Original Article

II. Quantitative Research

Assessing the Quality of Randomized Controlled Trials in Acupressure for Women Health: A Systematic Review (Supplementary material 5.)

Grace Hsiao-Hsuan Jen¹, May Mei-Sheng Ku², Mindy Szu-Min Peng¹, Abbie Ting-Yu Lin¹, Rene Wei-Jung Chang¹, Wilson Wei-Chun Wang¹, Sherry Yueh-Hsia Chiu², Chen-Yang Hsu¹, Lisa Li-Chen Hsieh⁴, Hsiu Hsi Chen¹

¹Institute of Epidemiology and Preventive Medicine, College of Public Health, National Taiwan University, Taipei, Taiwan

²Institute of Environmental Health, College of Public Health, National Taiwan University

³Department and Graduate Institute of Health Care Management, Chang Gung University, Tao-Yuan, Taiwan

⁴Ji-Din Clinic, Kaohsiung, Taiwan

Abstract

Background

The general application of the “Qi” practice though acupressure and cupping in orthodontic medicine of women health is hampered by the lacking of evidence in the efficacy. Although there are increasing clinical trials on this filed, the quality on the design, implementation, and reporting is barely addressed.

Material and Methods

A scoring system based on the guideline of Consolidated Standards of Reporting Trials (CONSORT) updated in 2010 first developed. Literatures on applying the acupressure and cupping for women health using a randomized controlled study design (RCT) were reviewed by the authors to ensure the relevance for the context of the study aim. The authors were trained and calibrated in the standard of scoring for each item. The total score and that normalized by the full score for each item and study were used to assess the absolute and relative quality, respectively.

Results

Among the 76 article searched from PubMed using the keyword of “acupressure” and “randomized controlled trial”, 13 of them were enrolled. Among the full total score of 66, the average score for the 13 articles was 39.8 (SD: 10.2) and the normalized rank was 60%. The score for the category of“Title and Abstract”, “Study design”, “Study implementation”, “Results”, and “Discussion” was 5.5 (SD: 1.7, rank: 69%), 10.0 (SD: 1.9, rank: 83%), 7.2 (SD: 3.6, rank: 45%), 9.7 (SD: 3.9, rank: 54%), and 6.6 (SD: 2.4, rank: 55%), respectively.

Conclusion

The quality of the study on evaluating acupressure in the field of women health using RCT is modest with the rank reaching 60% of the requirement of CONSORT standard. The main drawback was in the category of “Study Implementation”, which calls for improvement in conducting further study.

Introduction

The efficacy of the “Qi” practice through maneuvers such as acupressure in improving functionality and reducing disease symptoms have gained great attention in recent years (Au et al., 2015; Ernst et al., 2010). Due to the minimal risk of adverse effect compared with the use of chemicals such as pain control medications, anti-emetics, and sedatives, the applications of acupressure in the field of women health, especially for the treatment of dysmenorrhea and discomfort during labor including pain, nausea and vomit, and prolonged headache due to epidural procedures are of great interest. Considering the treatment of dysmenorrhea, the conventional approach in the field of orthodontic medicine including endocrinology and gynecological assessment to rule out the disease incurred by organic disease such as endometriosis. For subjects with primary dysmenorrhea the mainstay of treatment is pain control and hormonal therapy (Osayande et al., 2014; Wallace et al., 2010). However, the chronicity and periodical discomfort induces by dysmenorrhea is often a concern when these medication therapies are provided to women. The treatment for women with the symptoms associated with pregnancy and labor is also faced with safety considerations (Wallis et al., 2012; Ebrahimi et al., 2010; Caton et al., 2002).

Although the reported efficacy is satisfactory compared with the minimal risk of adverse effect, the main criticism in generalization and wide application of these maneuvers to clinical practice is that there is a lacking of evidence basis. Facing with such criticism, there are also studies tempting to assess the efficacy of acupressure following the principle of evidence based medicine with the randomized controlled study design (RCT). However, the quality of these studies in terms of the standards of reporting RCT, namely CONSORT checklist (Schulz et al., 2010), was not systematically evaluated.

To have a better understanding on the current evidences on efficacy of applying acupressure for the issue of women health, we thus performed a systematic review to summarized the findings and also the process of reaching the results. In this study, we further aimed to quantify and assess the extent of adherence to scientific principle for current evidences using a CONSORT-based scoring system.

Material and Methods

Systematic review for randomized controlled trial on the efficacy of acupressure on women health

The systematic review was conducted by searching the published articles from PubMed with the searching keywords of “acupressure”, clinical trial of article types, and free full text in English until June, 2018. The flowchart of retrieving literature is illustrated in Figure 1. Two authors (HHJ and SMP) independently searched the articles with the same strategy. The final decision of study selection was further reviewed by LCH to confirm the relevancy of the study topic of acupressure and women health. There were 13 articles related to gynecological health for evaluating the efficacy of acupressure and six authors (HHJ, SMP, MSK, TYL, WRC, and WCW) independently retrieved data and evaluated the quality of evidence by scoring system with CONSORT checklist (Schulz et al., 2010) elaborated as follows.

Development of scoring system assessing the quality of reporting randomized controlled trial

Since all of these studies are randomized controlled trial (RCT), the quality of reporting was assessed by using a scoring system derived from the CONSORT checklist guideline. A three-point scoring system was developed ranged from 0,1, and 2 representing the quality of “not addressed”, “addressed but with compromised quality”, and “fully addressed” for each item. Following the updated version of CONSORT checklist proposed in 2001 (Moher et al.), a total of 33 items depicting the necessary structure of reporting a randomized controlled trial including abstract, introduction, material and methods, results, and discussion was used as the backbone of the development of the scoring system for assessing the quality of collected studies. For the study with irrelevant item, a note of “not applicable” was filled. Excluding those not applicable items, it remains 33 items on average, so the total score is 66. For the purpose of calibrating the standard of scoring among the evaluators and validating the feasibility of using the scoring system on acupressure studies, an article published by LCH (Hsieh et al., 2006) was used as standard material before the evaluation of collected literatures.

Results

Literatures on acupressure and women health

Among these 13 articles, one conducted by Pouresmail et al. (2002) was to assess the efficacy of acupressure and Ibuprofen on primary dysmenorrhea and shown there were no difference between this two interventions, but both could reduce pain grade in comparison with sham acupressure (placebo); one conducted by Kashefi et al. (2011) was shown acupressure had more efficient to women general health than sham acupressure (placebo); and others were to explore the efficacy of acupressure for pregnant women before/after childbirth and shown acupressure could reduce morning sickness, nausea or vomiting, the intensity of pain, the length of labor stages, the labor duration and so on and also increase the infants’ Apgar scores (sTable 1).

Quality of reporting the efficacy of acupressure on women health

After scoring each studies to evaluate the quality of evidence with CONSORT checklist guideline, the mean of total score is 39.8 (range: 16-53; SD=10.2) and the mean of score are 5.5 (range: 4-8; SD=1.7), 10.0 (range: 7-12; SD=1.9), 7.2 (range: 2-14; SD=3.6), 9.7 (range: 6-15; SD=3.9), and 6.6 (range: 2-10; SD=2.4) in the “Title and Abstract”, “Study design” (including trial design, participants, interventions, outcomes, and sample size), “Study implementation” (including randomisation, blinding, and statistical methods), “Results”, and “Discussion”, separately (Table 1). The most discrepancy was in the “Study implementation” and “Results”.

For the studies related to the efficacy of acupressure associated with women health and labor, the complete statement was in the “Study design” and “Results”, and most studies did not mention about the “Randomisation” in the “Methods” section (including sequence generation, allocation concealment mechanism, and Implementation) (Table 2). The study with minimum score was conducted by Pouresmail et al. (2002), and only specified clearly in the “Introduction” section. The other study with maximum score was conducted by MafetoniI RR et al. (2016), the only weakness is in the “Discussion” especially for the information on registration, available protocol, and funding. In addition, it can be observed that the articles published in the recent years had higher score, and those published before 2014 tended to not identify as a RCT in the title.

In Figure 2, the study published by LCH (Hsieh et al., 2006) using as benchmark got 91% of normalized quality score and other studies related to women health and labor had 60% of normalized quality score. The score in each section is 69% (5.5/8) in the “Abstract” section, 83% (10/12) in the “Study design”, 45% (7.2/16) in the “Study implementation”, 54% (9.7/18) in the “Results”, and 55% (6.6/12) in the “Discussion”.

Discussion

By using a scoring system with the CONSORT checklist underpinning, we assessed and quantified the quality of current evidences on the efficacy of acupressure for women health following the guideline of scientific principle. Among the 13 enrolled articles using the randomized controlled study design, an overall rank of 60% (39.8/66) was observed, showing a compromised result for current published article in this field. There is also a remarkable variation across studies with the standard deviation estimated as 10.2 for the overall score. Considering the scores of the aspect of “Title and Abstract”, “Study Design”, “Study Implementation”, “Results”, and “Discussion”, the lowest rank was the “Study Implementation” (45% (7.2/16)), followed by the “Results” (54%, (9.7/18)) and “Discussion” (55%, (6.6/12)). The low rank in these three aspect demonstrating the aspect required for improvement in conducting and reporting a randomized controlled study for providing the evidence on the efficacy of acupressure with scientific background.

Our research focused on the application of acupressure for women health with randomized controlled trial study design. Given the increased attention on gathering scientific evidence for the Qi-based practice such as acupressure and its potential application on personalized medicine, the proposed scoring system can be extended to other study type such as observational study and include a wide range of research topic to be used as a first step for evidence synthesis.