Background
Osteoarthritis (OA) is the most common joint disease and a major public health problem throughout the world [
1,
2]. Since the publication of the first studies of total knee arthroplasty (TKA) in the 1960s, TKA has developed into a reliable surgical procedure to reduce pain, restore mobility and improve the quality of life for patients with osteoarthritis of the knee [
3,
4]. Meanwhile, many questionnaires for evaluating knee joint function have appeared [
5]. In 1989, The Knee Society Clinical Rating System was developed, and it became the most prevalent method to track and assess outcomes after TKA throughout the world [
6,
7]. However, with the improvement of equipment and surgical techniques, the age range of patients receiving TKA has expanded [
3]. Besides, the OA knee patients have high expectation and want excellent knee function by TKA. As a result, the reliability, validity and responsiveness of the Knee Society Clinical Rating System have been challenged [
2,
6].
Recently, the NKSS has been designed to be completed by both physicians and patients [
6]. The objective section of the NKSS, which includes the technical outcome of the procedure on the basis of pain, range of motion (ROM), alignment and stability, is completed by the physician, and the subjective parameters of the NKSS, which include patients’ knee function, satisfaction and fulfilment of expectations, is completed by the patient [
1,
6]. The internal consistency, construct and convergent validity and reliability of the NKSS have been confirmed by Noble PC. etc [
8] using statistical analysis in the United States and Canada. The NKSS has been cross-culturally adapted into Dutch [
9] and has demonstrated satisfactory psychometric properties when applied to patients with TKA.
It is well known that TKA is the most effective intervention to improve the quality of life for patients with the terminal stage of osteoarthritis [
5]. China has the largest population in the world, and the prevalence ratio of radiographic knee OA is 42.8 % in women, 21.5 % in men. Symptomatic OA knee occurred in 15.0 % of women and 5.6 % of men in Beijing. In mainland China, rural areas have a higher incidence f symptomatic OA knee than urban regions [
10,
11]. Many assessment questionnaires have been translated into simplified Chinese versions, including the Oxford Knee Score [
12] and Visual Analogue Score [
13,
14]. However, there is no so-called “gold standard” that optimally evaluates the outcome of TKA [
15,
16]. Therefore, development of the SC-NKSS may provide more choice for doctors or physicians to evaluate clinical outcomes of patients with OA knee. The aim of this study was to translate and cross-culturally adapt the NKSS into a simplified Chinese version and to evaluate its psychometric properties, and we specially tested the reliability, validity of the SC-NKSS.
Results
Cross-cultural adaptation
“Leisure recreational activities” are always understood as activities of playing card or watching TV and so on, in mainland China, these activities always require people to sit indoor. Therefore, the expert committee reached a consensus to use “go for a walk” instead of “leisure recreational activities” in our study. The distance measure of “a block” is unpopular in mainland China, so are “Inch” and “LBS”, therefore, the expert committees reached a consensus to use “100 m” instead of “a block”, centimetre instead of “inch”,“kilogram” instead of “LBS” in our study.
Score distribution and acceptability
There were no floor or ceiling effects in the target population, which indicated a good distribution for the SC-NKSS. The average completing questionnaires time was 428.7 ± 24.2 s. All participants completed the questionnaire without any difficulties and no issues with missing or multiple responses.
Reliability
Fifty participants completed the questionnaire again. Mean score of the retest was 82.08 ± 18.01, which was similar to the former result (78.42 ± 16.07). The ICC between the two sessions was 0.92 (95 % confidence interval 0.86, 0.95), which demonstrated exceptional test-retest reliability. We also found that the Cronbach alpha coefficient was 0.90 (0.71–0.85 for each subscale), which indicated excellent internal consistency for the overall SC-NKSS (Table
2).
Table 2
Correlation of each item and total SC-NKSS scores
Syptom modified | 6.75 ± 2.61 | 0.709 | 0.639 |
Satisfaction score | 12.10 ± 4.42 | 0.820 | 0.504 |
Expectation score | 10.85 ± 1.88 | 0.777 | 0.665 |
Functional activity score | 40.95 ± 12.74 | 0.847 | 0.788 |
Validity
Pearson’s coefficients are shown in table
3. The coefficient was −0.78 (95 % CI, −0.85, −0.69) between the SC-NKSS and OKS, -0.70 (95 % CI, −0.60, −0.78) between the SC-NKSS and VAS. These data demonstrate that the SC-NKSS strongly correlated with both OKS and VAS. The correlations between the SC-NKSS and the PF, BP, GH of the SF-36 (0.74, 95 % CI, 0.63, 0.81; 0.63, 95 % CI, 0.46, 0.76; 0.51, 95 % CI, 0.28, 0.67) were strong. However, the SC-NKSS exhibited weak correlations with RP (0.46, 95 % CI, 0.32, 0.59), VT (0.25, 95 % CI, 0.05, 0.43), SF (0.38, 95 % CI, 0.17, 0.54), RE (0.24, 95 % CI, 0.03, 0.44) and MH (0.27, 95 % CI, 0.05, 0.46) of the SF-36.
Table 3
Pearson correlations among the SC-NKSS, OKS, VAS and SF-36 (n = 105)
OKS | −0.78b (−0.85,−0.69) | |
VAS | −0.70b (−0.60,−0.78) | 0.74b (0.63, 0.82) |
SF-36 |
PF | 0.74b (0.64, 0.81) | −0.69b (−0.79,−0.54) |
RP | 0.46b (0.32, 0.59) | −0.42b (−0.56,−0.26) |
BP | 0.63b (0.46, 0.76) | −0.58b (−0.71,−0.40) |
GH | 0.51b (0.28, 0.67) | −0.51b (−0.66,−0.33) |
VT | 0.25b (0.05, 0.43) | −0.33b (−0.49,−0.12) |
SF | 0.38b (0.17, 0.54) | −0.46b (−0.60,−0.27) |
RE | 0.24a (0.03, 0.44) | −0.32b (−0.50,−0.10) |
MH | 0.27b (0.05, 0.46) | −0.29b (−0.47,−0.08) |
Discussion
The Knee Society Clinical Rating System appeared in 1989 and has become the most popular method for tracking and reporting outcomes of TKA [
1,
8]. However, as technology and patient population have changed, defects in this scoring system have emerged. First, the number of young and active patients receiving TKA has increased rapidly. Patients expect better functional reconstruction and recreational activity, and they are unwilling to accept physical limitations [
25,
26]. The Knee Society Clinical Rating System was focused on objective parameters, neglecting the patients’ subjective experiences [
11,
15,
16,
27‐
29]. Second, many other questionnaires, including the OKS, VAS and SF-36, have frequently been used to evaluate outcomes of TKA, and they have received positive responses. The defects and ambiguities of the Knee Society Clinical Rating System challenge its utility and validity in TKA evaluation. Recently, the NKSS was developed and has been demonstrated to have good reliability and validity in the United States, Canada and The Netherlands. Our study has tested the reliability and validity of the simplified Chinese new Knee Society Scoring System (SC-NKSS). Despite differences in culture and lifestyle, the results were excellent compared with the English and Dutch versions.
There were no ceiling or floor effects in our study, which demonstrated that the distribution of the SC-NKSS was satisfactory. All patients in our study completed the questionnaire without any difficulties, which indicated the questionnaire had good cultural acceptability.
In our study, the interval time between the first and second test was 1 week. This was suitable to assess the test-retest reliability, because it was long enough to prevent recall but short enough to ensure that clinical changes had not occurred [
20,
21]. The high value of the ICC (0.92) indicated excellent reproducibility, similar to the Dutch version of the NKSS [
9]. The high Cronbach’s alpha coefficient (0.71–0.85) indicated a good-to-excellent internal consistency between different domains of the SC-NKSS, which is comparable with the original English version (0.68–0.95) [
3] and Dutch version (0.84–0.96) [
2].
The OKS was designed to evaluate the pain and physical function of patients who had undergone TKA, and the VAS objectively reflects the patients’ subjective perception of pain. Previous studies have demonstrated the validity and reliability of the simplified Chinese versions of the OKS and VAS [
12‐
14]; therefore, the OKS and VAS scales were used as comparative criteria. In our study, we found the Pearson’s correlation coefficient was −0.78 between the SC-NKSS and OKS, −0.70 between the SC-NKSS and VAS. The strong correlation among the SC-NKSS, OKS and VAS indicated good construct validity for the SC-NKSS, similar to the English version of the NKSS [
8].
The construct validity was also assessed by comparing the SC-NKSS with the SF-36 subscale. The results demonstrated that correlations between the SC-NKSS and the PF, BP and GH of SF-36 were significant. As expected, the correlations between the SC-NKSS and the RP, VT, SF, RE and MH were weak. These results are similar to the construct validity of NKSS of the Dutch version [
9], demonstrating that the SC-NKSS has appropriate construct validity.
We also notice that there are some limitations in our study. First, we did not examine the responsiveness of SC-NKSS, therefore, the SC-NKSS couldn’t completely reflect improvement of patients who receivedTKA, and we plan to address it in a subsequent prospective study. Second, all patients in the sample had severe knee osteoarthritis and were awaiting TKA. Patients with mild to moderate knee osteoarthritis were not included in the study, which might result in a lower score. Third, most of our patients were from the south of China, which may not adequately represent the whole Chinese population, and we plan to make a multiple centre study in the subsequent study.
Conclusion
The English version of the NKSS has been translated into a simplified Chinese version and has been shown to be reliable, valid and internally consistent. The SC-NKSS is easy to understand and complete. Our results suggest that the SC-NKSS is a good method for the surgeon to assess the expectations, satisfaction and physical activities of patients before and after TKA. However, its validity and reliability need further research with a larger population.
Competing interests
The authors declare no conflict of interest. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
Authors’ contribution
LD, HX and ZW contributed equally to this work as co-first authors. LD participated in translating English version of NKSS into Chinese and drafted the manuscript, HX and ZW was in charge of statistical analysis. WW and LD were in charge of collecting clinical data of patients. ZY was in charge of collecting radiology data and evaluating radiological change of knee. All authors read and approved the final manuscript.