Introduction

Traditional Chinese Medicine Body Constitution (TCMBC) originates from the Yellow Emperor’s Canon of Medicine and the theory of TCMBC was first proposed by traditional Chinese medicine (TCM) scholars in the late 1970s (Wang 2012; Wang et al. 2013; Sun et al. 2018). The TCMBC is a new division from TCM (Li et al. 2020 a). The most recognised version of TCMBC is proposed by Wang Qi, which contains nine types of body constitutions and it provides insight into personalised medicine (Wang 2005; Li et al. 2020b). The nine types of TCM body constitutions can be divided into healthy and unhealthy categories which reflect the balanced constitution and eight types of biased constitutions respectively. The TCM body constitution indicates the dynamic healthy state of an individual and can be influenced by innate and acquired factors (Low et al. 2016; Liang et al. 2020). 

The balanced constitution becomes the protective factor against disease and also a preservation factor for health (Yap et al. 2021; Zhang et al. 2021b). The eight biased TCMBC mostly became the predictor for death and morbidity of diseases such as cancer, depression, dysmenorrhea, hypertension, metabolic syndrome, coronary heart disease, diabetes mellitus, cerebrovascular disease and others (Liu et al. 2016; Zhu et al. 2017; Chong et al. 2018; Liang et al. 2020; Liao et al. 2021; Yap et al. 2021; Zhang et al. 2021b; Yap et al. 2022). Over the past decade, researchers also integrated TCMBC with genomics, proteomics and metabolomics (Wu et al. 2010; Yu et al. 2017a; Yang et al. 2019; Deng et al. 2021; Hou et al. 2021; Zhang et al. 2021a). Researchers found that the single nucleotide polymorphisms (SNPs), the adrenergic receptor beta-3 (ADRB3) gene, lipid metabolic genes (diacylglycerol acyltransferase (DGAT2), acyl-CoA synthetase (ACSL1) and ATP-binding cassette subfamily A member 1 (ABCA1) tended to be high expression among Yang-deficiency constitution and Yin-deficiency constitution; uncoupling protein 1 (UCP1) tended to be high expression among phlegm-stasis constitution in obesity patients (Yu et al. 2017a; Hou et al. 2021). Apart from that, phosphatidylcholine (PC), Complement C3, C4a/C4b, A2M and SERPINF1 are the strong predicting biomarkers for phlegm-dampness constitution in hypertension patients (Yang et al. 2019; Zhang et al. 2021a).

The TCMBC could be referred as individualisation guidance for clinical prevention, diagnosis, treatment and prognosis of the disease (Han et al. 2013; Yu et al. 2017b; Ma et al. 2018; Liu et al. 2019; Liang et al. 2020; Hou et al. 2021; Zhang et al. 2021c). Hence, the personalised TCM body constitution can be used as an approach to improve public health, prediction, alleviation and prevention of illness (Wang et al. 2013; Li et al. 2019; Chen et al. 2021). 

Constitution of Chinese Medicine Questionnaire (CCMQ) is the survey instrument developed by Wang et al. in 2005 and it contains 67 questions to measure nine types of body constitution (Wang et al. 2005). The nine types of body constitution include; (i) Balanced constitution (BC); (ii) Qi-deficiency constitution (QDC); (iii) Yang-deficiency constitution (YADC); (iv) Yin-deficiency constitution (YIDC); (v) Phlegm-dampness constitution (PDC); (vi) Damp-heat constitution (DHC); (vii) Qi-stagnation constitution (QSC); (viii) Blood-stasis constitution (BSC); and (ix) inherited-special constitution (ISC). The validity and reliability of CCMQ (the spearman correlation coefficient of the nine sub scale was 0.76-0.90, internal consistency (Cronbach’s coefficient) for each sub scale was 0.72-0.80) was first conducted with 2500 participants from Beijing population (Zhu et al. 2007). Aside from China, Hong Kong China, Korea and Japan also had validated CCMQ in their respective country, this is due to the different culture, ethics, linguistics and geographical factors (Zhu et al. 2008; Wong et al. 2013; Li et al. 2015). The validation of the English version of CCMQ also had been conducted on American and Canadian Caucasian populations who lived in Beijing between 2011-2012 (Jing et al. 2018)The standard of classification of body constitution contributes a positive effect in TCM diagnosis, predictive and treatment model in clinical practice and public health (Wang et al. 2015).

In 2020, the top five causes of death in Malaysia were ischemic heart disease, pneumonia, cerebrovascular diseases, transport accidents and cancer (Department of Statistics Malaysia 2022). The personalised TCM body constitution can help in implementing the individualised management of those diseases. English is a widely spoken and written language in Malaysia (Vollmann &  Wooi 2019). Thus, we chose to validate the CCMQ in English version because it is easy to be self-administer and can be widely applied as a TCM screening tool by the multi-ethnic populations in Malaysia. In addition, this is the first study conducted to provide empirical evidence to validate the TCMBC in Malaysia.

The objective of this study was to adapt and validate the English version of CCMQ in Malaysia to establish the evidence on the nine domains with a total of 67 questions.

materials and Methods

Study Design

The cultural adaptation and validation of the English version of CCMQ were carried out in three phases (Figure 1). Phase one was the translation and adaption phase where translating the Chinese version of the survey instrument into the English version. Phase two, a Delphi was used to review, amend and validate the translated English version. The Delphi method was chosen as it is commonly used to gain consensus among a panel of experts on a complex issue or lack of knowledge (Jorm 2015). It has also been widely applied in health and mental health research (Jorm 2015). In phase three, a survey was conducted using the instrument derived from phase two. Online questionnaires were then distributed to young adults (aged 18 to 30 years old) in Malaysia via google form. The sample size of this study was calculated using the G* Power programme based on the medium effect size and power of 80%. The minimum required sample size was 114 subjects. Confirmatory factor analysis was used to analyse the instrument from the findings obtained from phase three. This study obtained ethical approval from the Universiti Tunku Abdul Rahman Ethics Committee for research involving human subjects (U/SERC/50/2020). 

Phase One: Translation and Adaption Process

In phase one, translation of the original Chinese version of CCMQ, which originated from China into English. First, the forward translation process (Chinese to English) was conducted by one academic TCM practitioner. Then another academic TCM practitioner was appointed to identify any ambiguities and discrepancies in the words, sentences, grammar, and meaning. Lastly, the translated version was given to three independent native English speakers. All were never seen the original version of CCMQ. They were appointed to determine their understanding of the English version. 

Phase Two: Delphi Technique

In Phase Two, the Delphi method was used to obtain and synthesise the CCMQ which originated from China. Email invitations were sent to eight experts, consisting of five TCM practitioners and three academic TCM practitioners. The experts were invited to fill in their demographic characteristics, qualifications and experiences in traditional medicine practices. They are required to review the CCMQ by answering the following open-ended questions i.e. “How many domains you suggested in the CCMQ?”, “What is in the domain and questions in the CCMQ?”, “Any additional questions to the domain in the CCMQ which has not been included?”. All the sent emails were followed-up by the researchers. Experts were required to complete the first round of Delphi within two weeks.

The results from the first round were compiled and organised. Any repeated additional questions were eliminated, any suggestions on amending the word in the instrument were taken notes and amended. 

In the second round, the instrument was sent to the same experts to review and assessed their level of agreement with the questions in each main domain using a five-point Likert scale. Competency ranking on a scale of 1 (strong disagreement) to 5 (strong agreement), mean scores were above 4 (i.e. agree or strongly agree) and were within the acceptable interquartile range (IQR) (Von der Gracht 2012).

In the third round, a face-to-face discussion was conducted by the researchers with the same expert to confirm the content validity of the competency items which will be surveyed in phase two (factor analysis). The experts reconfirmed and strongly agreed the outcome.

Phase Three: Confirmatory Factor Analysis (Convergent Validity and Discriminant Validity)

Data collected were analysed using partial least square (PLS) software SMART PLS 3.0. This is for the confirmatory factor analysis to determine the pilot test instrument’s reliability and validity. The reliability and validity were tested through internal consistency reliability, indicator reliability, convergent validity and discriminant validity. The study process was shown in the diagram (Figure 1).

Funding

This research was supported by the Universiti Tunku Abdul Rahman Research Fund (UTARRF), (Project number IPSR/RMC/UTARRF/2019-C2/F01).

Ethical statement

The study was conducted according to the Declaration of the UTAR Research Ethics and Code of Conduct guidelines, Code of Practice for Research Involving Humans, and approved by the UTAR Scientific and Ethical Review Committee (SERC) (U/SERC/50/2020).

Results

During the first round of phase two study using the Delphi method, the experts agreed and supported that CCMQ was classified into nine domains with a total of 67 questions, which were similar to the original CCMQ. However, some of the questions were rephrased; Domain 1 - Balanced Constitution consisted of 8 questions (2 questions were rephrased); Domain 2 - Qi Deficiency Constitution consisted of 8 questions (3 questions were rephrased); Domain 3 - Yang Deficiency Constitution consisted of 7 questions (2 questions were rephrased); Domain 4 - Yin Deficiency Constitution consisted of 8 questions (3 questions were rephrased); Domain 5 - Phlegm Damp Constitution consisted of 8 questions (2 questions were rephrased); Domain 6 - Damp-Heat Constitution consisted of 7 questions (3 questions were rephrased); Domain 7 - Blood Stasis Constitution consisted of 7 questions;  Domain 8 - Qi Stagnation Constitution consisted of  7 questions (2 questions were rephrased); and Domain 9 - Inherited-special Constitution consisted of  7 questions (Table 1).

In the second round of the Delphi method, all the questions exceeded the agreed consensus point mean of 4.72 and interquartile range of 0.63. The consensus was agreed occurring when at least 70 % of the respondents score 3.5 or more on the five-Likert scale for each question. Thus, in this study, an interquartile range (IQR) of less than 1 meant that more than 50% of all opinions fell on a certain point on the scale; this showed that they had reached a consensus. An IQR of zero indicated a perfect consensus among panel members: the higher the IQR, the greater the dispersion of the data. The consensus point mean was shown in Table 2. As a result, the CCMQ was remained in the nine domains with a total of 67 questions (Table 2).

In the third round, the researchers met all the experts face-to-face in the final checking on the questions even though round 2 demonstrated a high consensus. All eight experts agreed upon the domains and questions. The agreed CCMQ classification was then tested in the phase three for confirmatory factor analysis.

Table 3 summarised all the assessments performed on internal consistency, indicator reliability and convergent validity. Table 3 showed all the nine domains of composite reliability (CR) value are more than 0.70, indicating CCMQ had good and satisfactory internal consistency reliability and average variance extracted (AVE) value of more than 0.5 demonstrated an acceptable convergent validity.

Table 4 showed the heterotrait-monotrait ratio of correlations (HTMT) to assess discriminant validity. The HTMT ratio value for all the domains is below 0.90, which fulfilled the criteria of discriminant validity.

Discussion

The original CCMQ was in the Chinese version and has also been translated into Japanese, Korean and English versions (Wang 2005; Zhu 2008; Li et al. 2015; Jing et al. 2018). The CCMQ was carried out for the Malaysian population. It needs to be culturally adapted and validated in the target language. This study reported the validity, reliability and feasibility of self-administered Malaysia’s English version of CCMQ which was conducted on the types of body constitution among young adults in Malaysia. All the nine domains in CCMQ demonstrated CR values of more than 0.70, AVE values of more than 0.5 and HTMT values of below 0.90. This indicated that CCMQ had good and satisfactory internal consistency reliability, convergent validity and discriminant validity, respectively. The results obtained from this study were similar to the results obtained from numerous international validations studies (Zhu et al. 2008; Wong et al. 2013; Li et al. 2015). 

There are limited validated healthcare questionnaires in Chinese medicine even though it has been playing important role in prevention and treatment in China for 5000 years (Dong 2013). To broaden the usage and application of Chinese medicine to foreigners, the English-validated version of CCMQ was validated by American and Canadian (aged 15 to 71 years old) in Beijing (Jing et al. 2018). Due to the targeted group, culture and geographical differences, our validation study’s questionnaire word had been rephrased to cope with the understanding of young adults in Malaysia. 

The current validated CCMQ can be used as a predictive medicine tool by TCM practitioners for disease diagnosis and by researchers to conduct research in syndrome differentiation and prevention medicine studies. Through the application of CCMQ, the individualised body constitution is a way to improve the health literacy of Malaysians. It could be the protective and preventive factor to recognise and manage a person’s health status (Ye et al. 2019).

The targeted sample size of 114 was achieved to ensure adequate power for convergent and discriminant validity. However, the limitation of this study is the subject recruitment is mainly focused on young adults in Malaysia. We aimed to optimise the response rate from different populations in future studies.

Conclusion

This current statistically validated English version of the CCMQ can be considered as a valid and reliable instrument and can be applied to the young adult population in Malaysia.

Acknowledgement

The authors declare no conflicts of interest. The authors would like to express sincere gratitude to the TCM experts (Mr. Pang Jit Ken, Mr. Lock Wut Kean, Ms. Kong Ho Leh, Mr. Yek Shih Cherng, Ms. Tee Jing Xuan, Dr. Teh Lai Kuan, Dr. Pang Jit Seng and Ms. Sara Low Leng Keng) to review and assess the questionnaire. Sincere thanks go to all the participants in this study.