PEDIATRICS Vol. 121 No. 1 January 2008, pp. 9-14 (doi:10.1542/peds.2006-3673)
ARTICLE |
Estimating the Effectiveness of Screening for Scoliosis: A Case-Control Study
a Departments of Public Health
b Orthopedics, Erasmus Medical Center, University Medical Center Rotterdam, Rotterdam, Netherlands
c Youth Health Care Section, Municipal Health Service, Geleen, Netherlands
d Municipal Health Service, Rotterdam, Netherlands
e Youth Health Care Section, Municipal Health Service, Assen, Netherlands
f Department of Social Medicine, University Medical Center Groningen, Groningen, Netherlands
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
OBJECTIVE. The aim of this study was to test the hypothesis that screening for scoliosis is effective in reducing the need for surgical treatment.
METHODS. The study was a case-control study. A total of 125 consecutive patients who were treated surgically for idiopathic scoliosis between January 2001 and October 2004 and who were born on or after January 1, 1984, were invited; 108 agreed to participate. A total of 216 control subjects were selected randomly and anonymously, matched with respect to age and gender. For 279 adolescents, exact screening exposure and outcomes could be analyzed. Case subjects were recruited from 4 university and 6 nonuniversity Dutch hospitals; control subjects were recruited from all 37 municipal health services in the Netherlands.
RESULTS. Screen-detected patients received diagnoses at a significantly younger age than did otherwise-detected patients (10.8 ± 2.6 vs 13.4 ± 1.7 years). In total, 32.8% of the surgically treated patients had been screened between 11 and 14 years of age, compared with 43.4% of the control subjects. The odds ratio for being exposed to screening was 0.64. In total, 28% of the patients were diagnosed as having scoliosis before 11 years of age.
CONCLUSIONS. Our results showed no evidence that screening for scoliosis reduced the need for surgery. Abolishing screening seems justified, especially because the effectiveness of early treatment with bracing is still strongly debated. A randomized, controlled trial on the effectiveness of treating patients with idiopathic scoliosis with bracing is urgently needed.
Key Words: scoliosis • screening • case-control study • effectiveness
Abbreviations: IS—idiopathic scoliosis • RCT—randomized, controlled trial • OR—odds ratio • CI—confidence interval • MHS—municipal health service
Idiopathic scoliosis (IS) is defined as lateral curvature of the spine (minimal Cobb angle of 10°), of unknown origin, with concomitant vertebral rotation.1 Screening for IS was introduced in the United States and many other countries in the 1970s.2 Screening aims at detecting patients in an early stage of the clinical course, to allow brace treatment with the aim of preventing further progression and the need for surgical treatment.3 In the Netherlands, an estimated 80% of children are screened for IS at least once.
To date, however, the effectiveness of such screening (and early treatment) has not been established sufficiently, because of a lack of randomized, controlled trials (RCTs).4 Some studies concluded that screening for scoliosis is effective,2,5–7 whereas others doubted the effectiveness or even considered such screening to be unethical.8–10 The US Preventive Services Task Force recommends against the routine screening of asymptomatic adolescents for IS, because evidence has shown that the balance of benefits (a few) and harms (more than a few) is negative.11 Another important issue is that, because the cause of this form of scoliosis is unknown, the current screening test and the early treatment may not be the most appropriate ones.
Earlier, we found that screen-detected patients were detected at an earlier stage of the clinical course and that screen-identified patients had a 73% lower chance of needing surgery.12 However, overtreatment bias and length-biased sampling, leading to overestimation of the effect of screening, could be serious problems in such a study.12,13
The present study investigated the effect of screening for scoliosis on reduction of the need for surgery. From a methodologic viewpoint, a RCT is the best design to establish the effectiveness of screening for scoliosis. Because IS is not a common condition, however, such a design would require a very large study population to gain sufficient power, which makes a RCT less feasible.14 The second-best design is a case-control study,15 which we performed with individual data on exposure to screening.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Design
This was a case-control study, in which the case group consisted of patients with IS who were treated surgically and the control group consisted of a random sample of Dutch youths. Control subjects were matched to case subjects with respect to age and gender. Matching with respect to age was performed to provide equal opportunities for case subjects and control subjects to be exposed to screening in the past. Because IS occurs more often in girls than in boys, matching with respect to gender was applied.
Study Population
Case Subjects
In the Netherlands, 
50 patients with IS each year need surgical treatment; >90% of these operations take place in 11 hospitals. Orthopedic surgeons in 10 hospitals (4 university hospitals and 6 nonuniversity hospitals) in the Netherlands where patients with IS are treated conservatively and surgically were willing to cooperate in this study. They were requested to report all patients who were treated surgically for IS between January 2001 and October 2004 and who were born on or after January 1, 1984. These patients (N = 125) were invited to participate in this study; 108 patients (86%) and their parents (if the patient was <16 years of age) gave informed consent for participation.
Control Subjects
The control subjects were matched to the case subjects with respect to age and gender; all control subjects were selected randomly from the source population. For each case subject, 2 control subjects were sampled. Data on the control subjects were collected from the databases of all municipal health services (MHSs) in the Netherlands (N = 37), which include almost all youths. Each time a case subject was included in the study, we selected randomly 2 MHSs, weighted with respect to the number of youths registered in each MHS. We requested the MHS to select a control subject whose date of birth and gender matched those of the case subject. Because it was possible that a MHS would have >1 match in its database, we requested that the MHS start searching family names from a certain letter of the alphabet until the first match was found; these letters were distributed randomly among the control subjects. If there was no match in the database, then the MHS was requested to search for a control subject who was born 1 day after the case subject. If that still did not lead to a match, then the MHS had to search for a control subject born 1 day before the case subject and then, if necessary, 2 days after the case subject was born, and so forth, until a match was found. Such adjustments needed to be made for only 9 control subjects (the maximal difference that was needed was 5 days). We emphasized that the gender of the control subject had to be the same as that of the case subject. Because the control subjects were kept anonymous for the researchers, informed consent from the control subjects was not necessary. The study was conducted according to the principles of the Declaration of Helsinki. Under Dutch law, observational health surveys are exempt from needing approval from a medical ethics committee.
Variables and Measurements
Case Subjects and Control Subjects
School physicians received a questionnaire that they were requested to complete. Data on being exposed to screening, ages at screenings, and whether Adam's forward bending test was performed were collected from youth health care files. Being exposed to screening was defined as being examined for scoliosis during a periodic medical examination or a "single scoliosis screening" with at least Adam's forward bending test.3 All examinations that were performed at the specific request of someone (for example, a gym teacher or parent) and that were not incorporated into the usual screening program were not considered exposure to screening.
Case Subjects
Data on the Cobb angle, curve type, brace treatment, age at detection otherwise (ie, not through screening), and age at diagnosis were collected by using medical files and telephone interviews with participating patients being treated for scoliosis. If the school physician had detected the scoliosis, then the date of detection was retrieved from the youth health care file. The orthopedic surgeon established the diagnosis of IS.
Statistical Analyses
Power Calculations
In total, 102 case subjects and 204 control subjects were needed to obtain a probability of 80% for establishing a 50% reduction in surgery for patients with IS, with 
= .05.
Statistics
Because some data were skewed and some subgroups were smaller than n = 30, we used the Mann-Whitney U test to evaluate significant differences between screen-detected and otherwise-detected patients with respect to median Cobb angles, age at detection, age at diagnosis, time period between detection and diagnosis, and age at surgery.
Only screenings that were performed before IS was diagnosed by an orthopedic surgeon counted as exposure to screening. For the control subjects, only screenings that were performed before IS was diagnosed for the matched case subjects were valid.16 The odds ratios (ORs) and their 95% confidence intervals (CIs) for being exposed to screening were calculated by using binary logistic regression analysis. To account for matching with respect to gender, gender was added as a categorical covariate. We did not use matched analysis (eg, conditional logistic regression analysis) because the matching factors did not influence the exposure measure such that they would lead to bias.
First, the OR for being ever/never screened before diagnosis was calculated. We then calculated the OR for being screened or not between 11 and 14 years of age (ie, the ages at which screening is recommended in the Netherlands). The latter was performed only for the case subjects and matched control subjects who were still eligible for screening (ie, scoliosis had not yet been diagnosed).
Estimation of Costs
Costs for screening for scoliosis were estimated on the basis of the cost of activities model for a MHS.4 This model considers that 42% of the screenings are part of a school physician's consultation, 20% a single screening performed by school physicians, 31% part of a school nurse's consultation, and 7% a single screening performed by school nurses. Estimations of the costs of IS surgery were based on Dutch health care fees.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
For 7 selected case subjects and 1 control subject, the youth health care files were not retrievable. For 1 selected case subject and 2 control subjects, we did not receive a completed questionnaire; for another 5 case subjects, the age of diagnosis was missing. One case was diagnosed before the age of 5 years. Because this case might have represented some sort of very early-onset scoliosis, we deleted this case subject and the matched control subjects from all analyses.
Table 1 shows the characteristics of the patients with IS that were treated surgically. The ratio of girls to boys was 4:1. Screen-detected patients had significantly smaller Cobb angles at diagnosis, compared with otherwise-detected patients (P < .01). After surgery, Cobb angles did not differ significantly between screen-detected and otherwise-detected patients. Approximately one half of the patients were treated with a brace before surgery. Although there was no significant difference in duration of brace treatment between screen-detected and otherwise-detected patients, screen-detected patients had an almost threefold greater chance of being treated with a brace before surgery (OR: 3.1; 95% CI: 1.3–7.0), compared with otherwise-detected patients. On average, brace treatment lasted for 2.5 years.
|
Table 2 shows mean and median ages at detection, diagnosis, and surgery and the time period between detection and diagnosis. Screen-detected patients were
2.5 years younger at detection (Fig 1) and at diagnosis. On average, there were >12 months between detection and diagnosis for screen-detected patients and 6 months for otherwise-detected patients. In total, 55% of the screen-detected patients and 66% of the otherwise-detected patients were diagnosed by an orthopedic surgeon within 4 months after detection. In Fig 1, the 3 peaks in the screen-detected curve correspond to the ages at which school physicians perform periodic medical examinations in the Netherlands.
|
|
The majority (74%) of children had been exposed to screening at least once (Table 3). The proportion of case subjects exposed to screening was slightly greater (80.5%) than the proportion of control subjects exposed to screening (74.0%). The OR for being exposed to screening was 1.44 (95% CI: 0.77–2.68; P = .25).
|
The OR for being screened at the age of 11, 12, 13, or 14 years was 0.64 (95% CI: 0.34–1.19; P = .16) (Table 4). The proportion of case subjects exposed to screening was smaller than the proportion of control subjects exposed to screening, but screening did not reduce the chance of surgery significantly. It should be noted that 30 case subjects (28%) were diagnosed as having IS before they were 11 years of age.
|
Costs to screen 80% of 1 Dutch birth cohort (
200000) were estimated at 3 million euros. Costs for 1 scoliosis operation were estimated at 11000 euros. Total surgical costs were estimated at 550000 euros, given 50 operations in 1 year. |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The results of this study did not show a significant reduction in the need for scoliosis surgery attributable to screening. Patients detected through screening were significantly younger at diagnosis than were patients who were detected otherwise. This means that the screen-detected patients had additional years of concern about the disease and they had a greater chance of brace treatment but without better final outcomes. With detailed data for 200 case subjects and control subjects, we had 80% power to show a 59% reduction in scoliosis operations. Our results confirm the conclusion of Wiegersma et al,10 who also reported that screening for scoliosis did not reduce the need for surgery.
Case-control studies are susceptible to different kinds of bias. In this study, all patients already suffered from the serious outcome that screening is supposed to prevent; therefore, ascertainment bias is not a serious problem. Because the control subjects were kept anonymous to the researchers, there was no nonresponse in the control group that could lead to nonresponse bias. The response rate for participation among the patients was high (86%). If we had considered the nonresponders as not being exposed to screening, then we would have found a positive effect of screening. However, this is a highly unlikely scenario, because we have no indications that the nonresponders would be different from the responders. Furthermore, being exposed to screening for scoliosis depends more on whether the youth health care department offers the screening for scoliosis in general than on the characteristics of the child or the parents; therefore, we do not expect selection bias to influence the results substantially. Recall bias could be a problem for otherwise-detected case subjects in relation to data on the detection date; patients might have underestimated the time between detection and diagnosis. This could also apply to the age at diagnosis if the medical chart was incomplete.
Our results do not show a significant reduction in the need for surgery attributable to screening for scoliosis among children 11 to 14 years of age (the ages at which screening is usually recommended). If we assume that the OR of 0.64 is the true size of the effect, then the costs of keeping 1 patient from the need for surgery are estimated at (at least) 130000 euros, and 5800 children would need to be screened. These are relatively high costs and involve considerable effort, given that severe scoliosis is neither common nor fatal. Furthermore, screening identifies some children who ultimately receive treatment but involves referral of many more who do not.17 Therefore, these costs are an underestimation of the real costs, because they exclude the costs of visits to general practitioners and orthopedic surgeons and of radiographs attributable to false-positive results. Yawn and Yawn18 calculated that case finding costs for screening were $10836 per child treated (conservatively or surgically) for scoliosis.
In our previous study, we found that 2 prerequisites for an effective screening program were met, that is, earlier detection and less surgery in the screen-detected group.12 However, overtreatment bias and length-biased sampling could not be ruled out. In the present study, we also found that screen-detected patients were diagnosed in an earlier stage and had a greater chance of being treated with a brace. However, we could not prove that exposure to screening led to less surgery. One explanation for this could be that screening for scoliosis may lead to overtreatment with a brace.11 Patients with a relatively small Cobb angle are more likely to be detected through screening than otherwise (eg, by themselves or by their parents). Some of these patients are treated with a brace, whereas they would not have visited an orthopedic surgeon and received or needed treatment at all if they had not been identified through screening.
The relatively low sensitivity (55%) of the screening program12 could perhaps explain why we did not find a beneficial effect. Furthermore, low levels of compliance with brace treatment could lead to ineffective treatment with a brace, which could result in more operations. More importantly, it is still unclear whether early intervention with a brace is an effective strategy in preventing surgery for patients with IS. Some authors consider a brace effective,19,20 whereas others conclude that the effectiveness of bracing is doubtful or they recommend a RCT on brace treatment.4,8,21,22 If we had found convincing evidence for the beneficial effects of screening for scoliosis, then this would have implied that early treatment with a brace is effective. Because we did not find convincing evidence that screening is effective, we need to determine whether early bracing is effective by means of a RCT; such a trial started in the Netherlands in 2006.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We think that abolishing screening for scoliosis seems justified, because of the lack of evidence that screening and/or early treatment with bracing is beneficial. For now, instead of screening large numbers of asymptomatic children, the appropriate approach would be to look at a child's back when there are indications that something is wrong. Such children should be examined and, if necessary, referred to a specialist. If a RCT on brace treatment establishes that bracing is effective, then it will be worthwhile to determine which children could benefit from a screening program.
|
ACKNOWLEDGMENTS |
|---|
This study was funded by the Netherlands Organization for Health Research and Development (grant 2200.0127). The funding source had no involvement in our work.
Members of the Netherlands Evaluation study on Screening for scoliosis (NESCIO) group were as follows: Henk D. Been, MD, PhD, Department of Orthopedics, Academic Medical Centre (Amsterdam, Netherlands); L. Napoleon J. E. M. Coene, MD, PhD, Department of Orthopedics, HaGa Hospital (The Hague, Netherlands); Bert de Gruijter, MD, PhD, Department of Orthopedics, Medical Center Alkmaar (Alkmaar, Netherlands); Luuk WL de Klerk, MD, PhD, Department of Orthopedics, Erasmus Medical Center Rotterdam (Rotterdam, Netherlands); Marinus de Kleuver, MD, PhD, Department of Orthopedics, Sint Maartenskliniek (Nijmegen, Netherlands); Hein J. A. Kruls, MD, PhD, Department of Orthopedics, Amphia Ziekenhuis (Breda, Netherlands); Piet J. M. van Loon, MD, Department of Orthopedics, Rijnstate Hospital (Arnhem, Netherlands); Frank de Nies, MD, Department of Orthopedics, Onze Lieve Vrouwe Gasthuis (Amsterdam, Netherlands); Hans E. H. Pruijs, MD, PhD, Department of Orthopedics, University Medical Center Utrecht (Utrecht, Netherlands); Lodewijk W. van Rhijn, MD, PhD, Department of Orthopedics, University Hospital Maastricht (Maastricht, Netherlands).
We thank Gerrit Draisma, MSc, Caspar Looman, MSc, and Rob Boer, PhD (Department of Public Health, Erasmus Medical Center) for their statistical advice. We also thank all school physicians and nurses of the MHSs who participated in this study by providing us with the necessary data on screening history (Gemeente Den Haag, Dienst Onderwijs, Cultuur en Welzijn (OCW), Gemeentelijke Geneeskunolige en Gezondheidsdienst Utrecht, Gemeentelijke Gezondheidsdienst Amstelland-de Meerlanden, Gemeentelijke Gezondheidsdienst Amsterdam, Gemeentelijke Gezondheidsdienst Drenthe, Gemeentelijke Gezondheidsdienst Eemland, Gemeentelijke Gezondheidsdienst Eindhoven, Gemeentelijke Gezondheidsdienst Fryslân, Gemeentelijke Gezondheidsdienst Gelre-Ijssel, Gemeentelijke Gezondheidsdienst Gooi and Vechtstreek, Gemeentelijke Gezondheidsdienst Hart voor Brabant, Gemeentelijke Gezondheidsdienst Hollands Midden vestiging. Gouda, Gemeentelijke Gezondheidsdienst Hollands Midden vest. Leiden, Gemeentelijke Gezondheidsdienst Kop van Noord-Holland, Gemeentelijke Gezondheidsdienst Midden-Nederland, Gemeentelijke Gezondheidsdienst Nieuwe Waterweg Noord, Gemeentelijke Gezondheidsdienst Noord-en Midden Limburg, Gemeentelijke Gezondheidsdienst Noord Kennemerland, Gemeentelijke Gezondheidsdienst Regio Ijssel-Vecht, Gemeentelijke Gezondheidsdienst Regio Nijmegen, Gemeentelijke Gezondheidsdienst Noord Veluwe, Gemeentelijke Gezondheidsdienst Regio Twente, Gemeentelijke Gezondheidsdienst Rivierenland, Gemeentelijke Gezondheidsdienst Rotterdam en omgeving, Gemeentelijke Gezondheidsdienst West-Brabant, Gemeentelijke Gezondheidsdienst Westfriesland, Gemeentelijke Gezondheidsdienst Zaanstreek-Waterland, Gemeentelijke Gezondheidsdienst Zeeland, Gemeentelijke Gezondheidsdienst Zuid-Holland West, Gemeentelijke Gezondheidsdienst Zuid-Holland Zuid, Gemeentelijke Gezondheidsdienst Zuid Limburg, Gemeentelijke Gezondheidsdienst Zuidhollandse Eilanden, Gemeentelijke Gezondheidsdienst Zuidoost Brabant, Hulpverlening Gelderland Midden, Hulpverleningsdienst Flevoland, Hulpverleningsdienst Groningen, and Hulpverleningsdienst Kennemerland). Last but not least, we are very grateful to the patients and their parents who participated in this study.
|
FOOTNOTES |
|---|
Accepted Jun 29, 2007.
Address correspondence to Eveline Bunge, MSc, Department of Public Health, Erasmus Medical Center, University Medical Center Rotterdam, PO Box 2040, 3000 CA Rotterdam, Netherlands. E-mail: e.bunge{at}erasmusmc.nl
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
- Campbell W, Canale S, Daugherty K, Crenshaw A Jr. Scoliosis and kyphosis. In: Canale ST, ed. Campbell's Operative Orthopaedics. St Louis, MO: Mosby; 2003:1751–1984
- Lonstein JE, Bjorklund S, Wanninger MH, Nelson RP. Voluntary school screening for scoliosis in Minnesota.
J Bone Joint Surg Am. 1982;64
:481
–488
[Abstract/Free Full Text] - US Preventive Services Task Force. Screening for adolescent idiopathic scoliosis.
JAMA. 1993;269
:2667
–2672
[Abstract/Free Full Text] - Korfage IJ, Juttmann RE, Das BV, Diepstraten AF, Hazebroek-Kampschreur AA, van der Maas PJ. Idiopathic scoliosis in adolescents; an inventory into the possibilities of studying the efficacy of screening and treatment [in Dutch]. Ned Tijdschr Geneeskd. 2002;146 :1228 –1233[Medline]
- Bunnell WP. Selective screening for scoliosis. Clin Orthop Relat Res. 2005;434 :40 –45[Medline]
- Soucacos PN, Soucacos PK, Zacharis KC, Beris AE, Xenakis TA. School-screening for scoliosis: a prospective epidemiological study in northwestern and central Greece.
J Bone Joint Surg Am. 1997;79
:1498
–1503
[Abstract/Free Full Text] - Montgomery F, Willner S. Screening for idiopathic scoliosis: comparison of 90 cases shows less surgery by early diagnosis. Acta Orthop Scand. 1993;64 :456 –458[Web of Science][Medline]
- Dickson RA, Weinstein SL. Bracing (and screening): yes or no? J Bone Joint Surg Br. 1999;81 :193 –198[CrossRef][Medline]
- Goldberg CJ, Dowling FE, Fogarty EE, Moore DP. School scoliosis screening and the United States Preventive Services Task Force: an examination of long-term results. Spine. 1995;20 :1368 –1374[Web of Science][Medline]
- Wiegersma PA, Hofman A, Zielhuis GA. The effect of school screening on surgery for adolescent idiopathic scoliosis.
Eur J Public Health. 1998;8
:237
–240
[Abstract/Free Full Text] - US Preventive Services Task Force. Screening for Idiopathic Scoliosis in Adolescents: Recommendation Statement. Rockville, MD: Agency for Healthcare Research and Quality; 2004
- Bunge EM, Juttmann RE, de Koning HJ, Steering Committee of the NESCIO Group. Screening for scoliosis: do we have indications for effectiveness? J Med Screen. 2006;13 :29 –33[CrossRef][Web of Science][Medline]
- Juttmann RE, Hess J, van Oortmarssen GJ, van der Maas PJ. Patient follow up screening evaluations: examples with regard to congenital hip dislocation and congenital heart disease.
J Epidemiol Community Health. 2001;55
:126
–131
[Abstract/Free Full Text] - Black N. Why we need observational studies to evaluate the effectiveness of health care.
BMJ. 1996;312
:1215
–1218
[Free Full Text] - Weiss NS. Application of the case-control method in the evaluation of screening.
Epidemiol Rev. 1994;16
:102
–108
[Free Full Text] - Connor RJ, Boer R, Prorok PC, Weed DL. Investigation of design and bias issues in case-control studies of cancer screening using microsimulation.
Am J Epidemiol. 2000;151
:991
–998
[Abstract/Free Full Text] - Yawn BP, Yawn RA, Hodge D, et al. A population-based study of school scoliosis screening.
JAMA. 1999;282
:1427
–1432
[Abstract/Free Full Text] - Yawn BP, Yawn RA. The estimated cost of school scoliosis screening. Spine. 2000;25 :2387 –2391[CrossRef][Web of Science][Medline]
- Rowe DE, Bernstein SM, Riddick MF, Adler F, Emans JB, Gardner-Bonneau D. A meta-analysis of the efficacy of non-operative treatments for idiopathic scoliosis.
J Bone Joint Surg Am. 1997;79
:664
–674
[Abstract/Free Full Text] - Nachemson AL, Peterson LE. Effectiveness of treatment with a brace in girls who have adolescent idiopathic scoliosis: a prospective, controlled study based on data from the Brace Study of the Scoliosis Research Society.
J Bone Joint Surg Am. 1995;77
:815
–822
[Abstract/Free Full Text] - Focarile FA, Bonaldi A, Giarolo MA, Ferrari U, Zilioli E, Ottaviani C. Effectiveness of nonsurgical treatment for idiopathic scoliosis: overview of available evidence. Spine. 1991;16 :395 –401[CrossRef][Web of Science][Medline]
- Goldberg CJ, Dowling FE, Hall JE, Emans JB. A statistical comparison between natural history of idiopathic scoliosis and brace treatment in skeletally immature adolescent girls. Spine. 1993;18 :902 –908[Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2008 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  B. S. Richards, J. H. Beaty, G. H. Thompson, and R. B. Willis Estimating the Effectiveness of Screening for Scoliosis Pediatrics, June 1, 2008; 121(6): 1296 - 1297. [Full Text] [PDF]
|
|
|
|
|
|
E. M. Bunge and H. J. de Koning Estimating the Effectiveness of Screening for Scoliosis: In Reply Pediatrics, June 1, 2008; 121(6): 1297 - 1298. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
