Article Text

other Versions


Development and validation of the Chinese version of the Massachusetts General Hospital Acupuncture Sensation Scale: an exploratory and methodological study
  1. David Tai Wai Yu1,
  2. Alice Yee Man Jones2,
  3. Marco Yiu Chung Pang2
  1. 1Physiotherapy Department, Queen Elizabeth Hospital, Hong Kong, China
  2. 2Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong, China
  1. Correspondence to Professor Alice Jones, Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, 11 Yuk Choi Road, Hung Hom, Hong Kong, China;{at}


Background The Massachusetts General Hospital Acupuncture Sensation Scale (MASS) is a tool to measure needle sensations. The aims of the present study were to develop a Chinese version and to assess its psychometric properties.

Methods This study was a methodological and exploratory study. The English version of the MASS was translated into Chinese using standardised translation procedures. Content validity was conducted by nine acupuncture experts. The prefinal Chinese version (C-MASS) was then administered to 30 acupuncture-naïve, healthy subjects. Electroacupuncture was performed on the right LI4 and LI11 acupoints for 30 min. A test–retest reliability measurement was administered 1–2 weeks later. Construct validity was examined by comparing results from C-MASS and the Short-Form McGill Pain Questionnaire (SF-MPQ). The construct validity was further assessed by the principle component analysis.

Results C-MASS demonstrated a content validity ratio on relevance and importance from −0.04 to 1.00. Convergent validity was demonstrated by its significant association with the sensory dimension of SF-MPQ (γ=0.63, p<0.05). Discriminant validity was demonstrated by its low association with the affective dimension of SF-MPQ (γ=−0.3, p=0.111). A five-factor structure of C-MASS was established by factor analysis. C-MASS demonstrated good internal consistency (Cronbach's α=0.71) and test–retest reliability (intraclass correlation coefficient=0.92). Since the descriptor ‘sharp pain’ was not a valid needle sensation related to deqi, this was removed from C-MASS. We renamed the scale as the Modified MASS-Chinese version (C-MMASS).

Conclusions A 12-descriptor C-MMASS was established and shown to be a reliable and valid tool in reporting needle sensations associated with deqi among healthy young Chinese people.

Statistics from


Acupuncture has been widely used in China for thousands of years.1,,3 One of the fundamental characteristics of acupuncture is to ‘obtain qi’ during acupuncture, a sensation referred to as ‘deqi’. Based on the concept of traditional Chinese medicine (TCM), qi must flow in correct strength and quality along the meridians so that health can be maintained. It was believed that restoration of health can only be achieved if the acupuncture technique is able to elicit deqi, thereby allowing the flow of qi to be altered.3 4 While investigations of the relationship between the therapeutic effectiveness of acupuncture and the deqi experience have been reported in the literature,3 5,,7 scientific evidence to support such a relationship is still lacking.7

When deqi occurs, the acupuncturist and the subject may experience some ‘unusual’ sensations around the needle. The acupuncturist may perceive heaviness or tenseness around the needle when qi arrives.8 However, the feelings of the acupuncturist are more often subjected to biased preconceptions of ‘what one ought to feel’ and thus hold greater likelihood of a biased subjective report.9 In recent years, researchers have focused more on sensations perceived by the subjects. Standard needle sensations associated with deqi experienced by the subjects are described as soreness, distension, heaviness and numbness.10,,12 Nevertheless, there is still no standard method in measurement and quantifying deqi sensations at present.

Over the last two decades, a number of studies have attempted to quantify deqi sensations.10,,12 Vincent and colleagues adopted the McGill Pain Questionnaire and created a new scale of 20 adjectives describing possible acupuncture sensations.10 In 2005, Park et al modified Vincent and colleagues' work by adding five extra descriptions of the sensation scale.11 However, these scales have limitations because both of these instruments measure a range of sensations that include pain as well as deqi. In 2006, a panel of 29 international acupuncture experts were involved in MacPherson and Asghar's study, which categorised 25 sensations during acupuncture as associated with deqi and acute pain at the needling site.12 However, the above study quantified deqi sensations based on the acupuncturists rather than the patient's/subject's perception.

To address the complexity involved in accurately assessing the acupuncture sensations, Kong and coworkers developed the ‘Subjective Acupuncture Sensation Scale’ (SASS), an inventory incorporating various sensations associated with deqi.13 The SASS listed nine descriptors of sensations associated with deqi sensations reported in the traditional literature.13 Each of the elements was presented on a 10-cm bar with the anchor words ‘none’, ‘mild’, ‘moderate’ and ‘severe’ spaced evenly along the continuum. Subjects were asked to quantity their sensations at each point by rating the intensity with which they experienced after acupuncture. The authors demonstrated that there were significant correlations between numbness and soreness items of the SASS.13 In order to apply SASS to a wider range of research projects, Kong and his coworkers further expanded the descriptors in the SASS, and this became the Massachusetts General Hospital Acupuncture Sensation Scale (MASS) (Appendix 1).7 The descriptors in MASS include soreness, aching deep pressure, heaviness, fullness/distension, tingling, numbness, sharp pain, dull pain, warmth, cold, throbbing, plus a blank supplementary row left for subjects to describe perceptions in their own words. The MASS index is then calculated and used to quantify the intensity of the needle sensations experienced by a subject. The MASS also included two supplementary scales for measurement of spreading of the acupuncture sensation and anxiety.7

Establishment of the MASS involved extensive review of relevant literature and appears to be the most comprehensive assessment tool in measuring the needle sensations. The MASS has subsequently been used for measurement of acupuncture sensations.14 15 The validity and reliability of the MASS, however, have not been properly established. Furthermore, the precise Chinese language terms that best describe the needle sensation have not been determined. If optimal needle sensation is essential for achieving positive acupuncture effects, it is necessary to have a valid and reliable instrument to be used in the clinical setting to systematically quantify and document the needle sensation associated with deqi, as this could be an important factor that determines the treatment effect of acupuncture intervention. The objectives of the present study therefore were to establish a Chinese version of the MASS and to test its psychometric properties.


The study was an exploratory, methodological study that involved two phases. Ethics approval was granted by the Human Subjects Ethics Committee of the university involved. The nature of the study was explained to the subjects and written consent obtained prior to data collection. All procedures were conducted in accordance with the Declaration of Helsinki.

Phase I: cultural adaptation of the MASS and assessment of content validity

Permission was obtained from the original authors of the MASS before the initiation of the study. The translation process was divided into five stages and was mainly based on the guidelines that described by Beaton et al 16 and Wild et al.17

The first stage was the forward translation of the MASS into two independent Chinese versions by a physiotherapist and a professionally trained translator whose mother tongue is Chinese. The professionally trained translator had no medical or clinical background (naïve translator).

The second stage involved the synthesis of the results of the translations. The translated versions were compared and any ambiguous wordings were identified. Thus, a single Chinese version of the MASS was first established.

The third stage was the back translation stage. This involved another two independent translators with physiotherapy and TCM backgrounds and they translated the Chinese version into English.

In the fourth stage, a panel consisting of a professor from a local university, seven physiotherapists and two nursing staff examined the preliminary versions in terms of the degree of agreement on the importance and relevance of each item on the Chinese version of the MASS to deqi sensation. All the committee members were bilingual individuals with Chinese as their mother tongue. Six of them had a Master's degree in acupuncture studies and four had a Bachelor's degree in TCM. They had an average of 9 years of clinical experience in acupuncture. They were asked to rate each item on a five-point scale from ‘strongly agree’ to ‘strongly disagree’. The content validity ratio (CVR) was then computed. The CVR=(ne–N/2)/(N/2), where ‘ne’ refers to the number of subject matter that the experts considered as essential measurement items and N is the total number of experts in the panel. The CVR ranged between −1.00 and +1.00, where a CVR of 0.00 means that 50% of the panel believes the item to be essential. With an expert panel of 10 members, a 0.60 CVR was required to meet a 0.05 significance level of importance.18 The prefinal version was then established based on the results on the degree of agreement on the importance and relevance in content validity measurement.

In the fifth stage, the prefinal version was pilot tested on 10 normal healthy young subjects who were naïve to acupuncture and had no prior knowledge of the original MASS. They received electroacupuncture on the right Hegu (LI4) and Quichi (LI11) acupoints (see phase II below) and they were asked to complete the prefinal Chinese version of the MASS. Subjects were asked to select from this prefinal version that best describe the sensations during electroacupuncture. In computing the MASS index, the subject's individual MASS rating scales were ordered from highest to lowest intensity, and the MASS index was calculated according to the following equation: Embedded Image

Where R indicates the ratings for different sensations from highest to lowest and n represents the number of deqi sensations on the MASS questionnaire.7

Based on the feedback obtained from the subjects who participated in the pilot testing, minor modifications were made to the inventory to further improve its clarity. The final Chinese version, the C-MASS, was adopted and used to examine the reliability and construct validity in this study.

Phase II: validation of the C-MASS

Subjects' selection and sample size estimation

Subject inclusion criteria included people of under 40 years of age, right-handed, with normal health, naïve to acupuncture, and able to understand Chinese and English, as the subjects were also required to complete the Short-Form McGill Pain Questionnaire (SF-MPQ) (English version). Subjects with known neurological, cardiovascular and psychological disorders were excluded. A convenience sampling method was adopted. Subjects were invited to a face-to-face interview and demographic data collected.

The software PASS 2008 was used for sample size calculation ( Sample estimation was based on the intraclass correlation coefficient (ICC) obtained during the pilot test–retest reliability study. Preliminary results showed that the test–retest reliability was good, with the ICC exceeding 0.95. Therefore, if the ICC was set at 0.95 with a type I error of 5% and power of 80%, 30 subjects were required.

Procedures for electroacupuncture administration

The acupuncture points Hegu (LI4) and Quichi (LI11) of the right arm were selected as appropriate points for stimulation. All subjects were in a sitting position with both arms supported by a pillow. After the acupoints were located, isopropyl alcohol was applied for disinfection. Then, two 40 mm×0.25 mm single-use, sterile, prepacked stainless needles with guide tubes were inserted into the LI4 and LI11 points. The needles were inserted for about 0.5 inch and then manually manipulated until deqi sensation was reported. In this study, deqi was defined as sensations perceived by the subjects as ‘soreness/numbness/distension/heaviness’ at intensity around a moderate level. This was to ensure that proper acupoints were located and the subjects could identify the types of sensations when deqi arrived. In order to standardise the stimulation throughout the stimulation period, electroacupuncture was used in the present study. Two electrodes were attached to the needles and connected to an electroacupuncture device (ITO, EX-160, Hannover, Germany). Electrical stimulation was then applied for 30 min at a frequency of 2 Hz, pulse duration at 0.6–0.8 ms.13 The intensity of electrical stimulation was increased gradually to a moderate level using a 10-point visual analogue scale.19 Subjects were checked every 5 min and the intensity was readjusted so as to ascertain that deqi sensation was present.

After the electroacupuncture stimulation, the subjects were asked to quantify their acupuncture sensations by first completing the C-MASS and then the SF-MPQ.20 Those who experienced deqi sensation at a moderate level during electroacupuncture in this session were invited back 1–2 weeks later for a test–retest reliability measurement. This period was chosen because it was relatively long enough to minimise the chance of recalling the answers but short enough to minimise the probability of the subjects having undergone real change on the domain of interest.21 Similar to the first assessment session, the subjects received electroacupuncture on the right LI4 and LI11 points in the second session and were then required to quantify deqi with the C-MASS. The MASS index obtained at the first visit was compared with that obtained at the second visit for the test–retest reliability assessment.

The Short-Form McGill Pain Questionnaire

The SF-MPQ is designed to provide a brief measurement of pain.20 It is a two-factor inventory (sensory and affective) construct for measurement of pain. The SF-MPQ consists of 15 descriptors (11 sensory and 4 affective) rated on an intensity scale as 0=none, 1=mild, 2=moderate or 3=severe.

Statistical analysis

Descriptive statistics were performed to indicate the intensity of each needle sensation. Construct validity was examined by comparing the C-MASS and the SF-MPQ. It was hypothesised that items in C-MASS would be highly correlated with the sensory component but not with the affective component of the SF-MPQ. Correlation matrices were created between the items of C-MASS and SF-MPQ. The data were compared by Pearson product-moment coefficient of correlation γ. A correlation coefficient above 0.50 to 0.75 indicates a moderate to good relationship, above 0.75 was considered as good to excellent relationship. The observed values of γ should be greater than or equal to the tabled value to achieve a significance level.21

The construct validity was further assessed by principle component analysis (PCA) of the C-MASS. A correlation matrix of all the items was created. The factor was extracted for factor analysis by two criteria: factors with eigenvalues greater than 1 and the Scree test criterion, obtained by plotting the eigenvalues against the number of factors. A factor loading of greater than 0.40 was considered significant. Orthogonal rotation using varimax rotation was used for the rotation of factors to improve the spatial structure of the variables so that distinct factors would be more visible.21 Naming of factors was based on those items that had the highest factor loading in each factor.21

For reliability, the internal consistency of the C-MASS was assessed using Cronbach's α coefficient, which was considered reliable if it approached 0.70.21 To examine the test–retest reliability, ICC two-way random model were computed. ICCs above 0.75 were indicative of good reliability whereas those between 0.5 and 0.75 were considered moderate reliability.21 The level of significance was set at 0.05 for all analyses. All of the statistical analyses were performed using SPSS V.17.0 software (SPSS, Chicago, Illinois, USA).


Phase I: assessment of content validity

The degree of agreement on the importance and relevance of each item associated with deqi sensation is shown in table 1 and table 2. It was found that the CVR of all items ranged from −0.4 to 1.0.

Table 1

Results of content validity of the prefinal Chinese version of the Massachusetts General Hospital Acupuncture Sensation Scale (‘importance’ content associated with deqi sensation) (n=11)

Table 2

Content validity of the prefinal Chinese version of the Massachusetts General Hospital Acupuncture Sensation Scale (‘relevance’ to content of deqi sensation) (n=11)

Phase II: validation of the C-MASS

Subject characteristics

A total of 30 normal healthy subjects were recruited by convenience sampling (18 men, 12 women). The mean age was 34.1 ±3.81 years. Three subjects experienced only a very low level of ‘soreness/numbness/distension/heaviness’ at the required intensity and were considered not to have achieved deqi in the first session of electroacupuncture. These subjects were excluded from the analysis as inclusion of vague and uncertain sensations from these subjects was considered inappropriate. Therefore, only 27 subjects were invited to receive the electroacupuncture stimulation again for the test–retest reliability assessment at 1–2 weeks after the first session. A comparison of the intensity of various deqi sensations during the first and second sessions is displayed in table 3. The results showed that fullness/distension, soreness, numbness, dull pain and aching scored highest among the 13 sensations. No subject added any other sensations in the blank row provided. The mean MASS index obtained in the first and second assessment sessions were 4.4 (SD=1.3) and 4.7 (SD=1.1) respectively.

Table 3

Comparison of different deqi sensations intensity between time 1 and time 2 and test–retest reliability of the C-MASS (n=27)

Validity analysis

The C-MASS showed moderate correlation with the overall SF-MPQ score with the Pearson's product moment correlation coefficient γ=0.56, p=0.001. A significant correlation was observed when the C-MASS was compared with the sensory dimension of the SF-MPQ (Pearson's product moment correlation coefficient γ=0.63, p<0.05) but not with the affective dimension of the SF-MPQ (Pearson's product moment correlation coefficient γ=−0.3, p=0.111).

For factor analysis, based on the criterion of an eigenvalue of more than 1, the data yield five factors and explained 77% of variance. Therefore, the results suggested there were 5 factors among the 13 sensations of the C-MASS (table 4). Naming of factors was based on locating the highest factor loading in the factor. Therefore, factor 1 was labelled as ‘soreness’, factor 2 was labelled as ‘heaviness’, factor 3 was labelled as ‘fullness’, factor 4 was labelled as ‘dull pain’ and factor 5 was labelled as ‘numbness’.

Table 4

Factor analysis on Massachusetts General Hospital Acupuncture Sensation Scale Chinese version (n=27)

Internal consistency and test–retest reliability measurements

The overall Cronbach's α of the C-MASS was calculated as 0.71 (table 5). For test–retest reliability, three subjects did not meet the criteria of achieving deqi sensation during the first session of electroacupuncture and were thus excluded from the test–retest reliability assessment. Based on the data of the 27 subjects, the test–retest reliability of individual items of the C-MASS was moderate to excellent (ICC3,1=0.55–0.98) and that of the MASS index was excellent (ICC3,1=0.92) (table 3).

Table 5

Item-total statistics of the Massachusetts General Hospital Acupuncture Sensation Scale Chinese version (n=27)


This study examined the measurement properties of the translated version of the MASS as an inventory of needle sensations associated with deqi. The results demonstrated that the translated version is a valid and reliable instrument for the assessment of needle sensations in Hong Kong Chinese people receiving electroacupuncture.

In the forward translation process, two translators carried out independent forward translations of the instrument to ensure that the translations could be compared, enabling detection of errors and divergent interpretation of ambiguous items in the original, thus reducing the potential bias.17 Our study also adopted two back translations that were carried out in parallel so as to ensure that the quality of the translation is literally and conceptually the same as the original one.16 17 The major discrepancy was on the translation of ‘aching’ and ‘dull pain’ as both terms are semantically and pragmatically similar in the Chinese translation. The reconciliation between the two forward translators was the conclusion that ‘aching’ and ‘dull pain’ should both be used in the Chinese version.

The convergent validity and the discriminant validity of the C-MASS were established by comparing with the SF-MPQ. The overall C-MASS score has moderate correlation with the overall score of the SF-MPQ. Since four of the items in the sensory dimension of the SF-MPQ have the same description as the MASS, a stronger correlation was therefore observed if the C-MASS was compared with the sensory dimension of the SF-MPQ (convergent validity). However, as we did not expect any affective effect from acupuncture, it is reasonable that only minimal correlation was observed when the C-MASS was compared with the affective dimension of the SF-MPQ (discriminant validity).

The original developers of the SASS conducted a PCA that supported the partition of three factors among the nine sensations.7 When PCA was performed on the C-MASS in this study, five factors were extracted under the ‘eigenvalues greater than 1’ rule. These five factors explained 77% of the total variance. Factor analysis performed in this study confirmed that the C-MASS is a five-factor scale. The factors we identified (soreness, heaviness, fullness, dull pain and numbness) are in fact similar to those reported in most TCM literature as the ‘standard’ deqi sensations perceived.3 4 7 The results of this study showed that the test–retest reliability of the C-MASS was high.

From the opinions of the content experts, most of the items in the C-MASS were found to be important and relevant in relation to the content of deqi measurement, with ‘deep pressure’, ‘sharp pain’, ‘warmth’ and ‘cold’ being the exception. These items have a low CVR. Previous studies supported that ‘pressure’ appeared to be an important characterisation of deqi,15 22 23 whereas ‘warmth and coolness’ sensations are frequently considered as important associates of health in ancient TCM literature (Suwen, Chapter 54).7 According to Suwen (Chapter 54), a patient with ‘excess’ syndrome would feel coldness under the needle when the yin qi arrives; however, a patient with ‘deficient’ syndrome would feel warmth under the needle when the yang qi arrives.7 Our study showed that our subject cohort did not experience any ‘cold’ sensation or only experienced low level of ‘warmth’ sensation during acupuncture needling. This is probably because our subjects were young and healthy and thus less sensitive to these sensations, which dominate only when there is an imbalance of qi in their health status. In view of the above, we considered deep pressure, warmth and cold sensations are important components of the deqi sensations to be included in the instrument.

However, although the original MASS includes ‘sharp pain’ as one of the descriptors, it is not considered as a sensation associated with deqi by the original authors of MASS and others.7 12 24 A recent pilot survey on the perception of deqi by Chinese and American acupuncturists also revealed that 50% participants classified ‘sharp pain’ as ‘not deqi’ and 42% believed it was harmful.24 Our study showed that with ‘sharp pain’ removed from the scale, the overall Cronbach's α only decreased by 0.02, indicating that the overall homogenicity of the scale was not affected (table 5). The test–retest reliability was still maintained excellent with the ICC3,1 increased from 0.92 to 0.93 (p<0.001). The modified version also showed moderate correlation with the overall SF-MPQ score (γ=0.53, p=0.03) and sensory dimension of the SF-MPQ (γ=0.6, p<0.001). The five-factor structure was also preserved and it explained 79.6% of the total variance. Since the validity and reliability of the modified scale was not much affected if ‘sharp pain’ was removed, we consider that it would be appropriate for the descriptor ‘sharp pain’ to be removed from the C-MASS for assessment of needle sensation associated with deqi. As the instrument now is one descriptor short of the original MASS, we thus renamed this Chinese version the ‘Modified’ Massachusetts General Hospital Acupuncture Sensation Scale, Chinese version (C-MMASS). The C-MMASS is a 12-item measure that includes soreness, aching, deep pressure, heaviness, full/distension, tingling, numbness, dull pain, warmth, cold, throbbing and the one supplementary row at the end for subjects to describe perceptions in their own words (appendix 2).

Deqi sensations are complex and highly individualised. In addition, the intensity and nature of deqi sensations are determined by many factors such as patient constitution, types of needle, acupoints, needling and allied techniques.25 In our study, we have minimised all these confounding factors by including only young healthy subjects, only one type of acupuncture needle was used, only the LI4 and LI11 points were chosen for stimulation, acupuncture performed by the same trained physiotherapist and a fixed protocol used in electroacupuncture.

Limitations of the study

There are several limitations in our study. First, we have only recruited healthy subjects aged below 40. Therefore, generalisation of the findings to other age groups of Hong Kong Chinese would not be deemed appropriate. Secondly, the present study only adopted the LI4 and LI11 acupoints, as they are the classical acupoints commonly used for investigation of deqi. It may be possible that other distinct needle sensations could be elicited if different acupoints of other meridians were stimulated. Thirdly, the present study did not use non-acupoints for comparison. Hence, we could not distinguish the sensations related to skin piercing and tissue damage from real acupuncture sensation elicited by acupoint stimulation. Lastly, manual and sham acupuncture were generally used in acupuncture studies that involved deqi sensations measurement.13 25 However, our study had not included these groups for comparison. We anticipate that subjects may rate the needle sensations differently, since manual, sham and electroacupuncture may work through different mechanisms and, thus, may affect the same individual differently in the perception of needle sensations.13

Summary points

  • The MASS is an accepted method of objectively scoring the sensation of de qi.

  • We have produced a Chinese version and then shown it to be valid and reliable.


Deqi is an important concept as it may well influence the therapeutic effect of acupuncture. Developing a valid and reliable tool to quantify needle sensations is an important step towards a better understanding of the basic mechanisms underlying the reported therapeutic effects. While there has been no standardised, valid and reliable tool to quantify the deqi, our study has established a Chinese version from the MASS and modified it. We demonstrated that this is a valid tool for assessment of needle sensation associated with deqi. This study has also shown that this Chinese version of MASS has good repeatability, producing consistent scores over a period of time, as well as good internal consistency. The five-factor structure of the translated version was also established. Further study with the involvement of manual acupuncture and sham acupuncture as study groups may confer greater applicability of the C-MMASS in quantifying deqi in acupuncture in the future.


View Abstract

Review history and Supplementary material

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

    • Web Only Data - This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval This study was conducted with the approval of the Human Subjects Ethics Committee of the Hong Kong Polytechnic University (HSEARS20080514001).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.