Abstract
Objective:
The aim of this study was to evaluate the treatments delivered during a pragmatic effectiveness study of acupuncture for irritable bowel syndrome (IBS) and to explore the roles of Traditional Chinese Medicine (TCM) diagnoses and preferences of the acupuncturists in treatment design.
Methods:
Patients allocated to an acupuncture arm of a study were offered up to ten sessions of acupuncture over 3 months. Acupuncturists followed a flexible treatment protocol that allowed for treatment individualization, use of additional therapies such as moxa, and provision of lifestyle advice. All treatments were recorded in logbooks and analyzed by content analysis.
Results:
Seven primary and eight secondary TCM patterns were identified among the 113 patients with IBS, and were combined in various ways to produce unique diagnoses for 84% patients. Liver Qi Stagnation and Damp Heat were the most commonly reported patterns. Of the 126 acupuncture points used, a distinct core and supporting group of points was associated with each primary pattern. Each practitioner also utilized a distinct core and supporting group of points that reflected his/her preferences. Use of additional therapies (e.g., moxa) and provision of lifestyle advice (e.g., diet) were generally consistent with a particular practitioner.
Conclusions:
Data suggest that a patient's TCM diagnosis, the practitioner's preferences, and a patient's individual characteristics beyond his/her diagnosis influence treatment delivery. In particular, TCM diagnoses appear to influence the acupuncture (i.e., point selection) aspect of treatment more than the selection of additional therapies and lifestyle advice. From another perspective, the treatments incorporated pragmatic, individualized, and disease-specific approaches with combinations that produced both treatment commonalities and diversities.
Introduction
In addition to the diverse array of symptoms, patients with IBS also have various comorbidities and experience fluctuations in symptom severity, which, together, make IBS a complex condition. In research, these problems are complicated further by heightened placebo responses. Trials of biomedical interventions have reported placebo response estimates between 16% and 71%, and for complementary and alternative medicine (CAM) interventions average 43%. 7 Similarly, acupuncture is criticized because its beneficial effects are related to heightened placebo responses. 8 A pragmatic approach to treatment and a minimum 6-month study duration has been suggested as a way to overcome the heightened placebo responses associated with acupuncture and IBS studies. 9,10 A pragmatic study design also would potentially overcome challenges posed by the complexities of acupuncture as an intervention including: practitioners' preferences for particular acupuncture points; variety of styles (e.g., Traditional Chinese Medicine [TCM], Five Element, trigger point); and many modalities (e.g., moxa, cupping). 10,11
According to Verhoef et al., allowance for diagnosis differentiation and treatment individualization make the pragmatic design a more-valid approach to TCM research. 11 However, Unschuld has suggested that treatment individualization is only one of three common approaches to treatment. 12 In addition to treatment individualization, TCM practitioners also use disease-specific and/or pragmatic (i.e., utilization of a useful remedy) treatments. Moreover, the three treatment approaches may be used in combination, and they are also used by practitioners in other medical disciplines, including biomedicine. As such, emphasis on treatment individualization may overshadow if and/or how the different approaches would be combined in practice.
To date, approximately, 24 trials of acupuncture for IBS have been conducted (see Appendix online at
Methods
Data for this article were derived from the acupuncture arm of a two-parallel-arm randomized controlled trial of acupuncture plus usual care versus usual care for IBS. The trial was conducted in North Yorkshire, United Kingdom and was approved by the York research ethics committee (#08/H1311/66).
Recruitment and randomization of patients with IBS to acupuncture were explained previously in the trial protocol, 14 and articles regarding main outcomes have been published elsewhere. Patients who were allocated to acupuncture were offered ten sessions of acupuncture. Each patient was assigned to an acupuncturist based on that patient's preference for clinic location and appointment time.
Acupuncture practitioners were recruited on the basis of their membership in the British Acupuncture Council, years of experience (≥3), locations respective to the general practitioner (GP) practices from which patients were recruited, and the acupuncturists' styles of acupuncture. Acupuncturists in North Yorkshire primarily practice TCM; therefore, 9 TCM-trained acupuncturists were recruited.
The protocol for the pragmatic intervention was derived from the pilot study 15 and from practitioner interviews that mapped the routine treatment of IBS. 5 From this, a treatment protocol was developed that instructed practitioners to deliver routine acupuncture treatments that could incorporate additional therapies, such as moxibustion, cupping, or t'ui na, as well as lifestyle advice that were considered to be appropriate. Use of herbs and magnets was prohibited, and the practitioners were not allowed to recommend probiotics, as these interventions were outside the scope of the trial's objective. The emphasis on routine treatments allowed for flexibility in treatment design and necessitated use of a systematized logbook to record treatment details.
The systematic design of the logbook was intended to facilitate comparisons across treatments. Each logbook was used to record a detailed account of the patients' treatments and included the use of both checkboxes and free-text questions. Specifically, free-text questions required the acupuncturists to identify each patient's chief complaint, TCM diagnosis, treatment principle, additional therapies, points used and reason for their selection, and lifestyle advice with a rationale for offering that advice. Checkbox questions required the acupuncturists to identify the TCM framework used to make their diagnoses.
Because the logbooks presented an atypical narrative account, the data were analyzed by content analysis. Content analysis allowed data to be extracted in categorical units, which were compared descriptively or reviewed statistically. 16 To address the difficulty of analyzing free text, data for particular categories were organized into phrase inventories. 17 For example, variations in free-text diagnoses were condensed into groups based on similar phrases, such as “Damp Heat,” “Damp Heat in the Large Intestine,” and “Damp Heat in the Lower Jiao.” Condensing free-text diagnoses according to similar phrases has also been reported by Coeytaux et al. 18
After analyzing the data from an overall trial perspective, such as the most commonly used acupuncture points, the data were analyzed according to diagnosis (i.e., diagnosis-specific) and acupuncturist (i.e., acupuncturist-specific). For example, data for patients diagnosed with Damp Heat were collectively analyzed to evaluate treatment consistencies. Similarly data for patients treated by the same acupuncturist were analyzed collectively.
In addition to analyzing the treatments delivered, an exploratory analysis was also conducted on the TCM diagnoses to develop a hypothesis about the relationships between an individual patient's TCM diagnosis and his/her age, IBS severity, and IBS duration. Data for this analysis were derived from the treatment logs and the patients' baseline questionnaires. A descriptive comparison and analysis of variance (ANOVA) were conducted, using SPSS 17.0.2.
Results
Over the course of the trial, the 9 acupuncturists delivered 1016 treatments to 113 of the 116 patients allocated to acupuncture with an average of nine sessions per patient. On average each acupuncturist treated 13 patients (range 3–22) and provided 113 acupuncture sessions (range 30–205). The acupuncturists used approximately 126 different acupuncture point locations and needled an average of seven points per treatment. Aspects of treatment were reported based on their overall frequency, diagnostic specificity, and/or acupuncturist specificity.
Background characteristics of the patients are presented in Table 1. Because patients diagnosed with IBS in primary care have a variety of symptoms, the acupuncturists were asked to identify patients' chief complaints. Approximately nine different symptoms were indicated as the patients' chief complaints, most commonly including alternating bowel movements, diarrhea, constipation, pain, and indigestion. These symptoms were commonly accompanied by bloating and/or flatulence.
Severity was measured by the IBS symptom severity scale for which scores range from 0 to 500 and indicate no IBS<75, mild IBS 75–175, moderate IBS 175–300, and severe IBS>300.30
IBS, irritable bowel syndrome.
In addition to the chief complaint, at the initial session, the acupuncturists also identified the framework/s they used to diagnose each patient. The acupuncturists most commonly used the Zang-Fu framework (99% of diagnoses) and incorporated six other frameworks including: Qi–Blood–body fluids, Eight Principles, Five Elements, pathogenic factors, four levels, and six channels. Acupuncturist-specific analyses suggested that the practitioner's training and preference influenced framework selection. For example, 1 practitioner with Five Elements training routinely used a combined Five Element and Zang-Fu approach to diagnosis.
As discussed in Methods, free-text diagnoses were condensed into phrase inventories based on common phrases. Seven distinct primary diagnostic classifications emerged from the analyses, among which Liver Qi Stagnation and Damp Heat were the most-common diagnoses (Table 2). Analyses of the secondary diagnoses indicated that Liver Qi Stagnation was also among the most-common diagnoses, overall affecting 67% of patients. Secondary diagnoses not among the primary list included Food Stagnation and Phlegm. Combined analyses of the primary and secondary diagnoses suggested that 84% of patients received a unique diagnosis. The exploratory ANOVA suggested that the mean age, IBS severity, and IBS duration were not statistically different across diagnoses at baseline (p=0.099, 0.442, and 0.140, respectively). Therefore, none of these factors appeared to be predictive of TCM diagnosis; however, more research is necessary. The acupuncturist-specific analyses suggested that there was a positive relationship between the diagnoses and the practitioner; however, sample size limitations prevented a statistical comparison of these findings.
IBS, irritable bowel syndrome.
Of the 126 different acupuncture points used, four (LR 3, LI 4, St 36, and Sp 6) comprised the core group as defined by use in ≥50% of treatments. An additional five points (CV 12, GB 34, KI 3, Sp 9, and St 25) comprised the support group as defined by use in 25% to 49% of treatments. Combining the core and supporting points accounted for 57% of the 7185 reported points used in treatment. The remaining 117 (93%) points comprised the variable group as defined by use in <25% of treatments. Among the variable points, 43 (37%) were only used on 1 patient.
Diagnosis-specific analyses indicated that two to six points comprised the core and support groups and that eleven of the nineteen points only appeared in one or two core and/or support groups (Table 3). This indicated that diagnosis may be an important feature of point selection. Similar to the diagnosis-specific analyses, the acupuncturist-specific analyses identified core and support groups for each practitioner (Table 4). In addition the majority of acupuncture points most practitioners used comprised their variable group. Practitioners' preferences were highlighted by use of 55 points (44%) by 1 acupuncturist as well as variations in the composition of core and support groups. As such, the practitioners' variety of points appeared to be independent of the number of his/her patients but did appear to be related to his/her preferences and the individuals.
Bolded points are core points and nonbolded points are support points. Mean usage indicates the average number of sessions a particular point was used per patient.
Each acupuncturist's number of patients is noted in parentheses.
As part of the acupuncturist-specific analysis, a longitudinal comparison of each practitioner's set of treatments revealed three distinct treatment approaches: (1) fixed nucleus of points with moderate variations in additional points; (2) diverse treatments that lacked a consistent nucleus; and (3) repetitive points with minimal variations. Although each acupuncturist used each approach, the fixed nucleus approach was the most common and was more commonly used by more-experienced practitioners. As such, this finding distinguishes between practice style and years of experience. The longitudinal comparison also illustrated how practitioners changed their treatment emphases over time to account for concomitant symptoms or conditions in addition to bowel problems. Specifically, the acupuncturists noted treatment alterations for headaches, cold and flu symptoms, night sweats, low-back pain, cough, and emotional issues.
With regard to additional therapies, 25 patients (22%) received one or more therapies in 11% of the treatments (Table 5). Diagnosis-specific analyses indicated that diagnosis was a minor contributing factor to additional therapy usage, whereas the acupuncturist-specific analyses revealed a strong relationship between practitioners' preferences both in type of therapy and the frequency of use (Table 4). From the analyses, it was inferred that some practitioners were more comfortable delivering additional therapies and/or placed more importance on them. Despite accounting for the acupuncturist and diagnosis, some potentially important factors, such as patient characteristics (e.g., willingness) and length of treatment were not explored.
Mean number of stimulations per participant receiving that additional therapy.
Mean number of stimulations per participants receiving acupuncture.
Avg., average.
The acupuncturists gave lifestyle advice to 77 (68%) patients across multiple categories (Table 6). Of the patients who received advice, 53 (47%) received advice from multiple categories. Advice in the “other” category included comments such as “warm abdomen with hot water bottle” and “stop smoking.” The 7 patients who received advice outside the scope of the trial were informed about probiotics, herbs, and/or colonic irrigation.
GP, general practitioner.
While the three most common types of advice remained consistent across diagnoses, a detailed analysis of the relationship between diagnosis and advice only revealed consistency between dietary advice and a Yang Deficient diagnosis. Yang Deficient patients were typically advised to eat warm/cooked foods and avoid raw/cold foods. Patients with other diagnoses received a mixture of dietary advice including: eat warm/cooked foods; avoid wheat, sugar, and dairy; limit alcohol; set regular meal times; and manage food portions. The apparent lack of consistency between primary diagnosis and advice is discussed further in the subsequent section of this article (Discussion).
Acupuncturist-specific analyses determined that 4 practitioners provided advice to all of their patients, while the other practitioners provided advice to 40%–69% of their patients. The analyses provided additional evidence on the importance of practitioner's preference and training, specifically in additional modalities such as colonic irrigation, to overall treatment. Additionally the analyses highlighted differences among the practitioners' perceived importance of advice as an adjunct to treatment.
Discussion
Overall, the findings suggest that TCM diagnosis differentiation is relatively more important to the acupuncture aspect of treatment (i.e., point selection) than it is to the use of additional therapies or lifestyle advice. The use of additional therapies and lifestyle advice appear to be more related to the acupuncturist's preference and the patient's characteristics beyond his/her diagnosis. With regard to the treatment approaches described by Unschuld (i.e., pragmatic, disease-specific, individualized), 12 the data demonstrated that the acupuncturists delivered treatments that combined all three. It is important to highlight that, within each of the approaches the acupuncturists may have incorporated their preferences, thereby creating an idiosyncratic aspect of treatment, which is consistent with routine care that was mapped prior to the trial. 5
Although a summary may be distilled from the most-common aspects of treatment, a comprehensive description that incorporates diagnosis-specific and/or acupuncturist-specific data from pragmatic treatments posed a reporting challenge. In contrast, efficacy studies that us a standard treatment may utilize relatively straightforward reporting such as the STandards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) guidelines. 19 As such, the guidelines provide a platform for reporting precisely defined treatments that other studies may choose to replicate. The STRICTA guidelines do provide useful information on both types of studies; however, the flexible nature of pragmatic studies may generate dozens of points and techniques that are not reported. Because pragmatic studies are not intended for repetition, in the way efficacy studies are, and are instead intended to mimic routine practice, we suggest that researchers who conduct pragmatic studies should make an effort to examine the extent to which the treatments are pragmatic and/or indicative of routine practice. For example, pragmatic studies may use process evaluation to explore treatment delivery, whereby a mixture of quantitative and qualitative methods are used to assess the routineness of delivery. 20 The routine mapping of IBS treatment prior to this study and a logbook question about treatment limitations indicated that treatments were relatively routine. The acupuncturists' comments suggested that they were partially limited by the restrictions against herbs and probiotics, and desired more sessions for some patients. A measure or discussion on the routineness of trial treatments may have practical implications, because practitioners may be more willing to incorporate findings from pragmatic trials than efficacy (e.g. fixed) studies into practice. 21
With regard to the diagnoses, the seven primary diagnoses illustrate the concept that a single biomedical diagnosis may be associated with many TCM diagnoses. 22 A study by Tan et al. reported that IBS is associated with Liver Qi Stagnation, Spleen Qi Deficiency, and a combination of the two. 23 Although the diagnoses were among the ones reported in this study, the methods used by Tan et al. imply that their diagnoses were predetermined, which provides a potential source for the difference between the studies. Alternatively, a study by Forbes et al. reported that practitioners were allowed to make unrestricted diagnoses, which consequently influenced the selection of treatment principles and acupuncture points. 24 However, the authors of that study did not identify the diagnoses. Therefore, the current study is the first to provide empirical evidence for TCM diagnoses associated with IBS.
In terms of the relative importance of the TCM diagnosis, both this study and the Forbes study 24 suggest that TCM diagnosis plays a role in treatment design. In contrast, a low-back pain study by Sherman et al. reported a lack of consistency among diagnoses and treatments, which the authors attributed to variations in practitioners' idiosyncratic approaches to treatment. 25 In conclusion, Sherman et al. suggested that the lack of consistency may reflect the nonimportance of the diagnosis to treatment design. As mentioned elsewhere, 26 these studies may demonstrate that some TCM diagnoses are more influential on treatment design than others. Alternatively, the consistency in treatment between TCM diagnoses or lack thereof may be related to the patients' chief complaints. Although low-back pain may be associated with many TCM diagnoses, the patients may report similar chief complaints (e.g., pain). In comparison, IBS is also associated with many TCM diagnoses, but patients reported a range of potentially dissimilar chief complaints, such as bowel problems and fatigue. Further research is necessary to determine if particular diagnoses and/or chief complaints have a greater impact on treatment design than others.
Two other aspects of treatment—additional therapies and lifestyle advice—contribute to the discussion on the relative importance of diagnosis. Analyses of the additional therapies suggested that diagnosis was less of a contributing factor to usage than practitioner's preference. Cassidy reported a similar finding wherein the most common additional therapy (moxa) remained consistent regardless of diagnosis. 27 In comparison, the diagnosis and acupuncturist-specific analyses of lifestyle advice also suggest that preference outweighed diagnosis as an influential factor. Considering that lifestyle advice is an area of treatment that practitioners can readily apply additional forms of training to complement their TCM practice—such as reflexology or colonic irrigation—it may be beneficial for researchers to explore the relationships among patients' diagnoses, advice given, and practitioners' preferences.
In addition to this study, there were 2 other relatively pragmatic trials of acupuncture for IBS, and 1 qualitative study on routine treatment. The study by Forbes et al. did not specify differential diagnoses or the acupuncture points used in treatment, thereby limiting comparability. 24 However the pilot study 15 for the current trial and routine mapping of treatment 5 identified acupuncture points that were consistent with the core and support groups, specifically CV 12, LR 3, LI 4, St 36, and Sp 6. Furthermore, a study of routine practice incorporating a variety of conditions reported an identical core group of points, with the exception of SJ 5, to this study. 28 In this study, Napadow et al. noted that the most commonly used points have many indications and may be used to address a variety of conditions, thereby contributing to the popularity of these points. In addition, the current authors suggest that the points may also be popular because of their locations and accessibility. Given that four of the five points are consistent across many studies reaffirms Napadow et al.'s conclusion that a small core of points may be used to treat a variety of conditions and that those points are supplemented by an idiosyncratic group of points. 28 Further studies are necessary to explore whether the core group, the idiosyncratic group, or the combination of the two drives outcome. Researchers may use a two-arm trial design, wherein all participants receive usual care and are allocated to receive either modified acupuncture treatments that stimulate the four to five core points or pragmatic treatments
Conclusions
Patients received TCM treatments that incorporated pragmatic, disease-specific, and individualized selection of acupuncture points. This study contributes knowledge to the field through the variation in analytical perspectives, which allowed for a broader exploration of what motivates practitioners to provide certain types of treatments. The disease- or diagnosis-specific influence played a more prominent role in point selection than use of additional therapies or lifestyle advice. The latter two aspects were more influenced by the practitioners' preferences. Treatment analyses highlight the need for more research on the relative importance of TCM diagnosis to treatment design, and the contribution of core points, support points, and variable points to patients' outcomes.
Footnotes
Acknowledgments
The authors would like to acknowledge Sally Brabyn, M.S. Helen Tilbrook, M.S., Karen Bloor, PhD and Karl Atkin, PhD for their contributions to this work. The authors would also like to thank the patients and practitioners who participated in the trial.
Disclosure Statement
No financial conflicts exist.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
