Simplified Chinese version of hip and knee replacement expectations surveys in patients with osteoarthritis and ankylosing spondylitis: cross-cultural adaptation, validation and reliability

Total joint arthroplasties (TJAs) have proven to be effective surgical interventions for patients with moderate to severe hip or knee disorders [1]. The number of TJA procedures has doubled over recent decades and is expected to increase over the coming two decades [2]. Total hip arthroplasty (THA) and total knee arthroplasty (TKA) are extremely effective in reducing pain and restoring function in patients with end-stage osteoarthritis (OA) and ankylosing spondylitis (AS) [3‐5].

More recently, increasing emphasis is focused on patient satisfaction and patient-reported outcomes of the surgery [6, 7]. Previous researches have indicated that fulfillment of preoperative expectations was one of the most important factors affecting satisfaction and was a large contributor to functional outcomes [8‐13]. Fulfilled expectations correlate with increased patient compliance with postoperative recommendations, follow-up care and monitoring [14‐17]. Several studies have shown that compared with the actual ability after surgery, preoperative expectations were higher [11, 18]. There was a large discrepancy between what patients expected and what they truly achieved. Additionally, more disabled patients often have higher preoperative expectations than their surgeons [19, 20]. Unrealistically high expectations may weaken the doctor-patient trust and cause patients to be discouraged after surgery [21]. Nowadays, as a consequence of the complexity of the healthcare environment, the doctor-patient relationship is increasingly tensional in China [22, 23]. Patients requiring TJA often have overly optimistic preoperative expectations, and many of them do not trust their surgeons. Thus, it is of great significance to investigate patients’ expectations before surgery. Currently there are no standard methods to assess expectations based on patient-derived measures.

Recent surveys have been developed and validated to capture patients’ expectations before THA or TKA [24‐27]. The Hospital for Special Surgery Total Hip Replacement Expectations Survey (HSS-THRES) [27] and the Hospital for Special Surgery Total Knee Replacement Expectations Survey (HSS-TKRES) [25] are patient-reported surveys assessing patients’ preoperative expectations successfully, which have also been cross-culturally adapted to Dutch [28], French [29] and German [30]. To our knowledge, the HSS-THRES and the HSS-TKRES are the first two questionnaires developed to assess preoperative expectations for patients undergoing TJA. However, a Chinese version has yet to be developed. In addition, no study has proven to be a reliable and valid assessment of the HSS-THRES for patients with AS undergoing THA.

Therefore, the aims of this study were (1) to translate the original version of the HSS-THRES and the HSS-TKRES into Chinese, and (2) to evaluate the psychometric properties of the translated surveys in patients with AS and OA. We additionally compared preoperative expectations of AS patients to those of OA patients undergoing THA.

OA is a highly prevalent disease in the older population [43]. Although the prevalence of AS is lower than OA, the amount of patients with AS is also considerable because of the huge population of China, which lead to a heavy burden. TJA is an effective way in reducing pain and improving function in patients with end-stage OA and AS. As the demand for TJA continues to grow, understanding the factors associated with the outcomes of this procedure has become even more important. Nowadays clinical surgeons are placing more emphasis on self-reported outcome assessment. But patients’ preoperative expectations have seldom been studied. Analyzing patients’ expectations may help surgeons to understand better what is important for their patients and therefore enhance patients’ satisfaction and the doctor–patient relationship. In addition, the differences of age and state between AS and OA may result in discrepancies of expectation, of which may be unrealistic and need to be educated before surgery. The HSS-THRES and the HSS-TKRES are widely used for patients undergoing TJA, but to our knowledge, these widely used tools have not been validated in a Chinese population. So it is greatly meaningful to translate the questionnaires into Chinese. The translation process was nicely developed. All patients completed the questionnaires without any difficulties, which meant the translated questionnaires had good cultural acceptability. No significant floor and ceiling effects were observed in our study.

The interval time was two weeks between the first and second test in our study. It was appropriate to assess the test-retest reliability for the reason that the interval was long enough to keep patients from completing the questionnaires from memory, and short enough to prevent dramatic changes in expectations. According to the results, both simplified Chinese surveys showed good test-retest reliability. The ICCs for the total scores (0.878, 0.893 and 0.806 for hip OA group, AS group and knee OA group, respectively) were all higher than the minimum criterion of 0.80 recommended by Fleiss et al. [40], similar to the results of validated Dutch [28] and French [29] versions. As a result of no available information regarding the reliability of the original tool, we did not compare our results with the English version. In addition, no bias between measurements of the two sessions was observed in the Bland-Altman analysis, which was in accordance with other cross-cultural adapted studies [28, 29]. Determined by Cronbach’s α, the internal consistency of the simplified Chinese surveys were all good (0.814, 0.815 and 0.808 for hip OA group, AS group and knee OA group, respectively), which was also conformed with the criterion suggested by Terwee et al. [33]. With respect to surveys applied in patients with hip OA, the Cronbach’s α (0.814) was above the original English version (0.77) [27] and the French version (0.72) [29], but slightly lower than the Dutch version (0.86) [28]. As for the knee OA group, present value (0.808) was in line with the original version (0.79) [27] and the French version (0.82) [29], but lower than the Dutch version (0.91) [28].

Construct validity was demonstrated by assessing the correlation between the SC-THRES/TKRES scores and Expectation WOMAC scores. Until now, no so-called “gold standard” can be used to optimally evaluate patients’ expectations. We employed the Expectation WOMAC for the reason that (1) it had been first presented and demonstrated by Haddad et al. [26] and had been used in other studies [28, 29]; (2) the Chinese WOMAC is proved to be reliable and valid [37, 38]. And we made a slight modification to result in the Expectation WOMAC. On the basis of the criterion, the Pearson’s correlations between the SC-THRES/TKRES and Expectation WOMAC were moderate (0.49 for hip OA, 0.541 for hip AS and 0.465 for knee OA). Despite this moderate result, we found a systematic bias in the Bland-Altman analysis. Compared to the SC-THRES/TKRES, the mean scores of the Expectation WOMAC were higher. However, the same trend was found in the Dutch and French version [28, 29]. Van et al. [28] thought that the way the Expectation WOMAC was adapted from the original WOMAC resulted in answers whereby the patients also considered the current status. However, Neuprez et al. [29] believed that the response’s categories and codification of the two measures were different. Our view was different from theirs. The subscales of Expectation WOMAC included pain, stiffness and functional disability. These were the main symptoms which patients wanted to get rid of as soon as possible. In addition to the items related to these symptoms, the THRES/TKRES also included psychological well-being, sexual activity and employment, which were not that important for patients. These sections may lower the mean scores of the THRES/TKRES.

The mean total score of AS group was higher both in the Expectation WOMAC and the SC-THRES. The majority of the mean score for an individual item revealed the same trend. It was suggested that item was of some importance to patients when the average score was ≥2 points. The majority of the average scores were over 2.5 and some were even over 3, indicating they were important to patients. Patients with AS undergoing THA were of worse functional status and had higher expectations than those with hip OA [44]. Not only hip OA patients, but also AS patients showed the highest expectations to improve walking. This was because patients in both groups were almost end-staged with severe pain and dysfunction in the hips, which limited their abilities to walk. Patients with hip OA showed higher expectations for pain relief during daytime than those with AS. This can be explained because many patients in AS group were bony ankylosis with less pain currently. The mean age of hip OA group and AS group were 63.8 ± 9.4 and 41.3 ± 12.1, respectively. The majority of patients in hip OA group were retired from work and they showed lower expectations to be employed. However, patients with AS were always younger, and they were unable to work because of their physical disabilities. They were expected to be endorsed by family and society through rehabilitation. Similarly, patients with AS had more expectations in sexual activity than those with hip OA.

There were two strengths of our study. First, all the patients completed the questionnaires in the outpatient room of our hospital to avoid missing values. Second, to our knowledge, it was the first time that the surveys were validated in patients with AS.

With no doubt, there were some limitations in the current study that should be considered. First, considering the absence of validated tools assessing preoperative expectations, studying such questionnaire remains difficult. Although the Chinese WOMAC is reliable and valid, the psychometric properties of the Chinese version Expectation WOMAC are unknown. Without a doubt, this may result in some issues. For example, patients may take this adapted WOMAC as the original and consider the current state. Taking the expectations of orthopedic surgeons as reference would be an alternative way [28]. However, this is doubtful owing to the discrepancies between the expectations of orthopedic surgeons and those of patients [19, 20]. Thus we took the closest tool as a reference to evaluate the validity in our study. Second, the sample was limited in size, which may not be representative of the whole Chinese population. It was noteworthy that the sample of AS was too small for assessing the structural validity, we conducted this assessment among the OA and AS sample together. Last but not least, we translated the surveys into Mandarin, the official language of China, but traditional Chinese languages are still widely used in several areas of China. So it will be necessary to translate the surveys into traditional Chinese languages in the future.

In summary, the HSS-THRES and HSS-TKRES have been nicely adapted into Chinese versions with good psychometric properties. As self-reported questionnaires, the SC-THRES and SC-TKRES are reliable and valid for a Chinese population with OA or AS undergoing TJAs. Therefore, we suggest that the two questionnaires can be used to investigate patients’ expectations before surgeries. And patients’ expectations should be part of preoperative assessments. Future studies would be conducted to assess the content validity, cross-cultural validity and responsiveness of these two questionnaires.