J Manipulative Physiol Ther 2001 (July); 24 (6): 369–377
Gert Bronfort, DC, PhD, Roni L. Evans, DC, Paul Kubic, MD, PhD, Patty Filkin
Wolfe-Harris Center for Clinical Studies, Northwestern Health Sciences University, Bloomington, MN 55431, USA.
Read the Criticism of this study
Objectives: The first objective was to determine if chiropractic spinal manipulative therapy (SMT) in addition to optimal medical management resulted in clinically important changes in asthma-related outcomes in children. The second objective was to assess the feasibility of conducting a full-scale, randomized clinical trial in terms of recruitment, evaluation, treatment, and ability to deliver a sham SMT procedure.
Study Design: Prospective clinical case series combined with an observer-blinded, pilot randomized clinical trial with a 1-year follow-up period.
Setting: Primary contact, college outpatient clinic, and a pediatric hospital.
Patients: A total of 36 patients aged 6 to 17 years with mild and moderate persistent asthma were admitted to the study.
Outcome Measures: Pulmonary function tests; patient- and parent- or guardian-rated asthma-specific quality of life, asthma severity, and improvement; am and pm peak expiratory flow rates; and diary-based day and nighttime symptoms.
Interventions: Twenty chiropractic treatment sessions were scheduled during the 3-month intervention phase. Patients were randomly assigned to receive either active SMT or sham SMT in addition to their standardized ongoing medical management.
Results: It is possible to blind the participants to the nature of the SMT intervention, and a full-scale trial with the described design is feasible to conduct. At the end of the 12-week intervention phase, objective lung function tests and patient-rated day and nighttime symptoms based on diary recordings showed little or no change. Of the patient-rated measures, a reduction of approximately 20% in 2 bronchodilator use was seen (P = .10). The quality of life scores improved by 10% to 28% (P < .01), with the activity scale showing the most change. Asthma severity ratings showed a reduction of 39% (P < .001), and there was an overall improvement rating corresponding to 50% to 75%. The pulmonologist-rated improvement was small. Similarly, the improvements in parent- or guardian-rated outcomes were mostly small and not statistically significant. The changes in patient-rated severity and the improvement rating remained unchanged at 12-month posttreatment follow-up as assessed by a brief postal questionnaire.
Conclusion: After 3 months of combining chiropractic SMT with optimal medical management for pediatric asthma, the children rated their quality of life substantially higher and their asthma severity substantially lower. These improvements were maintained at the 1-year follow-up assessment. There were no important changes in lung function or hyperresponsiveness at any time. The observed improvements are unlikely as a result of the specific effects of chiropractic SMT alone, but other aspects of the clinical encounter that should not be dismissed readily. Further research is needed to assess which components of the chiropractic encounter are responsible for important improvements in patient-oriented outcomes so that they may be incorporated into the care of all patients with asthma.
From the Full-Text Article:
The Medical Expenditure Panel Survey estimated that almost 7% of the United States population sought unconventional health care in addition to conventional medical care in 1996.  The most common of the unconventional therapies was chiropractic care. Parents frequently seek care for their asthmatic children from chiropractors; however, there has been little scientific evidence to support such practices. Our study was an initial step in evaluating the scientific evidence of chiropractic spinal manipulation for children with asthma.
We considered it important to perform a pilot study combined with a prospective case series before embarking on a full-scale trial. First, if clinically important changes were not observed prospectively in either lung function, asthma severity, day and nighttime symptoms, or asthma specific quality of life, we had decided a priori not to conduct a full-scale clinical trial with this study design. Second, we believed it was necessary to determine the study feasibility before undertaking a costly and time-consuming full-scale, randomized clinical trial. Could patients be recruited in sufficient numbers to ensure adequate statistical power? What were the most cost-efficient methods of recruitment? Would patients and providers comply with study protocols? Could a sham SMT procedure be effectively delivered? We did establish that it is feasible to conduct a full-scale trial, although recruitment was slow and difficult. Patients and providers complied well with our protocols, and it appeared that patients and guardians were successfully blinded to the chiropractic treatment and had similar experiences of overall satisfaction regardless of group allocation.
The prospective case series part of this study demonstrated that after 12 weeks of SMT combined with optimal medical management, there were no clinically important changes in pulmonary lung function (PEFRs, FEV1, forced expiratory flow 25% to 75%, and hyperresponsiveness), patient-rated day and nighttime symptoms, and parent-/guardian-rated assessment of the child's quality of life and asthma severity. However, clinically important changes were found in patient-rated quality of life (particularly the activity domain) and patient-rated asthma severity and improvement. The discrepancy between child and parent/guardian ratings is consistent with those reported by other investigators. 
A study of similar design with a larger sample size was recently reported by Balon et al.  When we compare our results with the Balon et al trial, we note that similar and clinically important changes in asthma-specific quality of life and severity were found in both studies in the active SMT groups. However, the Balon et al study showed that these changes occurred in the sham SMT group as well. Although the improvements tended to be greater in the active group in most of the quality of life domains, they found no clinically important or statistically significant differences between active and sham SMT. There were no clinically important changes in lung function and airway hyperresponsiveness in either study. Patients in both studies had either chronic mild or moderate persistent asthma, but because they were optimally medically managed, their asthma was “under control.” In terms of lung function, there was therefore not much room for improvement, although we would expect a reduction in ß2-agonist use to accompany any reduction in patient-rated asthma severity. There was also a similar decline in ß2-agonist use in both studies. Again, this reduction in the Balon et al  study was of equal magnitude in both the active and the sham SMT groups.
What then are likely explanations for the patient-rated improvements in quality of life and patient-rated asthma severity? Recent research has shown that physical treatments such as massage appear to be beneficial in the management of children with pulmonary dysfunction and chronic asthma.  An RCT by Field et al  found that children who received 1 month of daily massage therapy by their parents showed decreased behavioral anxiety and increased cortisol levels. Thus it is possible that the physical contact involved in the spinal manipulation and the accompanying soft tissue palpation and massage used in the chiropractic studies may explain some of the benefits observed in our study.
Patient education is another important part of the successful management of asthma. It is important for patients to understand their asthma, recognize its triggers, and learn to practice necessary management skills. Family support is essential in their efforts. It has been shown that patients and parents/guardians are better able to focus on clinicians' recommendations after major concerns and fears have been addressed. [26, 27] Sometimes psychosocial dysfunction in the family may have a negative impact on the child with asthma. [27, 28] A recent systematic review concluded that family therapy for pediatric asthma appeared to reduce the severity of asthma and improve lung function, and it may be a useful adjunct to medication therapy.  In our study, a substantial amount of time was spent educating parents and children on how to recognize and rate their asthma symptoms and how to perform peak flow measurements, assess readings, and use ß2-agonists appropriately. The increased sense of control and knowledge about the asthmatic condition is likely to have resulted in anxiety reduction, contributed to proper medication use, and thus may also explain some of the observed improvement in outcomes.
The daily use of asthma diaries might in itself account for improvement in both the active and sham SMT groups. A recent randomized trial showed that by having asthma patients write about stressful life events, pulmonary lung function was increased.  It is possible that by having patients in our study subjectively evaluate and rate their asthma symptoms, this expression of their asthma-related stressful events resulted in increased asthma-related quality of life.
The frequency of care and the subsequent social connection that likely developed between the chiropractor and patients also deserves comment. Several studies have indicated that increased socialization is associated with positive health outcomes,  and this too may account for some of the improvements noted in this study.
Overall, our study corroborates the findings of the Balon et al  and Nielsen et al  studies; collectively, the studies suggest that factors other than the specific effects of SMT are contributing to most of the changes in quality of life and patient-rated asthma severity and improvement observed in these studies. If these factors are mainly nonspecific or placebo effects, what does this mean to patients, clinicians, and policy makers in considering the use of chiropractic care in the management of asthma?
The placebo or nonspecific treatment effect has traditionally been regarded as a confounding or nuisance factor to be controlled for or eliminated. However, this nonspecific effect is an important and often powerful aspect of any therapy,  and depending on the patient's experience with the therapeutic encounter, it may potentiate the patient's own healing capacity to different degrees. There is some evidence to suggest that this effect may be mediated by the brain through neural endocrine influences capable of modulating the function of the immune system.  The theory that the brain is capable of influencing the immune system has been confirmed in several RCTs, which used interventions such as suggestion, self-hypnosis, imagery, and relaxation techniques.  The nonspecific therapeutic effect has several known and likely several unrecognized dimensions. In the context of a clinical trial, the manner in which the informed consent is given, the expectations of the patient, and the enthusiasm and the attention of the treatment provider are factors that can have an impact on the patient-experienced outcome. It has been shown in practice-based studies that the doctor's attitude toward therapy, whether positive or negative and either with or without confidence, has an influence on the outcome of treatment. 
It has been argued that it is unreasonable to discard a therapy or consider it worthless if it is only a little better or even no better than a suitable placebo. What matters is the magnitude of effect on patients' outcomes when compared with commonly used treatments and in particular, no-treatment controls, if a patient wishes to decide more rationally which interventions a health care service should pay for.  Two placebo-controlled trials examining the effect of adding chiropractic spinal manipulation to the optimal medical management of chronic asthma in either children or adults showed no important difference between the active and placebo arms. [18, 23] However, in both trials a clinically important improvement in asthma-related quality of life and a reduction in patient-reported asthma severity appeared to result in both active and sham SMT groups. These improvements are unlikely to occur solely as a result of the natural history or regression to the mean. On that basis, it may not be appropriate to deem the addition of chiropractic care to medical management worthless and to proscribe its use.
When interpreting the findings of our prospective clinical series, it is impossible to make any causal inferences. The improvements observed in patient-oriented outcomes may be the result of a multitude of factors. Considering the chronicity of the disorder, the changes were unlikely the result of natural history. Some of the changes may be explained by regression to the mean because patients often enroll in studies when their symptoms are most severe. However, in this study patients went through a 2-month baseline period during which time their medical management was optimized. Changes in outcomes were measured from the end of this baseline period. In addition, it is possible that the specific effect of spinal manipulation may have been masked by the effect of the medications. Ethical considerations prevent the assessment of spinal manipulation alone in mild to moderate asthmatics. However, it is possible to design a study in which patients are given spinal manipulation in addition to medication and then monitored to see if medications can be reduced. According to the most recent guidelines, such a “step down” in medication should not be considered until the asthmatic condition has remained stable and has improved for at least 3 months. Thus for this to be assessed adequately, patients should be managed for longer periods to see if the reduction in medications can occur and if so, be maintained.
Another consideration is that during the study period, almost half the children had upper respiratory infections. These children had substantially poorer outcomes compared with the children who did not have upper respiratory infections. Upper respiratory infections are extremely common in children with asthma and tend to mask improvement from ongoing therapy. Finally, it is possible that the SMT may have specific effects in only certain subgroups of patients, which we were unable to identify in our study given the relatively small sample.
One of the strengths of this study is that it reflects to a certain extent what is occurring in health care today. Substantial numbers of patients are seeking “unconventional” health care in addition to medical care rather than as a replacement for it.  Future studies should continue to assess the multidisciplinary comanagement of asthma to enhance study generalizability and even more importantly, to optimize patient care. We submit that future studies should focus their attention on assessing what aspects of the chiropractic clinical encounter are responsible for the improvement in important patient-rated outcomes observed in the studies performed to date. Is it the touch and attention? Is it the relaxation that may ensue from the physical nature of the chiropractic interventions? Maybe it is the filling out of diaries, the continuous monitoring, or increased patient focus on their condition, resulting in a better compliance, decreased anxiety, and appropriate use of prophylactic and abortive medications. It is possible still even in the light of the current scientific evidence that spinal manipulation does have worthwhile specific effects in certain patient populations. Likely, it is a combination of some or all of these factors. In any case, something is occurring that makes the patients feel better, as indicated by changes in well-recognized measures of quality of life.  For this reason alone, future exploration is warranted. In addition, studies with larger sample sizes will be necessary to identify if worthwhile specific effects can be demonstrated in subgroups of patients.
The personal dimension of the clinical encounter offers a rich potential for useful interventions.  The generic elements of empathy and verbal and nonverbal communication (including listening and touch) need to be explored. The complexity of the physical, psychologic, and sociologic components of the chiropractic clinical encounter must be acknowledged, and it is likely that new methods for assessing these complex effects will need to be developed. 
After 3 months of combining chiropractic SMT with optimal medical management for pediatric asthma, the children rated their quality of life substantially higher and their asthma severity substantially lower. These improvements were maintained at the 1-year follow-up assessment. There were no important changes in lung function or hyperresponsiveness at any time. The observed improvements are unlikely to be the result of the specific effects of chiropractic SMT, but other aspects of the clinical encounter that should not be readily dismissed. Further research is needed to assess which components of the chiropractic encounter are responsible for important improvements in patient-oriented outcomes so that they may be incorporated into the care of all asthmatic patients.