? A crosssectional and prospective study of young twins. BMC Musculoskelet Disord. 2006;7:27. 3. Tsiros MD, Coates AM, Howe PR, et al. Obesity: the new childhood disability? Obesity Rev. 2011;12:26–36. 4. Wearing SC, Hennig EM, Byrne NM, et al. Musculoskeletal disorders associated with obesity: a biomechanical perspective. Obes Rev. 2006;7:239–250. 5. Tsiros MD, Buckley JD, Howe PRC, et al. Day-to-day physical functioning and disability in obese 10-13 year olds. Pediatr Obes. 2012;8:31–41. 6. Tsiros MD, Olds T, Buckley JD, et al. Health-related quality of life in obese children and adolescents. Int J Obes. 2009;33: 387–400. 7. Hainsworth KR, Davies WH, Khan KA, et al. Co-occurring chronic pain and obesity in children and adolescents: the impact on health-related quality of life. Clin J Pain. 2009;25:715–721. 8. Nissinen M, Heliovaara M, Seitsamo J, et al. Anthropometric measurements and the incidence of low back pain in a cohort of pubertal children. Spine. 1994;19:1367–1370. 9. Poussa MS, Heliovaara MM, Seitsamo JT, et al. Anthropometric measurements and growth as predictors of low-back pain: a cohort study of children followed up from the age of 11 to 22 years. Eur Spine J. 2005;14:595–598. 10. Salminen JJ, Erkintalo M, Laine M, et al. Low back pain in the young. A prospective three-year follow-up study of subjects with and without low back pain. Spine. 1995;20:2101–2107. 11. Lake JK, Power C, Cole TJ. Back pain and obesity in the 1958 British birth cohort. cause or effect? J Clin Epidemiol. 2000;53: 245–250. 12. Erskine A, Morley S, Pearce S. Memory for pain: a review. Pain. 1990;41:255–265. 13. de Sa Pinto AL, de Barros Holanda PM, Radu AS, et al. Musculoskeletal findings in obese children. J Paediatr Child Health. 2006;42:341–344. 14. Taylor ED, Theim KR, Mirch MC, et al. Orthopedic complications of overweight in children and adolescents. Pediatrics. 2006;117:2167–2174. 15. Bell LM, Byrne S, Thompson A, et al. Increasing body mass index z-score is continuously associated with complications of overweight in children, even in the healthy weight range. J Clin Endocrinol Metab. 2007;92:517–522. 16. Podeszwa DA, Stanko KJ, Mooney JF, et al. An analysis of the functional health of obese children and adolescents utilizing the PODC instrument. J Pediat Orthop. 2006;26:140–143. 17. Deere KC, Clinch J, Holliday K, et al. Obesity is a risk factor for musculoskeletal pain in adolescents: findings from a population-based cohort. Pain. 2012;153:1932–1938. 18. Hainsworth KR, Miller LA, Stolzman SC, et al. Pain as a comorbidity of pediatric obesity. ICAN. 2012;4:315–320. 19. Stovitz SD, Pardee PE, Vazquez G, et al. Musculoskeletal pain in obese children and adolescents. Acta Paediatr. 2008;97:489–493. 20. Wake M, Salmon L, Waters E, et al. Parent-reported health status of overweight and obese Australian primary school children: a cross-sectional population survey. Int J Obes. 2002; 26:717–724. 21. Wilson AC, Samuelson B, Palermo TM. Obesity in children and adolescents with chronic pain: associations with pain and activity limitations. Clin J Pain. 2010;26:705–711. 22. Kashikar-Zuck S, Goldschneider KR, Powers SW, et al. Depression and functional disability in chronic pediatric pain. Clin J Pain. 2001;17:341–349. 23. Gauntlett-Gilbert J, Eccleston C. Disability in adolescents with chronic pain: patterns and predictors across different domains of functioning. Pain. 2007;I131:132–141.
www.clinicalpain.com |
587
Tsiros et al
24. Watson KD, Papageorgiou AC, Jones GT, et al. Low back pain in school children: the role of mechanical and psychosocial factors. Arch Dis Child. 2003;88:12–17. 25. Chiang HY, Jacobs K, Orsmond G. Gender-age environmental associates of middle school students’ low back pain. Work. 2006;26:197–206. 26. Szpalski M, Gunzburg R, Balague F, et al. A 2-year prospective longitudinal study on low back pain in primary school children. Eur Spine J. 2002;11:459–464. 27. Masiero S, Carraro E, Celia A, et al. Prevalence of nonspecific low back pain in school children aged between 13 and 15 years. Acta Paediatr. 2007;97:212–216. 28. Kovacs FM, Gestoso M, Gil del Real MT, et al. Risk factors for non-specific low back pain in school children and their parents: a population based study. Pain. 2003;103:259–268. 29. Baranowski T. Validity and reliability of self-report measures of physical activity: an information procesing perspective. Res Q Exerc Sport. 1988;59:314–327. 30. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1240–1243. 31. Wickramasinghe VP, Cleghorn GJ, Edmiston KA, et al. Validity of BMI as a measure of obesity in Australian white caucasian and Australian Sri Lankan children. Ann Hum Biol. 2005;32:60–71. 32. Cole TJ, Flegal KM, Nicholls D, et al. Body Mass Index cut offs to define thinness in children and adolescents: international survey. BMJ. 2007;335:194. 33. Varni J, Thompson K, Hanson V. The Varni/Thompson Pediatric Pain Questionnaire. I. Chronic musculoskeletal pain in juvenile rheumatoid arthritis. Pain. 1987;28:27–38. 34. Gragg RA, Rapoff MA, Danovsky MB, et al. Assessing chronic musculoskeletal pain associated with rheumatic disease: further validation of the Pediatric Pain Questionnaire. J Pediatr Psychol. 1996;21:237–250.
588 | www.clinicalpain.com
Clin J Pain
Volume 30, Number 7, July 2014
35. Trost SG. State of the art reviews: measurement of physical activity in children and adolescents. Am J Lifestyle Med. 2007;1:299–314. 36. Tanner JM. Growth at Adolescence. 2nd ed. London: Blackwell Scientific; 1962. 37. Little RJ. A test of missing completely at random for multivariate data with missing values. J Am Stat Assoc. 1988; 83:1198–1202. 38. Hao L, Naiman D. Introduction. 2007. Quantile Regression. SAGE; 2007:1–7. Available at: http://ezlibproxy.unisa.edu.au/ login?url=http://SRMO.sagepub.com/view/quantile-regression/ SAGE.xml. 39. Rice WR. Analyzing tables of statistical tests. Evolution. 1989;43:223–225. 40. Stinson JN, Kavanagh T, Yamada J, et al. Systematic review of the psychometric properties, interpretability and feasibility of self-report pain intensity measures for use in clinical trials in children and adolescents. Pain. 2006;125:143–157. 41. Wells JC, Fewtrell MS, Williams JE, et al. Body composition in normal weight, overweight and obese children: matched casecontrol analyses of total and regional tissue masses, and body composition trends in relation to relative weight. Int J Obes. 2006;30:1506–1513. 42. Shultz SP, Anner J, Hills AP. Paediatric obesity, physical activity and the musculoskeletal system. Obes Rev. 2009;10: 576–582. 43. Shultz SP, Sitler MR, Tierney RT, et al. Effects of pediatric obesity on joint kinematics and kinetics during 2 walking cadences. Arch Phys Med Rehabil. 2009;90:2146–2154. 44. Kujala UM, Taimela S, Viljanen T. Leisure physical activity and various pain symptoms among adolescents. Br J Sports Med. 1999;33:325–328. 45. Gordon-Larsen P, McMurray RG, Popkin BM. Determinants of adolescent physical activity and inactivity patterns. Pediatrics. 2000;105:e83.
r
2013 Lippincott Williams & Wilkins
Musculoskeletal Pain in Obese Compared With Healthy-Weight Children Margarita D. Tsiros, PhD,* Jonathan D. Buckley, PhD,* Peter R. C. Howe, PhD,* Jeff Walkley, PhD,w Andrew P. Hills, PhD,z and Alison M. Coates, PhD*
Objectives: To investigate whether obesity is associated with musculoskeletal pain in children. Materials and Methods: Obese (n = 107) and healthy-weight (n = 132) 10- to 13-year-old children (132 males, 107 females) participated in an observational case-control study. Children self-reported pain location (excluding abdominal pain), pain intensity (current and prior week), and pain prevalence (overall and lower limb) using the Pediatric Pain Questionnaire. Body composition was assessed (dual-energy x-ray absorptiometry) and children wore an accelerometer for 8 days. Results: After adjustment for accelerometry (weekly average counts per hour) and socioeconomic status, obese children had more intense pain (worst pain, P = 0.006), pain in more locations (Pr0.005), and a higher prevalence of lower limb pain (60% vs. 52% respectively, P = 0.012) than healthy-weight children. Significant relationships were observed between body mass index and total pain locations (Pr0.004, unadjusted and adjusted) and worst pain intensity (Pr0.009, adjusted for socioeconomic status/accelerometry). There were no significant relationships between percent body fat and pain variables (unadjusted/adjusted analyses, P = 0.262 to 1.0). Discussion: Obesity in children was associated with increased overall and lower limb musculoskeletal pain, for which body mass index was a stronger predictor than adiposity. Clinicians treating obese children should screen for pain and prescribe exercise programs that take their symptoms into account. Key Words: body mass index, child, pain, adiposity, obese
(Clin J Pain 2014;30:583–588)
P
ain in childhood is a frequent occurrence with around 15% to 25% of children and adolescents experiencing recurrent/chronic symptoms: specifically abdominal-related,
Received for publication January 8, 2013; revised October 7, 2013; accepted August 2, 2013. From the *Nutritional Physiology Research Centre, School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, SA; wSchool of Health Sciences, RMIT University, Melbourne, Vic; and zMater Mothers’ Hospital, Mater Research and Centre for Musculoskeletal Research, Griffith Health Institute, Griffith University, Nathan, QLD, Australia. Present address: Peter R. C. Howe, PhD, Clinical Nutrition Research Centre, University of Newcastle, Callaghan, NSW, 2308, Australia. The authors declare no conflict of interest. This study was funded by a seeding grant from the Physiotherapy Research Foundation and internal funds (ATN Centre for Metabolic Fitness). M.D.T. was supported by an Australian Post Graduate Award and the Nutritional Physiology Research Centre, Adelaide, SA, Australia. Supported by the Australian Technology Network Centre for Metabolic Fitness and the Physiotherapy Research Foundation, Camberwell, Vic, Australia. Reprints: Margarita D. Tsiros, PhD, Nutritional Physiology Research Centre, School of Health Sciences, University of South Australia, P.O. Box 2471, Adelaide, SA 5001, Australia (e-mail: margarita. [email protected]). Copyright r 2013 by Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
musculoskeletal-related, or headache-related pain.1 Obesity has been identified as a possible contributor to musculoskeletal pain in children because of increased loading and/or biomechanical alterations imposed by additional fat, possibly compounded by periods of rapid growth that are characteristic of childhood.2–4 Children with obesity experience lower health-related quality of life (HRQoL) and greater disability than their healthy-weight peers5,6 and obesity-related pain may be a contributing factor.7 A detailed review of pain in obese children and adolescents, concluded that most research has focused on low back pain (LBP),3 with prospective studies finding no link between body mass index (BMI) and LBP incidence.8–11 However, it was unclear in these studies whether obese children were adequately represented.8–11 In addition, many such studies assessed lifetime pain prevalence which has questionable recall reliability in children, especially in relation to chronic pain.12 Although few studies have specifically investigated overall musculoskeletal pain in obese children, findings tend to suggest a higher pain prevalence in obese compared with nonobese children.13–17 Most notable was a recent population-based survey of adolescents which found that those with obesity were more likely to report musculoskeletal pain (odds ratio of 1.33) with greater symptom severity than nonobese children17; although generalizability to all children is limited given the sample included only 17-year-olds. Other limitations of remaining research include a lack of validated pain tools,13–15 relying on physical examinations or interviews,13–15,18 not including a control group,16,18 or recruiting participants from an orthopedic clinic leading to a possible recruitment bias.16 Lower limb pain prevalence has also been positively associated with obesity in children13,18,19; however, studies have been limited to treatment-seeking samples and may not be generalizable to obese children in the community. Moreover, most research examining overall musculoskeletal pain has been limited in that it only assessed whether pain was present13–16,19–21 and on few occasions has the frequency of pain been captured.13,19 Three studies have assessed pain intensity which is particularly relevant as it has been linked with functional limitations in general pediatric samples.1,22,23 However, these studies have either focused on a limited age range that may not be generalizable to children of other ages,17 included nonmusculoskeletal pain (eg, abdominal pain)21 or have focused on severely obese treatment-seeking youths18 who may represent an extreme. Early adolescence has the potential to be a particularly high-risk period for the development of obesity-related pain given the many tumultuous changes occurring during this time (eg, hormonal changes, rapid growth, changes in physical activity behaviors, emotional/psychosocial changes). From an intervention perspective, early adolescence is also a better time to intervene to prevent obesity-related pain given www.clinicalpain.com |
583
Clin J Pain
Tsiros et al
that parents still tend to have some influence over their children’s lives. However, scant research has specifically focused on early adolescence. Taylor et al14 found that obese children (mean age 12.8 y) were 4.04 times more likely than nonobese children to have musculoskeletal pain, whereas Bell et al15 found that the odds of experiencing musculoskeletal pain increased >2-fold per unit increase in BMI z score (6 to 13 y olds). In contrast, studies limited to examining low back pain concluded that such pain is unrelated to BMI in 10- to 14-year-olds.7–9,24–26 Physical activity is another important consideration as it could confound any relationship between weight status and pain. Yet few studies (examining LBP only) have addressed the potential influence of physical activity.10,24,25,27,28 Although some have reported LBP to be associated with limited activity or certain types of sport/leisure,10,24,27,28 they have not considered overall physical activity levels nor used objective measures of physical activity, instead relying on self-report measures which are subject to recall bias.29 Furthermore, studies have not sought to examine whether physical activity levels could confound relationships between weight status and pain. There are many possible ways that physical activity could confound the obesity-pain relationship. For example, although speculative, physical activity could increase children’s tolerance to uncomfortable sensations, thereby causing them to have higher pain thresholds, reporting less intense pain. Alternatively, particular types of physical activity may stress or strain certain parts of the body resulting in pain. The majority of studies (21 of 22 cited in our recent review3) have not used internationally accepted criteria for defining weight status using BMI such as International Obesity Task Force criteria,30 thereby reducing the generalizability of their findings. Moreover, BMI may not provide an accurate indication of adiposity,31 which can be quantified by dual-energy x-ray absorptiometry (DXA). In essence, prior research has many methodological limitations and few studies have examined the impact of weight status on musculoskeletal pain with the exception of LBP. Therefore, the present study aimed to examine if children with obesity have increased musculoskeletal pain compared with their healthy-weight peers and to explore the relationship between adiposity and musculoskeletal pain. This is the first study in children to use a validated pain assessment tool, control for objectively assessed physical activity levels, and characterize body composition using internationally accepted criteria and DXA.
MATERIALS AND METHODS Obese and healthy-weight 10- to 13-year-olds, as defined by International Obesity Task Force criteria,30 were recruited from 3 Australian states as part of a larger observational case-control study.5 Recruitment methods included media advertisements/releases, flyers, school newsletters, and a pediatric clinic. Exclusions for the larger study comprised children who were underweight, overweight,30,32 intellectually/neurologically impaired, engaged in a weight-loss program in the preceding 3 months, had a past medical history of ear infections/balance issues, acute injury (prior 6 mo) necessitating medical management (eg, sprain or fracture), or a pathologic cause for their obese status (such as hypothyroidism). Ethical approval was provided by the University of South Australia, RMIT University, Queensland University of Technology and Flinders Medical Centre.
584 | www.clinicalpain.com
Volume 30, Number 7, July 2014
Pain Children completed age-appropriate versions of the Pediatric Pain Questionnaire (PPQ)33 under interviewer supervision. The PPQ is a reliable and valid pain measure in children33,34 which captures qualitative descriptors of pain/ discomfort, while also quantifying pain locations (using a body chart) and pain intensity using 100 mm horizontal visual analogue scales to report current pain and recall worst pain over the previous week. The PPQ assesses general pain—as the current study was concerned with musculoskeletal pain, abdominal/stomach pains were excluded from the analysis, whereas all other pain locations were included (ie, head/neck, trunk, upper and lower limb pains). Variables of interest included pain intensity (current and worst over past week), total number of current pain locations (ie, pain anywhere on the body with the exception of abdominal/stomach pains), and total number of current lower limb pain locations (ie, pain reported in the hips/ thighs/knee/shin/calf/ankle/foot/toe regions).
Body Composition Anthropometric and body composition measures were recorded as described in detail previously.5 Briefly, participants wearing only a hospital gown and underwear had body mass and height recorded to calculate BMI and BMI z score.4 Percent body fat was estimated from a whole-body DXA scan.
Physical Activity Participants wore a uniaxial accelerometer (Actigraph; MTI Health Services, Fort Walton Beach, FL, version 2.2 model 7164) on their right hip full-time for 8 days during the school-term, removing it only for water-based activities. Actigraphs sampled at 1-minute epochs and minimum data eligibility criteria included at least 4 weekdays and both weekend days with at least 10 “eligible” hours (defined as >3000 counts/h) of wear-time each day. Average counts per hour per day were then calculated to allow for a differing number of hours awake per day between participants. Weekday and weekend-day average counts per hour were then calculated and an overall average taken to provide an indication of activity over a given week (called weekly average counts per hour). Note that weekdays and weekend days were weighted equally as children spend approximately half of the total days of a year in school over a year (4 terms10 wk 5 d = 200 d, minus about 10 d of public holidays, and teacher-free days, and about 10 d of sickness = 180 d/y). For descriptive purposes, data were also reduced to average minutes of moderate to vigorous physical activity (MVPA) per day. Actigraphs have been reported to have strong interinstrument reliability and have been widely validated in children and adolescents.35
Background Information Information regarding average annual household income, maternal education, and child medical conditions/ medications were collected through a parent-completed questionnaire. The data on household income and maternal education were used as measures of socioeconomic status (SES). Children self-assessed their pubertal development using the Tanner scales.36
Statistical Analysis SPSS version 17 for WINDOWS (International Business Machines Corporation, New York) and STATA r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
Musculoskeletal Pain in Obese Children
TABLE 1. Pain Data
Obese N = 107
Healthy-Weight N = 132
Unadjusted Sig (P)
Adjusted Sig (Inc, Mat Ed) (P)
Adjusted Sig (Inc, Mat Ed, Accel) (P)
2.0 (8.0, 0.0-8.0) 26.0 (48.0, 5.0-53.0)
2.0 (5.9, 2.0-6.9) 19.0 (32.3, 4.5-36.8)
0.584 0.039
0.932 0.006*
0.790 0.006*
2.0 (2.0, 1.0-3.0) 1.0 (2.0, 0.0-2.0)
1.0 (1.5, 0.5-2.0) 1.0 (1.0, 0.0-1.0)
0.005* 0.037
0.006* 0.015
0.005* 0.010*
Outcomes Current pain (mm) Worst pain (last week) (mm) Total pain locations Lower limb pain locations
Data were not normally distributed and are presented as median (interquartile range, 25th to 75th percentile). *Significant difference between groups using an exact Mann-Whitney U test and a sequential Bonferroni significance level (Pr0.013). Monte Carlo P-values are reported. Obese and healthy-weight were defined using International Obesity Task Force criteria.30 Notably, the Mann-Whitney U test is a comparison of mean ranks rather than medians, thus the 2 groups can have the same medians, but still be statistically significantly different as was observed with lower limb pain locations Accel indicates accelerometry weekly average counts per hour; inc, household income; mat ed, maternal education; sig, significance.
(StataCorp LP, College Station, TX) were used for analyses. Abdominal/stomach pains were excluded from the analysis of pain location data for 4 participants in the obese group. Missing data were imputed by expectation maximization algorithms after applying Little test37 showing that data were missing completely at random (P = 0.61). Between-group differences were examined using the Mann-Whitney U test for non-normally distributed data. When data could not be transformed, variables were ranked before using analysis of covariance to adjust for potential confounders that significantly differed between obese and healthy-weight; namely, accelerometry, maternal education, and household income. Relationships between percent body fat or BMI and pain outcomes were examined using quantile regression—as such only unstandardized regression coefficients were available for report. Adjustments were made for the same confounding variables. Quantile regression uses the median rather than the mean, is robust against outliers and is also appropriate to use when the residuals are not normally distributed (as in the current study), thereby violating the assumptions of other regression techniques such as ordinary least squares in relation to hypothesis testing.38 To account for type I error, a post hoc sequential Bonferroni procedure was used.39 Data were not split by gender as sex was not found to be a significant effect modifier of the relationships between pain outcomes and weight status (P-values ranging from 0.062 to 0.712). Given the homogeneous age range of the sample, data were not stratified by age.
RESULTS Full-population demographics have been previously reported.5 In summary, 107 obese and 132 healthy-weight
children participated. Mean group ages were similar (11.8 ± 0.1 and 12.0 ± 0.1 y, P = 0.38) with a reasonably even distribution of sexes between groups (obese: 51 females, 56 males; healthy-weight: 56 females, 76 males, P = 0.42) and no differences in pubertal development (86% being Tanner stage 4 or less, P = 0.13) or height (1.55 ± 0.01 and 1.54 ± 0.01 m, P = 0.33). Notably, more obese children were from families of lower income (43% earning r$60 k vs. only 19% in the healthy-weight group) and lower maternal education (27% with a University degree or higher vs. 49% for healthyweight). Obese children had higher percent fat (45.4 ± 0.5 vs. 21.7 ± 0.6%, P < 0.01), BMI (29.6 ± 0.4 vs. 18.2 ± 0.2 kg/ m2, P < 0.01), and BMI z score (2.16 ± 0.02 vs. 0.03 ± 0.05, P < 0.01) and were also less physically active than their healthy-weight peers (27 037 ± 769 vs. 32 376 ± 1 091 counts/ h, P < 0.01 and 52.1 ± 2.6 vs. 73.8 ± 3.2 min/d of MVPA). The obese group reported pain in more body locations irrespective of accelerometry and/or SES (Pr0.005 for total number of pain locations, Table 1). Obese children reported experiencing significantly more intense pain than healthyweight children in the preceding week (worst pain adjusted for accelerometry and/or SES factors, P = 0.006, Table 1). Furthermore, lower limb pain was more prevalent in obese compared with healthy-weight children (60% vs. 52% respectively, P = 0.012), with obese children also reporting a greater number of lower limb pain locations when controlling for accelerometry and SES factors (P = 0.010, Table 1). There were significant relationships between BMI and the total number of pain locations (adjusted and unadjusted analyses, Pr0.004) and worst pain over the prior week (when adjusted for accelerometry and SES factors, P = 0.009) (Table 2). In contrast, there were no significant relationships between
TABLE 2. Relationships Between BMI and Pain Outcomes in Obese and Healthy-Weight Children
Unadjusted
Adjusted (Inc, Mat Ed)
Adjusted (Inc, Mat Ed, Accel)
Pain Outcomes
Unst. r
P
Unst. r
P
Unst. r
P
Current pain (mm) Worst pain (last week) (mm) Total pain locations Lower limb pain locations
0.00 0.977 *0.05 0.00
1.00 0.04 *0.001 1.00
0.02 1.24 *0.06 0.00
0.667 0.022 0.002 1.00
0.02 *1.38 *0.07 0.00
0.617 *0.009 *0.004 1.00
Quantile regression was used as the residuals were not normally distributed, hence only unstandardized regression coefficients were available for report. *Significant relationship between BMI and the pain outcome using a sequential Bonferroni significance level (Pr0.013). Accel indicates accelerometry weekly average counts per hour; BMI, body mass index; Inc, household income; mat ed, maternal education; Unst. r, unstandardized regression coefficient.
r
2013 Lippincott Williams & Wilkins
www.clinicalpain.com |
585
Clin J Pain
Tsiros et al
percent body fat and pain variables in unadjusted or adjusted analyses (P-values ranging from 0.262 to 1.0) (data not shown).
DISCUSSION Previous research examining pain prevalence and frequency suggests that children with obesity experience more musculoskeletal pain than their healthy-weight counterparts.14–16,19,28 The current study extends on these findings, indicating that obese children experience musculoskeletal pain in more body locations than healthyweight children. More intense overall musculoskeletal pain, combined with a higher prevalence of lower limb pain also featured in obese compared with healthy-weight children after correcting for differences in SES and/or accelerometry, which is consistent with prior research in an older sample.17 Although de Sa Pinto et al13 also reported a higher prevalence of lower limb pain in obese children, they did not use a validated pain assessment tool. As expected, children with higher BMIs tended to report more pain in terms of total body locations and more intense pain over the prior week. Considering the 11.4 kg/ m2 mean BMI differential between the obese and healthyweight group, the observed coefficients would translate into the obese group experiencing 15.7 mm greater intensity of worst pain over the prior week and having 0.8 more pain locations in their body than the healthy-weight group. Given that a change of B10 mm is generally considered “meaningful” on a 100-mm visual analogue scale,40 the relationship between BMI and worst pain intensity over the prior week has particular clinical relevance. In contrast, there was no correlational relationship between percent body fat and any musculoskeletal pain variables. These discrepant findings suggest that the total mass (used to calculate BMI) rather than the adiposity of the child may be of greater importance with regards to musculoskeletal pain. This premise fits with prior research reporting associations between increased weight/BMI and lower limb pain prevalence in a treatment-seeking sample of severely obese children.19 Obesity is characterized by excessive fat, but there are also compensatory increases in fat-free mass and skeletal mass41 which contribute to the total mass a child is required to move during activities of daily living. It is conceivable that children with obesity (as defined by BMI) may exert a greater load on their joints and tissues and that this excessively prolonged loading could lead to biomechanical deviations and musculoskeletal pain.42 Certainly, evidence of increased joint moments/forces has been reported in children with obesity during locomotion when compared with their healthy-weight peers.43 Although not the primary focus of this work, it would appear that SES and overall physical activity may play some role in the musculoskeletal pain experience, as for a number of outcomes, pain was significantly greater (or increased with higher BMI) only after controlling for lower physical activity and SES in the obese group. Hence, this raises the question of whether obese children would experience more pain if they were more active (thereby placing greater load/stress on their joints). Previous research links higher levels of leisure-time physical activity with an increased frequency of musculoskeletal pain (and vice versa) in general pediatric populations aged 10 to 16 years (N = 698), suggesting that sport may induce stress/acute injuries in the growing child44; a finding further supported
586 | www.clinicalpain.com
Volume 30, Number 7, July 2014
by studies of LBP in children.24,27,28 Alternatively, others have found that low leisure activity may be associated with LBP.10,27 It is also possible that the presence of pain may impact a child’s willingness and ability to participate in physical activity, as suggested by Wilson et al21 who found that pain intensity was a stronger predictor of difficulties with active pursuits than BMI. Although the current study extends on prior research by examining overall physical activity levels (measured objectively), treating it as a confounding factor, there is clearly much work that still needs to be done to dissect multivariate relationships between obesity, musculoskeletal pain, and physical activity, which was beyond the scope of this study. Regarding SES, it could be that children from higher SES backgrounds have a lower musculoskeletal pain threshold (or vice versa), thereby confounding obese/healthy-weight differences, although this explanation is speculative. Alternatively, SES factors could also influence pain indirectly by physical activity, given that a higher maternal education/ household income has been shown to promote higher MVPA levels in children (N = 17 766, 11 to 21 y olds).45 This could be due to a greater awareness of the health benefits of physical activity in higher educated families, or participation in user-paid sport/recreation. There is a paucity of literature regarding the impact of obesity on musculoskeletal pain in children.13–16,20 This study makes a novel contribution, offering insight into the intensity of musculoskeletal pain and how widespread it is in the lower limbs and throughout the body. It should be noted that pain from the abdominal region was excluded from the analysis of number of pain locations (impacting 4 obese participants) however, it is possible that if a participant experienced abdominal pain that this may have influenced their rating of pain intensity and hence these data should be interpreted with caution. Particular strengths of this study included the use of internationally accepted definitions of weight status and DXA to characterize adiposity, considerations for objectively measured physical activity, and the use of a validated pain assessment tool (PPQ). Despite this, the PPQ does not offer an exhaustive analysis of the pain experience; for example, the chronicity/recurrence of symptoms, whether the amount of time a child has been obese has any bearing on symptoms, and whether pain thresholds and coping strategies differ between obese and healthy-weight children remains to be explored. Because of the limited age range of children in this study (10 to 13 y old), the generalizability of these findings should be noted. While there were no differences in pubertal development between the healthy-weight and obese groups, these findings may not translate to older or younger children as we cannot rule out that the data could have been influenced by the developmental status of the participants. In addition, it is possible that differences in pain due to weight status may change for a variety of reasons as children develop. The findings of this study have a number of potential clinical implications. Pain is an unpleasant, noxious experience which is always characterized by a certain amount of suffering.22 Not surprisingly, a child’s well-being (otherwise known as HRQoL) is compromised by pain.7 Considering that childhood obesity is already linked with many adverse health outcomes including marked reductions in HRQoL,6 the coexistence of pain and obesity have the potential to further compound impairments in well-being; a premise confirmed by Hainsworth et al’s7 finding that cooccurring r
2013 Lippincott Williams & Wilkins
Clin J Pain
Volume 30, Number 7, July 2014
chronic pain and obesity increased the risk of impaired physical HRQoL nearly 3-fold compared with either condition in isolation. The presence of pain in children with obesity may also pose a barrier to their participation in physical activity,21 whereas certain types of activity could have the potential to induce/aggravate symptoms.24,27,2,8,44 In conclusion, this work suggests that even in children aged 10 to 13 years, obesity may be associated with increased overall and lower limb musculoskeletal pain, although overall mass appears to be of greater importance than adiposity. Physical activity and SES also appear to play a role, especially with regard to musculoskeletal pain intensity and lower limb pain, but further research is needed to dissect these effects. Clinicians treating children with obesity should screen for musculoskeletal pain and, if needed, refer for management (eg, medical review, physiotherapy), while taking care to design exercise programs that will not aggravate existing symptoms. Low-impact activity such as walking, swimming, or aqua exercise may be better suited to children with obesity experiencing pain, although further research is needed to confirm this. For some children, specialist prescription (eg, by a physiotherapist or exercise physiologist) of appropriate exercises for their musculoskeletal pain condition may also be required.
ACKNOWLEDGMENTS The Quality of Life study described in this paper was carried out using the PedsQL developed by Dr James W. Varni. The authors thank Professor Adrian Esterman, PhD (School of Nursing and Midwifery, Sansom Institute for Health Research, University of South Australia, Adelaide, SA, Australia) (advice on statistical analysis) and to Dr Paul Grimshaw, PhD (School of Mechanical Engineering, University of Adelaide, SA, Australia), Professor Timothy Olds, PhD (Health and Use of Time Group, School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, SA, Australia) and Dr Anthony Shield, PhD (School of Exercise and Nutrition Sciences, Institute of Health Biomedical Innovation, Queensland University of Technology, Brisbane, QLD, Australia) (general advice on study design). Technical assistance was provided with thanks to: Ms Keren Kneebone, B HSci (Human Movement) (School of Exercise and Nutrition Sciences, Deakin University, Burwood, Vic, Australia) Dr Rachel Wood, PhD (Centre of Research Excellence in Translating Nutritional Science to Good Health, University of Adelaide, Adelaide, SA, Australia) Dr Sarah Shultz, PhD (School of Sport and Exercise, Massey University, Palmerston North, New Zealand) Dr Masa Kagawa, PhD (Institute of Nutrition Sciences, Kagawa Nutrition University, Saitama, Japan) Ms Lara Taylor, B App Sci (Psych Hon) (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Mr Richard Mallows, PhD, (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Mr Kaine Grigg, B App Sci (Psych Hon) (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) Ms Fiona Spargo, M Sci Hons, (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia) and Ms Kate Greenway, PhD (Discipline of Exercise Science, School of Medical Science, RMIT University, Melbourne, Vic, Australia). r
2013 Lippincott Williams & Wilkins
Musculoskeletal Pain in Obese Children
REFERENCES 1. Roth-Isigkeit A, Thyen U, Stoven H, et al. Pain among children and adolescents: restrictions in daily living and triggering factors. Pediatrics. 2005;115:e152–e162. 2. Hestbaek L, Leboeuf-Yde C, Kyvik KO. Are lifestyle-factors in adolescence predictors for adult low back pain? A crosssectional and prospective study of young twins. BMC Musculoskelet Disord. 2006;7:27. 3. Tsiros MD, Coates AM, Howe PR, et al. Obesity: the new childhood disability? Obesity Rev. 2011;12:26–36. 4. Wearing SC, Hennig EM, Byrne NM, et al. Musculoskeletal disorders associated with obesity: a biomechanical perspective. Obes Rev. 2006;7:239–250. 5. Tsiros MD, Buckley JD, Howe PRC, et al. Day-to-day physical functioning and disability in obese 10-13 year olds. Pediatr Obes. 2012;8:31–41. 6. Tsiros MD, Olds T, Buckley JD, et al. Health-related quality of life in obese children and adolescents. Int J Obes. 2009;33: 387–400. 7. Hainsworth KR, Davies WH, Khan KA, et al. Co-occurring chronic pain and obesity in children and adolescents: the impact on health-related quality of life. Clin J Pain. 2009;25:715–721. 8. Nissinen M, Heliovaara M, Seitsamo J, et al. Anthropometric measurements and the incidence of low back pain in a cohort of pubertal children. Spine. 1994;19:1367–1370. 9. Poussa MS, Heliovaara MM, Seitsamo JT, et al. Anthropometric measurements and growth as predictors of low-back pain: a cohort study of children followed up from the age of 11 to 22 years. Eur Spine J. 2005;14:595–598. 10. Salminen JJ, Erkintalo M, Laine M, et al. Low back pain in the young. A prospective three-year follow-up study of subjects with and without low back pain. Spine. 1995;20:2101–2107. 11. Lake JK, Power C, Cole TJ. Back pain and obesity in the 1958 British birth cohort. cause or effect? J Clin Epidemiol. 2000;53: 245–250. 12. Erskine A, Morley S, Pearce S. Memory for pain: a review. Pain. 1990;41:255–265. 13. de Sa Pinto AL, de Barros Holanda PM, Radu AS, et al. Musculoskeletal findings in obese children. J Paediatr Child Health. 2006;42:341–344. 14. Taylor ED, Theim KR, Mirch MC, et al. Orthopedic complications of overweight in children and adolescents. Pediatrics. 2006;117:2167–2174. 15. Bell LM, Byrne S, Thompson A, et al. Increasing body mass index z-score is continuously associated with complications of overweight in children, even in the healthy weight range. J Clin Endocrinol Metab. 2007;92:517–522. 16. Podeszwa DA, Stanko KJ, Mooney JF, et al. An analysis of the functional health of obese children and adolescents utilizing the PODC instrument. J Pediat Orthop. 2006;26:140–143. 17. Deere KC, Clinch J, Holliday K, et al. Obesity is a risk factor for musculoskeletal pain in adolescents: findings from a population-based cohort. Pain. 2012;153:1932–1938. 18. Hainsworth KR, Miller LA, Stolzman SC, et al. Pain as a comorbidity of pediatric obesity. ICAN. 2012;4:315–320. 19. Stovitz SD, Pardee PE, Vazquez G, et al. Musculoskeletal pain in obese children and adolescents. Acta Paediatr. 2008;97:489–493. 20. Wake M, Salmon L, Waters E, et al. Parent-reported health status of overweight and obese Australian primary school children: a cross-sectional population survey. Int J Obes. 2002; 26:717–724. 21. Wilson AC, Samuelson B, Palermo TM. Obesity in children and adolescents with chronic pain: associations with pain and activity limitations. Clin J Pain. 2010;26:705–711. 22. Kashikar-Zuck S, Goldschneider KR, Powers SW, et al. Depression and functional disability in chronic pediatric pain. Clin J Pain. 2001;17:341–349. 23. Gauntlett-Gilbert J, Eccleston C. Disability in adolescents with chronic pain: patterns and predictors across different domains of functioning. Pain. 2007;I131:132–141.
www.clinicalpain.com |
587
Tsiros et al
24. Watson KD, Papageorgiou AC, Jones GT, et al. Low back pain in school children: the role of mechanical and psychosocial factors. Arch Dis Child. 2003;88:12–17. 25. Chiang HY, Jacobs K, Orsmond G. Gender-age environmental associates of middle school students’ low back pain. Work. 2006;26:197–206. 26. Szpalski M, Gunzburg R, Balague F, et al. A 2-year prospective longitudinal study on low back pain in primary school children. Eur Spine J. 2002;11:459–464. 27. Masiero S, Carraro E, Celia A, et al. Prevalence of nonspecific low back pain in school children aged between 13 and 15 years. Acta Paediatr. 2007;97:212–216. 28. Kovacs FM, Gestoso M, Gil del Real MT, et al. Risk factors for non-specific low back pain in school children and their parents: a population based study. Pain. 2003;103:259–268. 29. Baranowski T. Validity and reliability of self-report measures of physical activity: an information procesing perspective. Res Q Exerc Sport. 1988;59:314–327. 30. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320:1240–1243. 31. Wickramasinghe VP, Cleghorn GJ, Edmiston KA, et al. Validity of BMI as a measure of obesity in Australian white caucasian and Australian Sri Lankan children. Ann Hum Biol. 2005;32:60–71. 32. Cole TJ, Flegal KM, Nicholls D, et al. Body Mass Index cut offs to define thinness in children and adolescents: international survey. BMJ. 2007;335:194. 33. Varni J, Thompson K, Hanson V. The Varni/Thompson Pediatric Pain Questionnaire. I. Chronic musculoskeletal pain in juvenile rheumatoid arthritis. Pain. 1987;28:27–38. 34. Gragg RA, Rapoff MA, Danovsky MB, et al. Assessing chronic musculoskeletal pain associated with rheumatic disease: further validation of the Pediatric Pain Questionnaire. J Pediatr Psychol. 1996;21:237–250.
588 | www.clinicalpain.com
Clin J Pain
Volume 30, Number 7, July 2014
35. Trost SG. State of the art reviews: measurement of physical activity in children and adolescents. Am J Lifestyle Med. 2007;1:299–314. 36. Tanner JM. Growth at Adolescence. 2nd ed. London: Blackwell Scientific; 1962. 37. Little RJ. A test of missing completely at random for multivariate data with missing values. J Am Stat Assoc. 1988; 83:1198–1202. 38. Hao L, Naiman D. Introduction. 2007. Quantile Regression. SAGE; 2007:1–7. Available at: http://ezlibproxy.unisa.edu.au/ login?url=http://SRMO.sagepub.com/view/quantile-regression/ SAGE.xml. 39. Rice WR. Analyzing tables of statistical tests. Evolution. 1989;43:223–225. 40. Stinson JN, Kavanagh T, Yamada J, et al. Systematic review of the psychometric properties, interpretability and feasibility of self-report pain intensity measures for use in clinical trials in children and adolescents. Pain. 2006;125:143–157. 41. Wells JC, Fewtrell MS, Williams JE, et al. Body composition in normal weight, overweight and obese children: matched casecontrol analyses of total and regional tissue masses, and body composition trends in relation to relative weight. Int J Obes. 2006;30:1506–1513. 42. Shultz SP, Anner J, Hills AP. Paediatric obesity, physical activity and the musculoskeletal system. Obes Rev. 2009;10: 576–582. 43. Shultz SP, Sitler MR, Tierney RT, et al. Effects of pediatric obesity on joint kinematics and kinetics during 2 walking cadences. Arch Phys Med Rehabil. 2009;90:2146–2154. 44. Kujala UM, Taimela S, Viljanen T. Leisure physical activity and various pain symptoms among adolescents. Br J Sports Med. 1999;33:325–328. 45. Gordon-Larsen P, McMurray RG, Popkin BM. Determinants of adolescent physical activity and inactivity patterns. Pediatrics. 2000;105:e83.
r
2013 Lippincott Williams & Wilkins
