J Manipulative Physiol Ther 2003 (Jan); 26 (1): 1–8 ~ FULL TEXT
Hayden JA, Mior SA, Verhoef MJ
Institute for Work and Health, 481 University Avenue, 8th Floor, Toronto M5G 2E9, Ontario, Canada
BACKGROUND: Recent epidemiologic studies have estimated that the lifetime prevalence of low back pain (LBP) in children is approximately 50%, with almost 15% of children experiencing frequent or continual pain. A literature search revealed no published studies addressing conservative treatment of childhood LBP.
OBJECTIVE: To describe chiropractic management of LBP in patients between the ages of 4 and 18 years, as well as outcomes and factors associated with the outcomes.
METHODS: Prospective cohort study of consecutive pediatric patients with LBP seeing randomly selected chiropractors within the cities of Calgary, Alberta, and Toronto, Ontario, Canada. Follow-up data collection included the type and extent of treatment rendered and its outcome, which was measured with a 5-point subjective rating scale and a self-report pediatric visual analogue scale.
RESULTS: Fifteen chiropractors provided data on 54 consecutive pediatric patients with LBP. The average age of the patients was 13.1 years, 57% were male, 61% were acute, with 47% attributing onset to a traumatic event (most commonly sports-related); 24% reported an episode duration of greater than 3 months. Almost 90% of cases presented with uncomplicated mechanical LBP, most frequently diagnosed as lumbar facet dysfunction or subluxation. Patients were managed with manipulation, with a minority (7.7%) receiving some form of active management. "Important" improvement was seen in 62% and 87% on the visual analogue and subjective scales, respectively, within a 6-week course of management (Kaplan-Meier survival analysis). Patients with chronic LBP were less likely to respond within the median number of treatments (relative risk = 2.1).
CONCLUSIONS: Patients responded favorably to chiropractic management, and there were no reported complications. Future investigations should establish the natural history and compare chiropractic management with other forms of treatment to gain knowledge about the effectiveness of chiropractic in managing pediatric LBP.
From the Full-Text Article:
This study provides information for the sparse chiropractic literature on the treatment of pediatric LBP. We collected data prospectively using standard and validated data collection forms; extensive information was collected on the history, examination, and follow-up of the subjects, including a self-report pediatric pain scale. 
Pediatric patient demographics provided interesting data regarding characteristics of the presenting back pain. The patients’ mean age of 13 years is consistent with that reported in the literature for first episode of pain [8, 9, 36] ; however, almost 40% of patients reported a prior history of
LBP, suggesting that this appears to be a common and recurring problem in this study sample. As in previously reported studies, [7, 8, 37, 38] it was not surprising to find that the onset of pain in almost 50% of cases was attributed to a traumatic event, most commonly sports-related. This appears
to account for the acute presentation of the majority of cases in our study. Notably, however, about 24% of the cases in this study had an episode duration of greater than 3 months, and almost 32% were “getting worse.” This corresponds to the work of Taimela et al,  who reported chronic or recurrent LBP in approximately 30% of cases in their cross-sectional, school-based study of children 7 to 16 years old. In addition, questions regarding the “natural history of adult LBP” have recently been raised by Croft et al  and may be relevant also to the pediatric population, emphasizing that uncomplicated mechanical LBP may be recurrent in nature and not be resolved within the first month. Almost 90% of cases in this study presented with uncomplicated mechanical LBP, most frequently diagnosed as a lumbar facet dysfunction or subluxation. These findings contrast the traditional opinion, which holds that when children present to practitioners for this complaint, some specific or serious cause (eg, severe spondylolisthesis or infectious, inflammatory, or neoplastic disorder) is found in nearly half of the cases. [28, 41, 42] This discrepancy is likely due to the population sampled: medical and orthopedic patients versus chiropractic patients. This suggests that the findings of this study are likely not generalizable to nonchiropractic patient populations. However, the sample may be
similar to the general population, because epidemiologic studies have suggested that the majority of LBP in the general population is mechanical or nonspecific in nature.  One case (1.8%) of disk herniation with neurologic findingswas documented in our study population. This is consistent
with previous reports that about 3% of patients with disk herniation presenting for surgery were under the age of 20 years. 
Almost all the patients received manipulation as the principal intervention. This is consistent with other reports describing chiropractic management of musculoskeletal conditions,  and is not surprising when we consider the diagnoses reported and positive spinal palpation findings. It should be noted, however, that there is little literature supporting the validity and reliability of motion palpation. We were surprised by the infrequent use in this study population of active management strategies, recommended to only 7.7% of the patients. In consideration of the emphasis of activity and exercise in published guidelines and reviews, [19-21] and past reports,  we had expected this management strategy to be more prevalent. Also, this may be
especially important given the finding that chronic patients were less likely to respond within the median number of follow-up visits. These findings suggest that there is a need to educate chiropractors further regarding active patient management.
We used 2 constructs to measure clinically important patient improvement and found that patients respond favorably to chiropractic treatment. Over a typical course of management, which would be approximately 4 to 6 weeks, 55% to 62% of patients demonstrated important improvement by the stringent criteria of minimally clinically important change on the VAS pain scale. It is important to note, however, that the outcome measure used can affect the reported outcome; using the subjective criterion “much improved,” 82% to 87% of patients demonstrated improvement over this 4- to 6-week period.  This difference between the VAS criterion of important improvement and the subjective measure “much improved” may be more reflective of patients’ interpretation of their overall response to management, including patient satisfaction with the practitioner and treatment. However, this subjective measure may also overestimate the true treatment effect. It will be important for future studies to include appropriate measures to assess physical functioning and general well-being.
Several limitations of our methodology must be considered. The major limitations of this study are related to the small size and observational design of the investigation. This study cannot be used to establish cause and effect, because there was no natural history comparison group. However, it does suggest that pediatric patients with LBP do respond positively to chiropractic treatment. The study subject selection is another potential source of limitation. Although subject practitioners were randomly selected, the practitioner inclusion criteria (more than 2 pediatric patients per week) and the limitation to 2 urban geographic areas limit the generalizability of the study results. Lack of follow- up was also a limitation.
It is uncertain whether participation of the noncompliant chiropractors would have significantly altered our findings. The only difference in basic demographics between the compliant and noncompliant groups was the school of graduation, with a greater proportion of the compliant group having graduated from the CMCC. This may have contributed to the 2 main reasons provided by the noncompliant group for not submitting cases, namely, that reporting only LBP cases was too limiting, and that they saw children primarily for visceral or postural-related conditions.
This prospective study is one of the first to document pediatric patients with LBP and their outcome to chiropractic care. Our findings suggest that the majority of the patients present with localized LBP of acute onset. The vast majority of the patients were managed with manipulation/adjustment, but relatively few were provided with active care strategies. Patients responded favorably to chiropractic management, and there were no reported complications. Future investigations should establish the natural history and compare chiropractic management to other forms of treatment to gain knowledge about the effectiveness of chiropractic in managing pediatric LBP.