Low back pain in adolescent gymnasts. Prevalence and risk factors.
Abstract: Objective--To evaluate the prevalence, determinants and risk factors for Low back pain (LBP) in club-level competitive adolescent gymnasts compared to the general adolescent population.

Design--A cohort study.

Setting--School of Physiotherapy--University of Bologna (Italy).

Participants--ninety-one gymnasts aged 11-14 years were selected for this study and three-hundred and seventy-five age- matched general adolescents were used as a control.

Assessment of risk factors--A questionnaire concerning back pain, physical activity, social-behavioral factors, and the measurement of anthropometric parameters and lumbar range of motion were used.

Main outcome measurement--The prevalence of LBP.

Results--Low-level LBP was reported by 46% of the gymnasts and 60% of the students in the control group; medium/high level LBP was reported by 26% of the gymnasts and 36% of the control group. In both groups, the risk of LBP was higher in females, in adolescents with parents or siblings suffering from LBP and in children who spent more time in a sedentary position. For the gymnasts, an increased or reduced lumbar three-direction ROM and certain psychosocial factors seemed to be risk factors for LBP. No significant relationship emerged between LBP and height, weight, BMI, number of training hours, type of training, or years of sport practice.

Conclusions--The results are in line with other studies on similar populations (adolescent and medium agonistic level), but they differ from the findings of research carried out on elite or Olympic athletes. LBP prevalence is lower in competitive, club-level adolescent gymnasts than in adolescent non-gymnasts.
Article Type: Report
Subject: Teenagers (Health aspects)
Youth (Health aspects)
Low back pain (Risk factors)
Low back pain (Diagnosis)
Low back pain (Research)
Gymnasts (Health aspects)
Authors: Vanti, Carla
Gasperini, Margherita
Morsillo, Filomena
Pillastrini, Paolo
Pub Date: 04/01/2010
Publication: Name: Scienza Riabilitativa Publisher: Associazione Italiana Fisioterapisti Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Associazione Italiana Fisioterapisti ISSN: 1828-3942
Issue: Date: April, 2010 Source Volume: 12 Source Issue: 2
Topic: Event Code: 310 Science & research
Product: Product Code: E121930 Youth
Geographic: Geographic Scope: Italy Geographic Code: 4EUIT Italy
Accession Number: 255363952

Low Back Pain (LBP) is a phenomenon which affects not only the adult population, but also children and adolescents. [1,2] Because children who suffer LBP have a higher probability of suffering such pain as adults, [3] it is important to study prevalence and risk factors for LBP in the younger population.

According to the literature, the prevalence of LBP in adolescents ranges between 7% and 72% [4]): in fact, there is a remarkable difference between several studies [5,6,7]. Moreover, determinants and risk factors for LBP at this age are still uncertain. The most cited risk factors are: gender, anthropometric factors, physical activity and sports, psychological and social profiles, and habits such as smoking and carrying loads, particularly school backpacks. [8, 9 10] More specifically, the female gender, obesity, intense levels of physical activity [11,12,13] and agonistic sports have been shown to be related to LBP [13,14,15]. Likewise Hellstrom [16] and Kujala [17] reported a higher rate of LBP in general athletes, respectively 65% and 46%. However other studies did not confirm such a direct relationship. [18]

The prevalence of LBP in gymnasts ranges from 20% [19] to 23% [20] in medium-high level athletes, and is 65.4% in female and 84.6% in male national level athletes [21]. Therefore, LBP in gymnasts appeared to be significant, but the studies concerning elite or Olympian gymnasts were conducted on very specific and small samples. Sward [22] found a significantly higher rate of LBP symptoms in elite gymnasts than in the control group (79% versus 37.5%). Kolt [23] observed that high-level gymnasts report a higher prevalence of injuries than medium-level ones, but this observation is inverted when the percentage of injuries per training hours is considered.

Moreover several authors reported a higher incidence of lumbar spondylolysis and spondylolisthesis in gymnasts, in comparison with the non-athletic population, probably caused by repetitive lumbar extension. More specifically, the prevalence of spondylolysis in gymnasts ranged between 16.3% and 19.2%; according to Goldberg [24], the prevalence was 9.1% in pre-elite, 21.4% in elite and 12.5% in national level gymnasts. Several authors reported a higher incidence of pars interarticularis defects in elite gymnasts compared to the general caucasian population; on the other hand, some authors reported the absence of a significant relationship between structural abnormalities and LBP [25,26].

We must consider that the data collected from the small, high-level gymnast groups may not reflect the reality for club-level competitive gymnasts, who form the largest group of practising gymnasts. This cross-sectional study aims to evaluate the prevalence and possible associated risk factors for LBP in competitive club-level gymnasts aged 11-14 years, and to compare these findings with an aged-matched general population attending junior high school in the same geographic area (Regione Emilia-Romagna, located in the Northern part of Italy).


The study included two groups of adolescents, aged 11-14 years. Group#1 (G1): 91 competitive club-level gymnasts (85 females and 6 males) attending the main gymnastic clubs of the Emilia-Romagna Region. Group#2 (G2): 375 adolescents (173 females and 202 males) attending junior high school in 3 institutes of the Emilia-Romagna Region (one in the city centre of Bologna, one in the outskirts, one in a country village). Another inclusion criterion was the consent of both parents and of the adolescents. Researchers of the University of Bologna, in collaboration with educational and sports teams, developed this study from June 2006 to March 2008. It was carried out with the cooperation of physiotherapists who measured anthropometric parameters and lumbar range of motion (ROM), and collected information concerning back pain, physical activity, and social and behavioral factors, using a questionnaire. Anthropometric parameters (weight and height) were taken for each adolescent, using respectively an electronic weighting scales and an anthropometer. Due to practical reasons, the three-dimensional lumbar ROM was recorded only for the gymnasts by an electronic motion evaluation system named the "Vector Bte", produced by "Baltimore Therapeutic Equipment"--Hanover, Maryland, USA. This piece of equipment records the reciprocal three-dimensional position of two points targeted on the spine, located at S2 and T12 levels. The Vector system includes a PC based program; it has a database storage and report-generating capabilities. The external unit is lightweight and straps onto the back.

The questionnaire employed for this study was composed of 66 questions. The first section of the questionnaire contained questions concerning the students' environment and context from a socio-demographic perspective. It was asked if the adolescent had ever suffered from back pain. If the answer was affirmative, he/she then had to answer the questions contained in the second section, which gathered more detailed information regarding back pain. The third section contained information on physical activity and studied the frequency, amount and characteristics of training, the agonistic level, the amount and type of trauma and injuries, the behavior adopted when the LBP occurred. Other questions concerned lifestyle (for example how many hours were spent in a sitting position, watching TV and using a PC); smoking habits were also recorded. The last section of the questionnaire investigated the psychosocial area, particularly the self-perception and the quality of relationships with friends and parents. This questionnaire was given to the children to complete at home alone, or with the help of their parents.

In order to define back pain, the following questions were used: "Have you ever had a backache and with what frequency?" and "How would you rate your usual pain in a scale from 0 to 10?" Children who referred having a frequency of back pain of at least "sometimes" were included in "Profile 1: Low-level LBP Group" (P1); children who also answered "with an intensity [greater than or equal to] 4/10" were included in "Profile 2: Medium/High-level LBP Group" (P2). All data collected was registered in an electronic database. Keeping the identifying data separated from all questionnaire data ensured confidentiality.

A descriptive statistical analysis studied the prevalence and the characteristics of LBP. To emphasize the most relevant and statistically significant risk factors and their relationship to the two cited Profiles, the most common risk factors cited in literature were analysed. The analysis of the relationship between the risk of LBP and the kind of sport practiced was done through a univariate analysis. For the analysis of the relationship between lumbar ROM and risk of LBP a statistical technique called "Cluster Analysis" was used, which allowed identification of different groups of adolescents based on the measurement of lumbar ROM. A multivariate analysis of variance was employed for estimating the significant differences between the multi-dimensional mean value of the groups. A risk estimation (OR) for these factors was calculated. The method chosen for the OR estimation was the logistic regression for each group (gymnasts and students).


The two groups are similar for age (12.3 [+ or -] 3.63 for the gymnasts and 13.07 [+ or -] 0.95 for the students), but different for gender (the percentage of female in gymnasts group was 93% while in the other was 46,1%), height (144.9cm [+ or -] 12.2 for the gymnasts and 154.7cm [+ or -] 16.3 for the students), weight (38.5 kg [+ or -] 12.1 for the gymnasts and 49.8 kg [+ or -] 11.7 for the students), and BMI (17.9 [+ or -] 2.62 for the gymnasts and 20.3 [+ or -] 3.4 for the students).

We did not find any significant relationship between height, weight and BMI, and LBP. The prevalence of LBP in the two groups is described in Table 1. For both profiles, LBP was reported less by the gymnasts than by the other students. For both groups, the regions of the body most frequently involved in back pain were the lumbar region (75% of the gymnasts, 55% of the students) and the thoracic region (28% of the gymnasts, 35% of the students); the duration of the LBP was in the most part of cases (56% of the gymnasts; 76% of the students) only one day. The type and the location of pain were similar between gymnasts and students. The most reported causes for LBP provocation were "to strain oneself" (50% of the gymnasts, 38% of the students ); for an increase in LBP were "to strain oneself" (76% of the gymnasts, 58% of the students) and "to carry school backpack" (35% of the gymnasts, 56% of the students). During the practice of artistic gymnastics, the most reported cause for LBP provocation was floor exercise (P1=66.7%; P2=79.2%); for an increase in LBP were landing from a jump (P1=38.1%; P2=58.3%), pushing (P1=38.1%; P2=45.8%) and sustained static positions (P1= 26.2%; P2=33.3%). We did not find any relationship between LBP and the amount or frequency of training, the amount and characteristics of injuries or trauma, the type of training, the agonistic level, and the years of sport practice.

It is interesting to observe that only half of the gymnasts reported the pain felt during training to the trainer; all the other gymnasts did not. With respect to gender, the percentage of females was more relevant in the gymnasts' sample (93%) in comparison with non-gymnasts (46%): therefore, the estimation of risk related to gender was calculated on the entire adolescent group. The risk for both LBP Profiles was higher in females, but statistically significant only for the P2 profile (OR=1.86; P=0.004; CI=1.22-2.83). With respect to family, analysing the entire group, a statistically significant relationship between LBP in parents and/or siblings and P1 was found. In fact the risk for LBP doubles for adolescents who have at least one LBP-sufferer in the same family (OR=0.566; P=0.021; CI=0.348 0.919). When the analysis was conducted for each group individually, the same tendency appeared, but statistically significant only for the students group.

With respect to lifestyle, and particularly the amount of time spent sitting during the day, the gymnasts with LBP (P1 and P2) were in general more sedentary than the LBP students. The relationship between the amount of hours spent sitting by the gymnasts (especially in the afternoon and in the evening), and LBP-Profiles was statistically significant. For the students, the relationship between watching TV and P1 was statistically significant (See Table 2). We did not find any relationship between LBP and smoking habits. Concerning the practice of sport, obviously the totality of G1 was involved in agonistic sport and more than 80% trained from four to six times a week. In the G2 group, 42% of P1 adolescents and 47,5% of P2 adolescents played competitive sports, from two to four times a week. In the G2 group, the frequency of training for non-LBP adolescents was about the same. To make a comparison between artistic gymnastics and other kinds of sports (individual or team sports), and to better understand the risk factors for each category of sport, a univariated analysis was conducted for subgroups. The result of this comparison is a lower risk for LBP Profiles if the adolescent was involved in individual-agonistic sport, especially artistic gymnastics. Adolescents who practised agonistic soccer, basketball or volleyball had a 1,77 times more risk of low-level LBP (P1) than those who practised artistic gymnastics.

The measurement of lumbar ROM by "Vector Bte" on the gymnasts showed in general a greater ROM related to the standard adolescents, but we did not find any relevant relationship between the augmentation of a single movement ROM and LBP. Therefore, we decided to further analyse the total amount of ROM (flexion plus extension, plus left and right lateral bending, plus left and right rotation). Using the statistical technique named "Cluster Analysis", we found three "ROM groups", with similar characteristics within each group, but statistically different among groups: ROM1=poor ROM, ROM2=moderate ROM, and ROM3=large ROM. The results showed a statistically significant relationship between the "extremes" of ROM (poor ROM and large ROM) and the P1 profile (see Table 3). With respect to the psychosocial items, for the gymnasts the perception of a difficult relationship with their parents (especially his/her mother) was a risk factor for Profile 1 (OR=2,20) and for Profile 2 (OR=2,99); also the perception of non-acceptance by his/her classmates was a risk factor for Profile 1 (OR=2,09) and for Profile 2 (OR=1,87). For the students, we did not find any significant correlation between psychosocial aspects and LBP.


The results of this study show a lower prevalence of LBP in club-level competitive adolescent gymnasts, in comparison with the general adolescent population. This surprising result is in contrast with the findings of some research carried out on more specific gymnasts groups, such as elite or Olympic level athletes. The explanation for this difference could be related to the extremely high amount of training and competition and the consequent high percentage of injuries and trauma. These specific characteristics do not reflect the situation of the club-level gymnasts. Our results are comparable to a study on a similar population (adolescent and medium agonistic level): more specifically, in a study on 67 rhythmic adolescent gymnasts, Cupisti [27] concluded that the athletes showed lower prevalence of LBP than a matched general adolescent population. According to many authors, floor exercises are the most common culprit of LBP in gymnasts; movements which worsen pain are landing from a jump (probably because reaction forces at a lumbar level are sharply higher than impact forces during landing), pushing and sustained static positions, as reported by Hall [28].

With respect to gender, our results are in agreement with many other studies, which consider the female gender as being a significant risk factor for LBP in adolescent people. The statistically significant relationship between LBP in parents and/or siblings and LBP in adolescents is in line with the most part of scientific studies. We cannot confirm or deny anything about the association between familiarity and back pain in gymnasts, because in literature it is uncommon to find studies that suggest familiarity as a risk factor in athletes. However, it is not clear if the familiarity factor should be attributed to a greater attention of the parents and/or siblings towards this problem.

The sitting position is one of the risk factors most commonly associated with LBP. In our results it is surprising to note that, in contrast to data reported by these authors, only the sitting position while watching TV is a statistically significant risk factor for the students, while for the gymnasts the amount of time in the general sitting position is a risk factor. The sedentary lifestyle has almost never been investigated in studies on athletes, because it might seem a contradiction. Only recently a study [29] showed that in female athletes an increase in the number of hours in the sitting position increases significantly the risk of LBP. With respect to the relationship between the sum of three-dimensional lumbar ROM and LBP in gymnasts, the literature does not deny or confirm the information above. These studies focused on the sagittal plane without considering the three direction of spinal movement in a comprehensive manner. This is probably because pain in gymnasts seems to be due to overuse, especially hyperextension movements. In our study it is not hyperextension alone which is a risk factor for LBP, but a relevant increase or decrease of the three-directional lumbar ROM. The study carried out by Ohlen [20] claims that both an increased lordosis and a total ROM on a sagittal plane of less than 100[degrees] could increase the overload of the spine, but he did not find any statistical significance. Likewise in our study, the measurement of flexion-extension ROM is not enough, because the results were similar for gymnasts with or without pain. With respect to the relevance of psychosocial factors for the gymnasts or for the athletes in general, these factors are studied mainly in relation to the performance in competition and not in relation to pain. Often high-level athletes underestimate pain in order to avoid the risk of missing a competition. This observation could explain the higher LBP reported by adolescents who practise agonistic team sports compared to artistic gymnastics. The underestimation of pain, especially in individual sports, is emphasized by a pressure linked to the expectations of coaches, parents and supporters and including, for certain types of sports, the pressure related to economic interests. To confirm that finding, our study showed that only half of the gymnasts who had LBP during training reported it, while the other half continued to train. It is interesting to note that a significant risk factor for LBP is a difficult relationship with classmates and not with friends or companions of gymnastics, perhaps because the high amount of hours spent each week in training (for some athletes even more hours than spent at school) creates a strong group adherence, which is probably not reflected at school. We conclude that sedentary lifestyle and some psychosocial factors seem relevant risk factors for LBP. The myth which considers artistic gymnastics as a sport harmful to the back of adolescents, at least for female club-level athletes, should be dispelled. The prevalence of LBP in adolescent club-level gymnasts is lower than in high-level gymnasts, and, finally, adolescents who practise an agonistic team sport seem more at risk for LBP than those who practise gymnastics.

ACKNOWLEDGMENTS: The authors thank Mrs. Caroline O'Reilly for her careful revision of the English text.


[1.] Hirsch C, Jonsson B, Lewin T. Low-back pain symptoms in a Swedish female population. Clin Orthop Relat Res. 1969; 63:171-176.

[2.] Salminen JJ. The adolescent back. A field survey of 370 Finnish schoolchildren. Acta Paediatr Scand Suppl. 1984;315:1-122.

[3.] Harreby M, Neergaard K, Hesselsoe G, et al. Are radiologic changes in the thoracic and lumbar spine of adolescents risk factors for low back pain in adults? A 25-year prospective cohort study of640 school children. Spine. 1995;21:2298-2302.

[4.] Jeffries LJ, Milanese SF, Grimmer-Somers KA. Epidemiology of adolescent spinal pain: a systematic overview of the research literature. Spine. 2007;23:2630-2637.

[5.] El-Metwally A, Salminen JJ, Auvinen A, et al. Risk factors for development of non-specific musculoskeletal pain in preteens and early adolescents: a prospective 1-year follow-up study. BMC Musculoskelet Disord. 2007;23: 8-46.

[6.] Harreby M, Nygaard B, Jessen T, et al. Risk factors for low back pain in a cohort of 1389 Danish school children: an epidemiologic study. Eur Spine J. 1999;6:444-450.

[7.] Kovacs FM, Gestoso M, Gil del Real MT, et al. Risk factors for non-specific low back pain in schoolchildren and their parents: a population based study. Pain. 2003;3:259-268.

[8.] Masiero S, Carraro E, Celia A, et al.Prevalence of nonspecific low back pain in schoolchildren aged between 13 and 15 years. Acta Paediatr. 2008;2:212-216.

[9.] Sheir-Neiss GI, Kruse RW, Rahman T, et al. The association of backpack use and back pain in adolescents. Spine. 2003;9: 922-930.

[10.] Salminen JJ, Pentti J, Terho P. Low back pain and disability in 14-year-old schoolchildren. Acta Paediatr. 1992;12:1035-1039.

[11.] Shehab DK, Al-Jarallah KF. Nonspecific low-back pain in Kuwati children and adolescents: associated risk factors. J Adolesc Health. 2005;1:32-35.

[12.] Watson KD, Papageorgiou AC, Jones GT, et al. Low back pain in schoolchildren: occurrence and characteristics. Pain. 2002;1-2:87-92.

[13.] Balague F, Troussier B, Salminen JJ. Non-specific low back pain in children and adolescents: risk factors. Eur Spine J. 1999;6:429-438.

[14.] Cakmak A, Yucel B, Ozyalcn SN, et al. The frequency and associated factors of low back pain among a younger population in Turkey. Spine. 2004;14:1567-1572.

[15.] 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;5:459-464.

[16.] Hellstrom M, Jacobsson B, Sward L, et al. Radiologic abnormalities of the thoraco-lumbar spine in athletes. Acta Radiol. 1990;2:127-132.

[17.] Kujala UM, Taimela S, Erkintalo M, et al. Low back pain in athletes. Med Sci Sports Exerc. 1996;2:165-170.

[18.] Grimmer K, Williams M. Gender-age environmental associates of adolescent low back pain. Appl Ergon. 2000;4:343-360.

[19.] Ohlen G, Wredmark T, Spangfort E. Spinal sagittal mobility related to low-back pain in the female gymnasts. Spine. 1989;8:847-850.

[20.] Kolt GS, Kirkby RJ. Epidemiology of injuries in elite and subelite female gymnasts: a comparison of retrospective and prospective findings. Br J Sports Med. 1999;5:312-318.

[21.] Sward L, Eriksson B, Peterson L. Anthropometric characteristics, passive hip flexion, and spinal mobility in relation to back pain in athletes. Spine. 1990;5:376-382.

[22.] Sward L, Hellstrom M, Jacobsson B, et al. Disk degeneration and associated abnormalities of spine in elite gymnasts. A magnetic resonance imaging study. Spine. 1991;4:437-443.

[23.] Kolt G, Kirkby RJ, Lindner H. Coping processes in competitive gymnasts: gender differences. Percept Mot Skills. 1995;3 Pt 2:1139-1145.

[24.] Goldberg MJ. Gymnastic injuries. Orthop Clin North Am. 1980;4:717-726.

[25.] Kujala UM, Taimela S, Erkintalo M, et al. Low back pain in adolescent athletes. Med Sci Sports Exerc. 1996;2:165-170.

[26.] Tertti MO, Salminen JJ, Paajanen HE, et al. Low back pain and disk degeneration in children: a case-control MR imaging study. Radiology .1991;2:503-507.

[27.] Cupisti A, D'Alessandro C, Evangelisti I, et al. Low back pain in competitive rhytmic gymnasts. J Sports Med Phys Fitness. 2004;1:49-53.

[28.] Hall SJ. Mechanical contribution to lumbar stress in female gymnasts. Med Sci Sports Exerc. 1986;6:599-602. Auvinen J, Tammelin T, Taimela S, et al. Associations of physical activity and inactivity with low back pain in adolescents. Scand J Med Sci Sports. 2008;2:188-194.

Carla Vanti, Physical Therapist, Lecturer of Manual Therapy Sciences--University of Bologna

Margherita Gasperini, Physical Therapist--University of Bologna

Filomena Morsillo, MStat (Statistica)--Unita Operativa di Medicina del Lavoro--University of Bologna

Paolo Pillastrini Physical Therapist--Associate Professor of Rehabilitative Sciences Department of Internal Medicine - University of Bologna
Table 1: Prevalence of LBP-Profiles in gymnasts and students.

                     G1 (Gymnasts)    G2 (Students)

P1: Low-level LBP     42 (46.15%)      228 (60.0%)

P2: Medium/High         24 (26%)        137 (36%)
level LBP

TOT                    91 (100%)        375 (100%)

Table 2: Risk estimation for LBP profiles in relation
to the hours spent in a sitting position.

Activity                P1: Low-level LBP

                          G1 (Gymnasts)

                     OR       P           CI

Studying           1.042    0.867    0.641-1.694
Watching TV        0.711    0.201    0.422-1.198
Using PC           1.175    0.535    0.705-1.957
Sitting position   1.621    0.012    1.111-2.367

Activity               P1: Low-level LBP

                         G2 (Students)

                     OR       P           CI

Studying           1.062    0.533    0.879-1.283
Watching TV        1.180    0.046    1.003-1.388
Using PC           1.017    0.869       0.832
Sitting position   0.983    0.822       0.852

Activity             P2: Medium/High level LBP

                          G1 (Gymnasts)

                     OR       P           CI

Studying           1.200    0.490    0.714-2.016
Watching TV        0.582    0.089    0.312-1.086
Using PC           0.883    0.621    0.541-1.442
Sitting position   1.929    0.003    1.257-2.958

Activity             P2: Medium/High level LBP

                          G2 (Students)

                     OR       P           CI

Studying           1.066    0.500    0.885-1.284
Watching TV        1.045    0.573    0.895-1.220
Using PC           1.067    0.509    0.879-1.296
Sitting position   1.010    0.890    0.877-1.163

Table 3: Distribution of ROM groups in LBP Profiles.

       Groups                 PROFILE 1

                        YES (n=42)   NO (n=49)

ROM1=poor ROM            16 (42%)     22 (58%)
ROM2=moderate ROM        7 (28%)      18 (72%)
ROM3=large ROM           19 (68%)     9 (32%)

Risk estimation *             PROFILE 1

ROM1=poor ROM             1.87 (0.63 - 5.53)
ROM2=moderate ROM **             1
ROM3=large ROM            5.43 (1.66 - 17.6)

       Groups                 PROFILE 2

                        YES (n=24)   NO (n=67)

ROM1=poor ROM            9 (24%)      29 (76%)
ROM2=moderate ROM        5 (20%)      20 (80%)
ROM3=large ROM           10 (36%)     18 (64%)

Risk estimation *            PROFILE 2

ROM1=poor ROM              1.24 (0.36-4.25)
ROM2=moderate ROM **              1
ROM3=large ROM             2.22 (0.64-7.74)

* Unadjusted risk ratio (equal Odds Ratio)

** In this table the odds ratio is given for each group of ROM
using moderate ROM as the reference group. The reference group
is indicated by a value of 1 for the odds ratio.
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.