|Home | About | Journals | Submit | Contact Us | Français|
PFPS is one of the most frequently occurring overuse injuries affecting the lower limbs. A variety of functional and self-reported outcome measures have been used to assess clinical outcomes of patients with PFPS, however, only the Anterior Knee Pain Scale (AKPS) has been designed for PFPS patients.
We followed international recommendations to perform a cross-cultural adaptation of the AKPS. The Arabic AKPS and the Arabic RAND 36-item Health Survey were administered to 40 patients who were diagnosed with PFPS. Participants were assessed at baseline and after 2 to 3 days assessed with the Arabic AKPS only. The measurements tested were reliability, validity, and feasibility.
The Arabic AKPS showed high reliability for both temporal stability, internal consistency (Cronbach’s alpha was 0.81 for the first assessment and 0.75 for the second), excellent test-retest reliability (Intraclass Correlation Coefficients ICC=0.96; 95% confidence interval (CI): 0.93, 0.98) and good agreement (standard error of measurement SEM=1.8%). The Arabic AKPS was significantly correlated with physical components of the RAND 36-Item Health Survey (Spearman’s rho=0.69: p<0.001). No ceiling or floor effects were observed.
The Arabic AKPS is a valid and reliable tool and is comparable to the original English version and other translated versions.
Patellofemoral pain syndrome (PFPS) is one of the most frequently occurring overuse injuries affecting the lower limbs and is especially prevalent in people who are physically active [2,3]. The syndrome is manifested by either retropatellar or peripatellar pain, or both, as a result of activities that involve loading of the lower extremity when an individual walks, runs, jumps, climbs stairs and sits or kneels for a prolonged period . The major symptom of PFPS is pain  and it is usually progresses to impairment of function. Based on the fundamental theoretical framework and existing research, number of factors such as weakness of the muscles, structural as well as biochemical alterations of the lower limbs, the way an individual moves and cognitive factors contribute to the development of PFPS [6,7].
There are numerous etiologies responsible for either PFPS, with different patients displaying different underlying pathology . Some individuals can have poor patella tracking due to underlying biomechanical etiology. On the other hand, some individuals can have a normal profile of the femoral or tibial bone and manifest with tibiofemoral-patellofemoral joint anatomical features. Anterior knee pain is linked with patella tracking that occurs laterally in the femoral trochlea .
Numerous functional and patient self-reported outcome (PRO) measures had been applied in the assessment of clinical outcomes following patellar dislocation or anterior knee pain . Most of those measurements were initially designed for people with joint disorders that are non-patellafemoral. The Kujala Patellofemoral Disorder Score, also known as AKPS was particularly designed and developed for the assessment of patients having anterior knee pain as well as patellofemoral conditions . This outcome measurement was subsequently demonstrated to be reliable, valid and responsive to patients with anterior knee pain and patellar instability [10,12,13]. Since, direct translation of a questionnaire from one language to another, may not be scientifically sound for clinical and research purposes, the standard AKPS written in English must be validated and adapted for use in an Arabic speaking population. This can be achieved by translating the Patient Report Outcome (PRO) measures in Arabic, then correlating the psychometric properties of the new version against the original version . The standard AKPS is widely used globally, and has shown strong representation of psychometric and normative data patterns seen in English speaking populations . It has been translated to different cultural settings and into many languages, including Turkish , Persian , Chinese , Dutch , and Brazilian-Portuguese . Data compiled from questionnaires targeting different cultures are useful in establishing a better understanding of the instrument’s strengths and limitations. The aim of this study was to translate, develop a cross-cultural adaptation, and perform psychometric properties testing of the Arabic version of the Anterior Knee Pain Scale (AKPS) in patients with PFPS.
The cross-cultural adaptation was conducted in two major stages: translation and cross-cultural adaptation and assessment of psychometric properties. The first stage was performed according to the guidelines published for the translation and cross-cultural adaptations of health-related questionnaires [20,21] and adopted by the American Orthopedic Surgeons Association (AOSA). The second stage employed the use of quality criteria for assessing properties of the questionnaire ; this included: (1) translation, (2) synthesis, (3) back–translation, (4) expert committee review, (5) pretesting, and (6) validation.
The initial stage in the process was forward translation of the AKPS. Two independent Arabic speakers who were native and also spoke fluent English translated the AKPS, which was in English, into Arabic. One translator was aware of the questionnaire concept, while the other was not. That strategy utilized version T1 which was the conceptual translation of the outcome being measured and version T2, that was a reflection of the linguistic practice which was not only standard but also without a scholarly influence .
The authors of this study and the two translators compared and synthesized versions T1 and T2 of the instrument and then produced Arabic versions of each measurement: the initial draft of the Arabic language version, developed as T12 .
Two professional translators who spoke both Arabic and English and did not know what the instrument measured translated version T12, the initial translation of the instrument into Arabic, back into English. These back-translations were titled versions B1 and B2, and compared with the initial English versions .
A committee of three rehabilitation specialists who were bilingual in Arabic and English was established. The translators (described previously) assisted the committee members whenever the need arose. Each of the committee members independently evaluated the semantic, idiomatic, experiential and conceptual equivalence of each item on the questionnaire. During that analysis process, the members were given the original English version of the AKPS scale, the Arabic version that was forward translated and the English version that was back translated. When a nonequivalent item was identified, the committee reviewed it until a conclusion was made and the final version of the instrument was adapted for use in Arabic culture .
The adapted Arabic version of the instrument was tested for cultural equivalence. In that stage, an option labeled “not applicable” was included in every item of the Arabic version of the scale in order to recognize questions that Arabs would not understand or activities they would not perform often . The “not applicable” option was used in pretesting and was removed from the final version of the instrument. After the survey was finalized, 15 patients diagnosed with PFPS who were receiving physical therapy treatment in Prince Sultan Medical City completed the questionnaire.
Later, the patients were asked about any difficulties they encountered while completing the questionnaire, and patients participated in a discussion about items that were “not applicable” or questions the patients did not answer. To develop the final Arabic version of AKPS, a 15% upper limit was set for the number of unanswered questions and “not applicable” items .
The assessment of psychometric properties was based on the quality criteria used to assess properties of the questionnaire  The details and results of the validation study of the Arabic version of the AKPS are provided in the next sections.
Forty volunteers native Arabic speakers with PFPS were recruited from the Prince Sultan Military Medical City in Riyadh and the Prince Faisal Bin Fahad Hospital in Riyadh. They completed both Arabic versions of the AKPS and the RAND 36-Item Health Survey at baseline and the Arabic AKPS only 48 to 72 hours later. The mean ±SD age of the participants was 34.7±9.31 years. The majority of participants were males (65%, n=26), and 67.5% (n=27) reported pain in the right knee (Table 1).
All patients were diagnosed by either general practitioners or an orthopedist. Inclusion criteria were as follows: age between 18 and 45 years old with untreated PFPS and symptoms for longer than two months. A range of ages was chosen to avoid difficulties in differentiating between PFPS, late symptoms of apophysitis (Osgood-Schlatter’s disease) and early symptoms of osteoarthritis. Patients included in the study were experiencing anterior or retropatellar pain from at least two of the following activities: prolonged sitting, stair climbing, squatting, running, kneeling and hopping/jumping, with symptom onset unrelated to a traumatic incident and experienced pain on palpation of the patellar facets or a positive physical symptoms on Waldron’s test [4,24,25]. We excluded patients with other knee injuries or pathology, such as knee osteoarthritis/arthritis, previous knee injury or knee operation, patellar tendinopathy and Osgood-Schlatter’s disease.
The AKPS, which is sometimes known as the Kujala Scale , is a self-report questionaire with 13 items that are knee-specific. It documents patients’ responses about six activities such as walking, running, jumping, climbing stairs, squatting, and sitting for a long period. The AKPS also documents symptoms such as limping, inability to bear weight in the affected extremity, swelling, abnormal movement of the patellar, muscle atrophy, and limited flexion of the knees. Based on the patient’s answers, a score between zero and 100 is given, with the lowest score indicating severe pain or disability. The scoring of the scale is hierarchical, using categories such as “absence of difficulty – not able” or “absence of pain – presence of severe pain.” Some sections included scoring the distance that the patient can either walk or run without pain. The AKPS is easy to understand and administer and can be completed quickly ; the test-retest reliability is good [11,27]. The authors of the AKPS scale have demonstrated its validity [11.27] and its sensitivity has been examined by numerous authors [12,26,28] (Appendixes 1, 2.)
Another scale used in this study was the Arabic RAND 36-Item Health Survey. The instrument, a multipurpose short survey with 36 questions, has eight subscales for assessing a person’s physical and mental health. The physical component (PCS) includes: physical functioning, physical role functioning, bodily pain, and general health. The mental component (MCS) includes: vitality, social functioning, emotional role, and mental health. The score of this scale ranges from 0 to 100 (higher scores indicating better health status). It has been validated in Arabic  (Appendixes 3, 4).
Patients participating in this study signed the consent form and were briefed about the study procedures at every stage. The study was approved by the Institutional Review Board (IRB) of Loma Linda University and the Ethical Committee of the Prince Sultan Military Medical City in Riyadh, Saudi Arabia. The first session involved completing the Arabic version of both the AKPS and the RAND 36-Item Health Survey. In the event that a patient had PFPS on both limbs, the patient completed the questionnaires for the more symptomatic side [12,26]. The Arabic AKPS was given again 48 to 72 hours after the initial session to assess for test-retest reliability [12,30]. This time interval was chosen because it is not long enough for participant’s health status to be altered but long enough for participants to have forgotten the earlier responses of the initial session [12,26]. For convergent validity we hypothesized a strong and moderate correlation between both the Arabic AKPS and the physical components of the RAND 36-Item survey (physical functioning, role-physical, bodily pain, and general health) . To assess divergent validity we hypothesized a weak correlation between both the Arabic AKPS and the mental components of the RAND survey (vitality, social functioning, role-emotion, and mental health) because those measure different constructs. Finally, to assess feasibility, ceiling and floor effects were measured . The questionnaires were considered to have ceiling and floor effects if 15% of participants had the theoretical maximum or minimum total scores .
Data analysis was conducted using the Statistical Package for the Social Sciences (SPSS). A sample size of 40 patients was required for a power of 80% and an alpha of 0.5 to carry out this study. The two scales used in the study were examined for internal consistency, test-retest reliability, construct validity, and feasibility. Using the Cronbach’s alpha index, we were able to assess the internal consistency of the Arabic AKPS with values of 0.70 to 0.90 considered adequate . For test-retest reliability, interclass correlational coefficient (ICCs) and corresponding 95% confidence intervals (CIs) were calculated. ICCs that were less than 0.40 were considered poor, 0.4 to 0.7 were considered moderate, and 0.7 to 0.9 was considered substantial, while values above 0.9 were regarded as excellent . Agreement was obtained by computing the standard error of measurement (SEM) from baseline assessment data and the assessment taken 48 to 72 hours later and expressed in similar units as the instrument used [12,30]. The SEM as a percentage of the total score provides a relatively good measure of agreement and is considered very good if it is ≤5%; good if it is between >5% and ≤10%, doubtful if it between >10% and ≤20% or negative if >20% . Taking the standard deviation of differences between the scores from the two testing sessions and dividing by the square root of two yielded the SEM . To obtain construct validity, the level of association was calculated using the Spearman’s rho correlation between both the Arabic AKPS and the RAND 36-Item subscales at baseline. Correlation coefficients of ≥0.7 are recommended for same-construct instruments, while moderate correlations of ≥0.4 to ≤0.70 are acceptable . We examined the ceiling and floor effects by calculating the percentage of participants who reached the highest or lowest possible scores in any instrument . Ceiling and floor effects were confirmed to have occurred when more than 15% of all respondents obtained the lowest or highest possible score . The level of significance was set at p≤0.05.
In the process of translating the AKPS into Arabic, we did not find any linguistic, semantic, or cultural differences and any inconsistencies were well illustrated and resolved amicably by the expert committee. During pretests all questions and options on cultural equivalence were well understood and answered satisfactorily by all 15 participants.
All participants completed Arabic versions of both AKPS and the RAND 36-Item Health Survey at baseline and the Arabic AKPS only 48 to 72 hours later. The mean ±SD of total scores on the instruments at baseline and 48 to 72 hours later are provided in Table 2.
Results showed that the internal consistency of the Arabic version of AKPS, with a Cronbach α of 0.81 at baseline and 0.75 after 48 to 72 hours later. Deleting an item from the construct did not significantly change the alpha level. Values ranged from 0.75 to 0.83 when an item was deleted at baseline (Table 3).
From test-retest reliability analysis, the Arabic AKPS showed excellent reliability (ICC=0.96: 95% CI: 0.93, 0.98). Also, analysis of individual ICC values ranged between 0.59 and 0.97. The percentage of the SEM to the total score was classified as very good (Table 4).
The Arabic AKPS was significantly correlated with physical components of the RAND 36-Item Health Survey (rho=0.69, p<0.001) and RAND 36-Item subscales: physical functioning (rho=0.63), role-physical (rho=0.57) and bodily pain (rho=0.49); only the general health subscale was weak (rho=0.24). For divergent validity, the correlation with mental components of the RAND-36 was not significant (rho=0.31, p=0.055), showing a non-significant correlation with social functioning subscales (rho=0.22), role-emotional (rho=0.34) and mental health (rho=0.42), and a strong correlation with vitality subscales (rho=0.53) (Table 5).
For this analysis, responses from participants at baseline and at 42 and 72 hours after baseline were used. None of the participants obtained the highest or lowest possible score on the Arabic AKPS; therefore, no ceiling or floor effects were observed at any of the assessment times. Regarding the RAND 36-Item, we observed a ceiling and floor effect in role-physical, while a floor effect only in vitality and role-emotional (Table 6).
The purpose of this study was to translate, modify, and adapt the Anterior Knee Pain Scale (AKPS) to suit the Arab population culturally.
The study was conducted using a sample of Arab-speaking patients with anterior knee pain. Results of this study showed that the Arabic version of the AKPS exhibited tolerable levels for reliability, validity, and feasibility and could be used as a subjective and functional assessment tool for Arab-speaking individuals presenting with AKP or PFPS.
The literature suggests that if possible it is preferable to use a scale developed in another language which had its reliability previously tested than to create a new instrument; in this way, the results can be compared with other studies . Therefore, we chose to perform the cultural adaptation and validation of the Arabic AKPS in patients with patellofemoral pain syndrome (PFPS) in Saudi Arabia instead of creating a new questionnaire. There is consensus in the literature that a direct translation of a questionnaire into another language is not appropriate; for this reason we chose a translation protocol for maximum attainment of semantic, idiomatic, experiential, and conceptual correspondence between the original and the translated questionnaire.
The process of translating and customizing a questionnaire to a different cultural group is not an easy one. It requires time, knowledge, skill and experience . Certain conversational terms, idiomatic expressions, and emotional expressive terms may be challenging to translate. Whereas, reviews of literature and expert opinions are needed while formulating such tools, the importance of focus groups and patient involvement in the process of cultural adaptation of PRO cannot be underestimated . In this study, we followed the guidelines of cross-cultural adaptations reported by Beaton et al. , and psychometric properties testing reported by Terwee et al. . Translation and cross-cutural adaptaion of the AKPS was performed in five stages: translation, synthesis, back translation, expert committee review, and pretesting. The role of the expert committee was crucial in the review of all translations, making critical decisions, reaching a consensus on any discrepancy, and putting together the different versions of the questionnaire.
The new tool was reviewed and modified at each point by the investigators and subjected to an additional review by the committee members to guarantee the quality of the final translation. The Arabic version did not need major or specific modifications and changes because the signs, symptoms, and activities evaluated by the scale are common in both English and Arabic populations. Also, the translation used simple everyday words commonly used in Arabic. Even so, it remains challenging to align literal terms with dialectic ones. We observed that in the questions on “Abnormal painful kneecap (patellar) movements, Stairs, and Squatting” were not clear to all participants, so we placed an Arabic slang term (rather than a classical Arabic term) between parentheses to be clearer to participants. After the cross-cultural adaptation phase had been completed, the questionnaire still was not yet ready for use. Further tests should be conducted on the psychometric properties of the adapted questionnaire.
The most important findings of our study was that the Arabic AKPS demonstrated an excellent internal consistency, reliability, and acceptable construct validity; in addition, no ceiling or floor effects were observed in patients with anterior knee pain. Furthemore, this is the first study to translate the AKPS to Arabic and valdiate it for use in patients with anterior knee pain.
Similar to other versions, the Arabic AKPS had good internal consistency (α=0.81) [15–19]. Reliability testing is one of the most important of psychometric properties of an outcome measurement . When we examined reliability, we used 48 to 72 hour intervals between the baseline session and the second session to give patients time to forget their initial responses and for symptoms not to vary substantially [12,30]. The Arabic version of the AKPS showed excellent reliability and very good agreement (ICC=0.96, 95% CI=0.93–0.98). These findings are in line with those obtained by studies of other versions of the instrument, Turkish , Persian , Chinese , Brazilian-Portuguese  and those conducted by Bennell et al. (ICC=0.96) , Crossley et al. , and Watson et al. (ICC=0.95) . The original Kujala scale and Dutch version did not examine test-retest reliability. The variation in reliability observed among different studies may be due to length of time intervals, population differences, and the type of statistical approach used. The agreement assessed by the percentage of the SEM in relation to the total score range was rated as very good and was in agreement with findings from previous studies that used the AKPS [19,26,28] (Table 7.)
To verify the validity of the AKPS, we studied the content and construct validity: construct validity was examined by convergent and divergent validity, and content validity by ceiling and floor effects. We found a good correlation between Arabic AKPS and PCS of the RAND-36 Item subscales: physical functioning, role-physical, and bodily pain. A poor correlation was found with the general health subscale. Divergent validity was expected and observed with the MCS of the RAND-36. These findings support our hypothesis that the AKPS and the PCS measure the same construct, while the AKPS and the MCS measure a different construct . In this study, the correlation between the Arabic AKPS and the RAND-36 subscales of physical functioning, role-physical, and bodily pain were higher than that of the Persian , Chinese , and Dutch  versions. The correlation between the AKPS and the mental components of the RAND-36 were similar to the results found in other translated versions  (Table 8).
In this study, no ceiling and floor effect was seen for the Arabic version of the AKPS; therefore, the Arabic AKPS has the ability to distinguish between different patients based on their signs and symptoms. This parameter supports the reliability and responsiveness of the scale  and is comparable to other translated versions [15,17–19].
Findings from this study provide clinicians and researchers with evidence backing the use of an AKPS tool on Arabic speaking patients with PFPS by Arabic researchers in everyday clinical settings . Having reliable and standardized instruments can improve the quality of research findings and enhance the value of scientific evidence since findings can be reported in a more unified way. This allows standardized comparison of findings through systematic reviews and meta-analysis . In addition, this standardized instrument enhances the quality of pooled data from various parts of the world with dissimilar cultures. Our study was concluded with recommendations for future study. Due to time restraints we did not conduct an analysis of the responsiveness of the AKPS, which is defined as the ability of an instrument to detect important clinical changes through time . Therefore, we feel that the measurement properties of the AKPS are similar to the original version and the majority of the different versions available in the literature. We understand that evaluating a cross-culturally adapted instrument is an ongoing procedure, and believe that the present study laid the cornerstone for that process. Based on this assumption, we suggest further studies on the AKPS with the purpose of increasing its coverage and evaluating measurement properties yet unknown.
From our findings, the Arabic AKPS is sufficiently reliable, valid, and appropriate for use as a patient reported outcome measure for Arabic speaking individuals with anterior knee pain and PFPS. It is also the first validated knee outcome measure in Arabic to assess knee pathology.
The authors thank Dr. Abdulrahman Alasmari at Prince Sultan Military Medical Hospital for supporting this research.
ANTERIOR KNEE PAIN (Sheet code: __________________ )
Name: ___________________________________________ Date: _________________
Duration of symptoms: ______ years _______ months
For each question, circle the latest choice (letter), which corresponds to your knee symptoms.
Instructions for completing the questionnaire: Please answer every question. Some questions may look like others, but each one is different. Please take the time to read and answer each question carefully by filling in the bubble that best represents your response.
Patient Name: ___________________________________________________________
The study protocol was approved by IRB of Loma Linda University (# 5140007), Loma Linda, California, United States and the Ethical Committee of Prince Sultan Military Medical City, Riyadh, Kingdom of Saudi Arabia.
Source of support: Departmental sources