Objectives Expectancy has been shown to affect patients’ responses to acupuncture therapy. However, no validated measure of expectancy for acupuncture is available in the Korean language. The Acupuncture Expectancy Scale (AES), a 4-item self-reporting questionnaire, is the validated instrument for measuring expectancies regarding acupuncture therapy. We translated the AES into Korean and examined its test–retest reliability and construct validity.
Methods The AES was translated using a forward and backward translation procedure. Internal consistency was assessed in 275 participants with item-total correlations. Construct validity was also assessed by performing principal component analysis and correlating scores on the scale with the participants’ intention to receive acupuncture therapy. Test–retest reliability was assessed in 33 participants by calculating Cronbach's α and Spearman rank correlation coefficients.
Results Internal consistency was high (Cronbach's α=0.910). Principal component analysis showed that expectancy of acupuncture treatment accounted for 79.2% of the variance. The AES was positively correlated with participants’ intention to receive acupuncture treatment (r=0.695, p<0.001). Test–retest reliability in 33 of the 275 participants was adequate, with the Spearman rank correlation coefficient ranging from 0.500 to 0.737 (p<0.001).
Conclusions The Korean version of the AES is a valid and reliable instrument for measuring patients’ expectancies regarding acupuncture treatment in the Korean population.
- Primary Care
- Public Health
Statistics from Altmetric.com
Expectancy is an important variable affecting patients’ responses to psychological and medical interventions.1 In particular, response expectancy, a form of outcome expectancy defined as ‘expectations held by the individual about one's own emotional and physiological response’ related to a situation or a treatment, may induce important clinical changes.2 Response expectancy has been identified as one of the most important aspects of non-specific therapeutic effects.3 Recent literature has suggested that there is a strong relationship between expectancy and clinical outcomes in the field of acupuncture because of the strong non-specific effect of acupuncture delivery as part of a complex intervention.4–7 Many studies have shown that patient expectations influence clinical outcomes in acupuncture trials as well as brain responses of pain during acupuncture treatment.8–10 According to Bayes’ theorem, the initial or prior odds can influence the current or ‘posterior’ odds with exposure to novel information, even in acupuncture treatment.11 ,12
Assessing response expectancies regarding acupuncture is important for understanding the non-specific effects and complex social-behavioural components of the acupuncture modality in clinical settings. The Acupuncture Expectancy Scale (AES) is a simple 4-item self-reporting questionnaire that is valid and reliable for measuring expectancy regarding acupuncture therapy.13 The metric properties of the original Chinese version have been assessed and its internal consistency and test–retest reliability have been confirmed on a Chinese outpatient population.13 It has been translated into English, and the reliability, validity and responsiveness of the English version have been confirmed in a US cancer population.14 However, no outcome measure assessing patients’ expectancies regarding acupuncture therapy is available in the Korean language.
In this study we sought to determine the reliability and validity of the AES translated into Korean.
The Acupuncture Expectancy Scale (AES)
The AES is a simple scale, validated in English and Chinese, used to measure patients’ expectancies regarding acupuncture therapy. It has four items which measure the expectation of illness improvement, enhanced coping, increased vitality and symptom alleviation as the result of acupuncture treatment. The four items are graded on a 5-point Likert scale ranging from completely agree (5 points) to completely disagree (1 point).13
Scale translation and adaptation
The AES was translated using a forward and backward translation procedure. Two bilingual translators with no prior familiarity with the instrument separately translated the original scale once. They were encouraged to strive for idiomatic rather than word-for-word translation.15 The investigators (two in the field of acupuncture and one each for Korean language and cognitive psychology) then reviewed the translations to make cultural and vocabulary adaptations. Each of the investigators proposed changes and a consensus meeting was subsequently held. The reviewed version was then back-translated into English to verify that the meaning of each scale item was preserved.
A total of 275 participants between the ages of 14 and 66 were recruited. Participants who were working or majoring in the medical field were excluded. The participants received a detailed explanation of the questionnaire and written informed consent was obtained. Basic demographic data of all participants such as age and gender were acquired from the questionnaire. Information was obtained from participants concerning whether or not they had previous experience of acupuncture treatment within the past 5 years and whether they had a pain-related condition. The investigation was conducted in accordance with the guidelines of the human subjects committee of Kyung Hee University, Seoul, Republic of Korea.
Testing the scale
All participants were asked to fill out the Korean version of the AES questionnaire independently. To determine test–retest reliability, the AES was administered to 33 random subjects 5 weeks later. To evaluate the validity of the Korean version of the AES, all participants also filled out the question assessing an individual's intention to receive acupuncture treatment. The question was graded on a 5-point Likert scale ranging from completely agree (5 points) to completely disagree (1 point). We hypothesised that a higher score in the AES was positively related to an individual's intention to receive acupuncture treatment. They were asked to evaluate their intention on the wording of the question: ‘Do you want to receive acupuncture treatment when you have certain kinds of diseases in the future?’ on a 5-point Likert scale ranging from completely agree (5 points) to completely disagree (1 point). Considering possible confounding factors, we checked the participants’ experience of acupuncture treatment between the test and retest period.
The internal consistency of the Korean version of the AES was evaluated by computing item-total correlations and Cronbach's α. Spearman rank correlation coefficients were calculated to determine test–retest reliability. The paired sample t test was also implemented to complement the correlation coefficient.
To test construct validity we performed a principal component analysis to explore the potential number and characteristics of the domains. Varimax rotation was performed to present more specific response patterns, with an Eigen value >1.0 as the criterion. We also evaluated construct validity by correlating scores on the AES with the question regarding participants’ intention to receive the acupuncture treatment. Data analyses were performed using SPSS V.20.0 (Chicago, Illinois, USA); p<0.05 was considered statistically significant.
Forward and backward translation was used to translate the scale into Korean. No cultural adaptation was needed, and some minor vocabulary adaptations were made. Consensus for the Korean version of the AES was obtained after the first meeting of the four investigators. The Korean version was retranslated into English to compare it with the original English version (table 1). This procedure allowed us to confirm that the translated version and the original version explored the same dimension.
A total of 275 subjects of mean age 35.15±12.87 years (201 (73.1%) women) were enrolled in the study. One hundred and three participants (37.5%) had no previous experience of acupuncture treatment while 172 (62.5%) had previous experience of acupuncture treatments within the past 5 years; 68 participants (24.7%) had pain-related conditions.
The response patterns to the scale from all 275 participants are shown in table 2. The scores of the final instrument were in the possible range of 4–20. The mean was 11.82, with a SD of 3.22 and a median of 12.
Cronbach's α was 0.910 for the mean of the item-total and did not increase when any of the four items was removed. Each item had an item-total correlation value >0.6 and three items (AES1, 2 and 3) had correlation values >0.8 (table 3).
The Korean version of the AES was administered twice to 33 participants at an interval of 5 weeks. Test–retest reliability was above standard, and all items were not significantly different at p<0.05 (table 4). All items exhibited high Spearman rank correlation coefficients ranging from 0.500 to 0.737. Of the 33 participants on whom test–retest reliability was examined, none received acupuncture during the 5-week interval.
The four items were subjected to factor analysis to examine factorial validity using principal component extraction and Varimax rotation with an Eigen value >1.0 as the criterion. Only one factor solution explained an acceptable amount of variance (79.211%) in the response, with factor loadings of 0.819 to 0.922 (table 5).
To test construct validity a correlation analysis was conducted between the scale and the question assessing the participants’ intention to receive acupuncture treatment. The question scored by the participant was on a 5-point Likert scale from 1 (completely disagree) to 5 (completely agree). The AES was positively correlated with participants’ intentions to receive acupuncture treatment (r=0.695, p<0.001).
This study shows that the Korean version of the AES has acceptable psychometric properties for use in assessing expectancy regarding acupuncture in Korean-speaking participants, and the results can be compared with international studies using other translated versions. The reliability coefficient of Cronbach's α was 0.910 with strong internal consistency and the Spearman rank correlation coefficient ranged from 0.500 to 0.737 (p<0.001). To test its validity, we conducted a correlation analysis between the AES and participants’ intentions to receive acupuncture treatment. The score of the scale was positively correlated with subject ratings regarding participants’ intentions to choose acupuncture therapy as a healthcare option (r=0.695, p<0.001). We therefore believe that using the AES to assess response expectancy may help reveal how patients choose their healthcare options in actual clinical settings.
Complementary and alternative medicine therapies have been criticised as nothing more than placebos, and acupuncture treatment has received similar criticisms.16 Recently, acupuncture has been increasingly accused of merely being a ‘placebo effect’ or ‘expectancy effect’ and lacking any specific effects, partly because sham acupuncture often turned out to be as efficacious as true acupuncture in clinical trials.17 In addition, other studies have revealed in different ways how patients’ expectations are either influencing clinical outcomes in acupuncture trials9 or how expectancies and beliefs modulate the neuronal substrates of pain treated by acupuncture.18 It is therefore important to understand the role of the expectancy of acupuncture treatment when evaluating its efficacy in clinical trials. We believe that the Korean version of the AES would be a useful tool to assess the expectancy of acupuncture treatment in clinical trials in Korea.
Although acupuncture treatment is an acknowledged clinical method in the Korean healthcare system and is widely used in South Korea, no specific instrument assessing expectancies regarding acupuncture for the Korean-speaking population has been developed. We tested the use of a Korean version of the AES in the Korean population. Based on several previous studies that have validated translated versions of questionnaires into different languages,19–21 a forward and backward translation method was adopted. Simple translation is usually better than the development of a new instrument if the culture of the target population is similar to the original population.19 As Korea and China are both on the Asian continent and the two countries share similar cultures and thoughts regarding acupuncture therapy, cultural adaptations during scale translations from Chinese are considered to be relatively small.
Several properties of the Korean version of the AES in the present study were different from previous studies (one performed in a Chinese acupuncture patient population and the other in an American cancer patient population).13 ,14 For example, in the US study there were flooring effects in the responses to individual items with approximately 20% firmly believing acupuncture was unlikely to have any effect (mean total AES score=9.8). In contrast, Chinese participants responded that acupuncture would have significant positive effects (mean total AES score=16.4). In the present study the mean total score was 11.8 in the Korean population, indicating a medium level of expectancy between the Chinese and US populations. Consistent with previous studies of the AES, we suggested that prior acupuncture experience was associated with increased response expectancies regarding acupuncture.14 In the present study we also found that expectancies regarding acupuncture were lower in the naïve group than in the experienced group (10.7 vs 12.1). Comparing the three studies, the Chinese participants were most experienced with acupuncture (79.0% had previous acupuncture experience), followed by the Korean participants (62.5%) and the US participants (9.0%). The differences in the total AES among the three countries are therefore closely associated with the participants’ prior experience of acupuncture.
The original AES was developed in an acupuncture clinic for the complaint for which they were being treated. In contrast, the Korean version of the AES in this study was mainly originated from the general population regarding expectancy for acupuncture treatment. In the current study, 68 participants (24.7%) had pain-related conditions. In order to investigate the influence of a diverse population such as chief complaints on the AES, we further compared the AES depending on the pain-related condition. No significant differences were found between pain-free participants and those with pain (11.9±3.3 vs 11.7±2.9, t±0.522, p>0.602). Thus, it is less likely that a diversity of population such as chief complaints has an influence on the AES in the current study. As we developed the Korean version of the AES in a general population, it could be more widely used in the assessment of expectancies for acupuncture in both clinical and experimental studies.
The Acupuncture Beliefs Scale (ABS), a 36-item self-report scale, has been developed for the measurement of beliefs in the effectiveness of acupuncture treatment in English.22 It includes three significant and meaningful factors (general beliefs, scientific credibility and adverse events). Since no validated Korean version of the ABS has been developed, construct validity was conducted using the correlation between the AES and the question assessing the participants’ intention to receive acupuncture treatment. In our study the AES was developed for measurement of expectancies regarding acupuncture therapy. We believe that it would be simpler and more specific to measure expectancy for acupuncture treatment.
In conclusion, the AES was successfully translated into Korean, which proved to have a good factorial structure and psychometric properties that replicated the results of other existing versions. With this study, expectancies regarding acupuncture treatment can be assessed in Korean-speaking population, enabling cross-cultural research comparing response expectancies in Korean, Chinese and English-speaking populations in the context of acupuncture.
In clinical trials it is important to measure expectation of acupuncture.
A Korean version of the Acupuncture Expectancy Scale proved valid and reliable.
Funding This research was supported by Basic Science Research programme through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Nos 2011-0009913 and 2005-0049404).
Contributors YC and Y-JK designed the study, monitored data collection and drafted and revised the paper. Y-JK, I-SL and H-SK conducted the experiment and drafted relevant sections of the paper. HL and H-JP designed the statistical plan and analysed the data. J-JM and HL co-designed the study, monitored data collection and analysed the data.
Competing interests None.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
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.