|Year : 2022 | Volume
| Issue : 3 | Page : 326-330
Cross-cultural adaptation and validation of gujarati version knee injury and osteoarthritis outcome score for patellofemoral subscale for early patellofemoral osteoarthritis population
Heta Pavan Patel, M Balaganapathy
Department of Musculoskeletal Sciences, Ashok and Rita Patel Institute of Physiotherapy, Faculty of Medical Sciences, CHARUSAT, Anand, Gujarat, India
|Date of Submission||25-May-2022|
|Date of Acceptance||11-Sep-2022|
|Date of Web Publication||27-Dec-2022|
Dr. Heta Pavan Patel
Department of Musculoskeletal Sciences, Ashok & Rita Patel Institute of Physiotherapy, CHARUSAT Campus, Highway off Nadiad-Petlad Road, Changa, Anand - 388 421, Gujarat
Source of Support: None, Conflict of Interest: None
Context: The condition of knee pain is common across Gujarat state but identification of joint structure involvement is burdensome in the early cases. The earliest symptom/s patients reported early are anterior knee pain and difficulty in weight-bearing activities. By reviewing the related literature, the researcher states that these symptoms frame the involvement of patellofemoral joints. A dialect tool is not available to evaluate the symptoms of Gujarati speaking population. Knee Injury and Osteoarthritis Outcome Score for patellofemoral (KOOS-PF) scale was developed in the year 2018 using COSMIN guidelines for spotting patellofemoral disorders. It was a valid tool published in the English language. Aims: The study aimed to cross-culturally adapt and validate of KOOS-PF scale in the Gujarati language for the early patellofemoral arthritis population. Settings and Design: This validation study was carried out in six regional orthopedic hospitals. Subjects and Methods: A translation and cross-cultural adaptation process was conducted using Beaton D. 2002 guidelines. It was administered to 144 early patellofemoral arthritis patients to determine the psychometric properties. Results: A total of 144 participants with a mean age of 46.61 ± 4.31 were recruited. The internal consistency measures showed low Cronbach's alpha (α = 0.124). The participants completed the second administration of the Gujarati version scale for good test–retest reliability, r = 0.938. There were no floor and ceiling effects. The convergent validity was confirmed with a moderate negative correlation with Gujarati version modified Western Ontario and McMaster Universities Arthritis Index (r = −0.527). Factor analysis explained that the scale was loaded on three factors. Conclusions: The analysis identified deficits in reliability and convergent validity of the Gujarati version of the KOOS-PF scale suggestive of the development of a new condition-specific tool for the early cases which can be used in research and clinical practice.
Keywords: Factor analysis, Gujarati version Knee Injury and Osteoarthritis Outcome Score for patellofemoral scale, patellofemoral disorder scale, psychometric properties, reliability
|How to cite this article:|
Patel HP, Balaganapathy M. Cross-cultural adaptation and validation of gujarati version knee injury and osteoarthritis outcome score for patellofemoral subscale for early patellofemoral osteoarthritis population. J Sci Soc 2022;49:326-30
|How to cite this URL:|
Patel HP, Balaganapathy M. Cross-cultural adaptation and validation of gujarati version knee injury and osteoarthritis outcome score for patellofemoral subscale for early patellofemoral osteoarthritis population. J Sci Soc [serial online] 2022 [cited 2023 Jan 27];49:326-30. Available from: https://www.jscisociety.com/text.asp?2022/49/3/326/365189
| Introduction|| |
Patellofemoral osteoarthritis (PFOA) is a condition seen as a sequence in patients who are suffering from anterior knee pain, crepitus, and stiffness., Duncan et al. stated that symptomatic patellofemoral joint is the initial marker for the future development of joint arthritis. The 48% prevalence rate of PFOA was found in symptom-based cohorts. However, it is a difficult task to identify it at the community level because most individuals are ignoring the initial pain and that landed to difficulty in daily activities. Crossley et al. developed Knee injury and Osteoarthritis Outcome Score for patellofemoral (KOOS-PF) disorders in the English language. There were several cross-cultural adaptation versions available of this KOOS-PF scale such as in Arabic, Saudi Arabic, Italian, Persian, Polish, Portuguese, Spanish, and Turkish. The study aimed to cross-culturally adapt and evaluate the psychometric properties of the Gujarati version KOOS-PF scale.
| Subjects and Methods|| |
Ethical approval was obtained for this validation study from the institutional ethical committee. The permission for cross-cultural translation was obtained from the original developer of an instrument KOOS-PF subscale. The Gujarati version of the KOOS-PF subscale was cross-culturally adapted according to the guidelines suggested by the original developer, Beaton et al. 2000 recommended by the American Association of Orthopedic Surgeons Outcome Committee. The validation of the translated Gujarati version was performed for reliability and validity analysis.
Translation and cross-cultural adaptation process
The original version subscale consisted of 11 items, including stiffness (1 item), pain (9 items), and quality of life (1 item); each item was rated on a Likert scale of 0–4; The following were the interpretation based on the score – 0 = none, 1 = mild, 2 = moderate, 3 = severe, and 4 = very severe. The total score interpretation was 100 indicating no problems and 0 with extreme problems. Based on Beaton et al. guidelines, cross-cultural adaptation was performed.
In the first stage, forward translation was carried out by two bilingual translators as their first language is Gujarati. A written report was generated with a compilation of comments on difficulties and the rationale for the choices for problematic questions; which were raised in the second stage – the synthesis process. The third stage was performed for backward translation. The two native English speakers with no medical background were involved in the backward translation process and two English versions were produced. During the fourth stage, an expert committee meeting was executed which was among two forward and two backward translators, one methodologist, one Gujarati language expert, and two clinicians and the investigators. They reviewed all the reports to reach a consensus on discrepancies between the first two translation and synthesis reports and backward translation reports. At the end, the prefinal Gujarati version of the KOOS-PF subscale was produced in the Gujarati language. Following this, the fifth stage of pilot testing was executed with 40 Gujarati-speaking patients who were suffering from patellofemoral disorder (i.e., patellofemoral pain syndrome and PFOA) through one-to-one interactive interviews. Here, the participants were asked for clarity and relevance of the item for completing the subscale. In this prefinal validation stage, some patients suggested three local terminologies for moving knee, carrying, and recreational activities for appropriate clarity in understanding sentences. These words were again reviewed by an expert panel and then included in the final Gujarati version of the KOOS-PF subscale. This version was subsequently used for the validation phase without further changes. The sixth stage was completed by sending the final Gujarati version of the KOOS-PF subscale; it is freely available on the website http://www.koos.nu.
A total of 144 patients with early patellofemoral arthritis were recruited from the outpatient department (OPD) unit of six orthopedic hospitals in Gujarat state using the snowball sampling method from November 2020 to June 2021. The recruited patient characteristics were patients having episodes of anterior knee pain in the past year, crepitus, stiffness, and difficulty in the weight-bearing activity of daily living such as stair ambulation, squatting, kneeling, running, and occasionally sitting to stand and walk.,, Written informed consents were obtained from the patients in vernacular language. The collected data were anonymous.
Evaluation of psychometric properties
Reliability Testing of Gujarati version Knee Injury and Osteoarthritis Outcome Score for patellofemoral scale
The measurement of reliability was performed using internal consistency and test–retest reliability of the scale. The internal consistency was the average correlation of all the item scales calculated through Cronbach's alpha α. A value of 0.7–0.9 was considered a good internal consistency of the scale. The patients were called after 2 weeks in the same hospital environment and the scale was administered a second time to test the test–retest reliability. The intraclass correlation coefficient (ICC) and 95% confidence intervals (CI) were calculated using a two-way mixed effect model, absolute agreement. An ICC value ≥0.70 indicated an acceptable level of reliability. Absolute reliability was determined by the standard error of measurement (SEM) using the formula SEM = standard deviation (SD) × ✔1-ICC, where SD was the average SD of the two-session scores. The minimal detectable change was calculated using the formula 1.96 ×✔2 × SEM. The limits of agreement between the two session scores were evaluated by plotting the difference in scores during the two testing occasions against the baseline scores in the Bland–Altman graph. The limits of agreement were calculated by the 95% CI within the difference score.
Factor analysis and validity assessment of Gujarati version Knee Injury and Osteoarthritis Outcome Score for patellofemoral scale
Convergent validity was used to identify the psychological attributes of scale through an empirical process. It was evaluated by Spearman's correlation coefficient and measured the expected correlation between the pain domain of the Numerical Pain Rating Scale (NPRS) and the functional domain of modified Western Ontario and McMaster Universities Arthritis Index (mWOMAC). The correlation coefficient values were considered >0.70 – strong, between 0.70 and 0.50 – moderate, 0.50–0.35 – weak, and <0.35 – no correlation. Hypothesis for convergent validity as follows: strong correlation for pain domain of KOOS-PF and NPRS and moderate correlation coefficient for functional domain between Gujarati version of KOOS-PF and Gujarati version of mWOMAC. Moreover, a principal component analysis was performed to establish the construct validity of the items in the scale. The acceptable level of commonalities and factor loadings for items were 0.5 and an eigenvalue of more than 1 would be considered for component factors. For the data reduction, the following norms were considered: principal component analysis, varimax rotation, communalities >0.4, factor loading >0.5 (as the study sample size was more than 120), KO/MSA>0.45, anti-image correlation matrix >0.45, correlation matrix >30%, and eigenvalue >1.
| Results|| |
The data were analyzed using IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY, USA: IBM Corp. Demographic details of the study participants are shown in [Table 1].
The internal consistency was very low (Cronbach's alpha α = 0.124).
Test–retest reliability (n = 144)
The ICC value based on the total scores of the first (test) and second (retest) assessment was 0.938 (ICC 2,1; 95% CI = 0.914–0.956; P < 0.001); and for different dimensions of the scale, stiffness −0.769 (ICC 2,1; 95% CI = 0.671–0.838; P < 0.001), pain −0.928 (ICC 2,1; 95% CI = 0.892–0.951; P < 0.001), and quality of life −0.997 (ICC 2,1; 95% CI = 0.995–0.998; P < 0.001). The test means was 59.40 ± 6.31 and the retest was 59.90 ± 6.01. The Bland–Altman plot showed the difference in total scores against the mean total scores. The SEM for scale was 1.51. Calculations revealed a minimum detectable change of 4.189 points for the scale (sc. mean difference approached zero, indicating that no bias had occurred and only a few outliers were seen outside the 95% CI limits). The Bland–Altman analysis showed that the mean difference was −0.501 ± 2.94 for the scale [Figure 1].
|Figure 1: Bland–Altman plot for the agreement between test–retest measurements of the KOOS-PF, KOOS-PF: Knee Injury and Osteoarthritis Outcome Score for patellofemoral|
Click here to view
The SAM scale skewness statistic showed no floor and ceiling effect for the scale as its value was between + 1 and −1 (0.116 for the data). The minimal clinically important differences were set as one of the SDs for the scale which was 6.31.
Convergent validity for Knee Injury and Osteoarthritis Outcome Score for patellofemoral
The pain intensity score (NPRS) showed a moderate negative correlation with KOOSPF (r = −0.357; P < 0.01) and the functional disability score (WOMAC) revealed a moderate negative correlation with KOOSPF (r = −0.527; P < 0.01), showing good convergent validity.
Initially, the factorability of the 11-item KOOSPF scale was examined. Several well-recognized criteria for the factorability of a correlation were used. First, all 11 items correlated at least 0.3 with at least one other item, and the determinant was 0.029. Moreover, an inspection of the correlation matrix revealed that more than 30% of correlations were significant at the 0.01 level, suggesting reasonable factorability. Second, the Kaiser–Meyer–Olkin measure of sampling adequacy was 0.809, above the recommended value of 0.6, and Bartlett's test of sphericity was significant (χ2(55) = 492.10, P < 0.05). The diagonals of the anti-image correlation matrix were all over 0.5, supporting the inclusion of each item in the factor analysis. Finally, the commonalities were all 4 and above, further confirming that each item shared some common variance with other items.
The principal components analysis was used because the primary purpose was to identify and compute composite coping scores for the factors underlying the scale. The initial analysis considering factors with more than one eigenvalue produced a three-factor solution with 56.83% total variance. The initial eigenvalues showed that the first factor explained 25.60% of the variance, the second factor was 21.62% of the variance, and the third factor was 9.6% of the variance.
| Discussion|| |
The cross-cultural adaptation was conducted using Beaton et al. guidelines and another aim was to evaluate the reliability and the validity of the Gujarati version of KOOS-PF for the early patellofemoral arthritis population. Using this type of outcome, the measure was to assess appropriateness for clinical setup, analyze disability assessment, and evaluation of condition-specific treatment protocols.
The results of this current study indicated the Gujarati version KOOS-PF showed low internal consistency - Cronbach's alpha (α) was 0.124. However, in the previous results during cross-cultural adaptation, the internal consistency was high. The population was 51.3% patients with patellofemoral pain syndrome and 48.7% with patellofemoral arthritis which is considered an advanced stage of arthritis. This heterogeneity in the population might be the cause of Cronbach's alpha (α) 0.87. The other cause might be a low number of questions for the early cases. The ICC was performed to estimate test–retest reliability. The ICC values for the overall scale were 0.938; which was similar to the Gujarati version (0.995) and Saudi Arabic validation (>0.95) of the KOOS-PF scale. Although the time frame for the second administration was different, this may include the reason for the memory effect. In the original English version, good test–retest reliability can be seen even after 2-week administration.
The KOOS-PF subscale showed a moderate correlation with NPRS for its construct validity, similar to the results of the original study. The Saudi Arabic version was validated with VAS, which showed similar results of negative moderate correlation (r = −0.568). The functional score KOOS-PF showed a moderate negative correlation with Gujarati version mWOMAC (r = −0.527) and good convergent validity. The targeted population included the ones with osteoarthritis of the knee; so, the gold standard scale was used for validation purposes. The Spanish version was evaluated with the Kujala score which showed a strong positive correlation (r = 0.71).
The relationship between the variables was established by principal component analysis. The result showed that there was loading on a three-factor solution with a 56.83% total variance. The initial eigenvalues showed that the first factor explained 25.60% of the variance, the second factor was 21.62% of the variance, and the third factor was 9.6% of the variance. Factor analysis was not performed in the previous study due to insufficient sample size. PF 2-How often do you experience knee pain after stopping activity? PF3-How often does pain limit your activity? and PF6-pain after squatting were related to the first factor, and their 0.828, 0.766, and 0.723, respectively, are shown in [Table 2]. Pain rising from sitting (including getting out of the car)-PF4, kneeling-PF5, and running/jogging-PF8 were related to the first and the second factor and so as pain after heavy household activity-PF7 after sport and recreational activities-PF 10 were loaded to factor second and the third.
|Table 2: Factor analysis for Gujarati version of Knee Injury and Osteoarthritis Outcome Score for patellofemoral|
Click here to view
The limitations related to study participants, Kellgren & Lawrence (KL) classification grades 0, 1, and 2 addressed as early arthritis but the study population was having Grades 1 and 2 because the participants who visited OPD, were included in the present study. A large number of sample was required to perform the confirmatory factor analysis.
| Conclusion|| |
The analysis identified deficits in reliability and convergent validity of the Gujarati version of the KOOS-PF scale for early cases of patellofemoral arthritis. It also suggested developing new condition-specific tools for the early PFOA population. It can be used in research and clinical practice to discover the important subgroup of arthritis.
The authors would like to acknowledge Thangamani Ramalingam Alagappan for providing valuable suggestions in statistical analysis. We also extend our gratitude to patients who participate voluntarily in the study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Kobayashi S, Pappas E, Fransen M, Refshauge K, Simic M. The prevalence of patellofemoral osteoarthritis: A systematic review and meta-analysis. Osteoarthritis Cartilage 2016;24:1697-707.
Schiphof D, van Middelkoop M, de Klerk BM, Oei EH, Hofman A, Koes BW, et al
. Crepitus is a first indication of patellofemoral osteoarthritis and not of tibiofemoral osteoarthritis. Osteoarthritis Cartilage 2014;22:631-8.
Duncan R, Peat G, Thomas E, Hay EM, Croft P. Incidence, progression and sequence of development of radiographic knee osteoarthritis in a symptomatic population. Ann Rheum Dis 2011;70:1944-8.
Peat G, McCarney R, Croft P. Knee pain and osteoarthritis in older adults: A review of community burden and current use of primary health care. Ann Rheum Dis 2001;60:91-7.
Crossley KM, Macri EM, Cowan SM, Collins NJ, Roos EM. The patellofemoral pain and osteoarthritis subscale of the KOOS (KOOS-PF): Development and validation using the COSMIN checklist. Br J Sports Med 2018;52:1130-6.
Patel H. Cross-Cultural Translation and Validation of Gujarati version of KOOS-PF Sub Scale for Indian population. Available from: http://koos.nu/
. [Last retrieved on 2021 Nov 20].
Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976) 2000;25:3186-91.
Peat G, Duncan RC, Wood LR, Thomas E, Muller S. Clinical features of symptomatic patellofemoral joint osteoarthritis. Arthritis Res Ther 2012;14:R63.
Haj-Mirzaian A, Mohajer B, Guermazi A, Roemer FW, Zikria B, Demehri S. Kneeling as a risk factor of patellofemoral joint cartilage damage worsening: An exploratory analysis on the Osteoarthritis Initiative. Eur Radiol 2021;31:2601-9.
Rattray J, Jones MC. Essential elements of questionnaire design and development. J Clin Nurs 2007;16:234-43.
Alghadir AH, Anwer S, Iqbal A, Iqbal ZA. Test-retest reliability, validity, and minimum detectable change of visual Analog, numerical rating, and verbal rating scales for measurement of osteoarthritic knee pain. J Pain Res 2018;11:851-6.
Nambi G, Rathod P. Translation and validation of Guajarati version of WOMAC and lequesne questionnaire in subjects with knee osteoarthritis. Int J Health Sci Res 2018;8:151-57.
Ateef M. Measurement properties of the knee injury and osteoarthritis outcome score patello-femoral questionnaire in Saudi Arabians. Peer J 2020;8:e9323.
Martinez-Cano JP, Vernaza-Obando D, Chica J, Castro AM. Cross-cultural translation and validation of the Spanish version of the patellofemoral pain and osteoarthritis subscale of the KOOS (KOOS-PF). BMC Res Notes 2021;14:220.
[Table 1], [Table 2]