332  |  may 2013  |  volume 43  |  number 5  |  journal of orthopaedic & sports physical therapy

[ research report ]

Patellofemoral pain syndrome (PFPS) is a clinical condition described as retropatellar and/or peripatellar pain associated with activities involving the lower limbs, such as ascending or descending stairs, squatting, kneeling, or sitting for

long periods.7 PFPS is known to be mul-tifactorial,10,13 and its point prevalence ranges from 15% to 18%.3,12

Assessment of individuals with PFPS is common in clinical practice and re-search. The numeric pain rating scale (NPRS) is one of the instruments cur-rently used in Brazil for this purpose.11 Although the NPRS has already been translated and cross-culturally adapted for the Brazilian population,5 its measure-ment properties have not been tested in patients with PFPS. The Global Perceived Effect scale (GPE) assesses global impres-sion of recovery. It has also been trans-lated and cross-culturally adapted for the Brazilian population5; however, it was not tested in a sample of patients with PFPS. The Anterior Knee Pain Scale (AKPS),18 the Functional Index Questionnaire (FIQ),4 and the Pain Severity Scale (PSS) for PFPS20 are condition-specific instru-ments used to assess patients with PFPS (TABLE 1) that have not yet been translated, cross-culturally adapted, and tested for their measurement properties in a sample of Brazilian patients with PFPS.

To our knowledge, there are no data in the literature regarding which of the

TT STUDY DESIGN: Clinical measurement study.

TT OBJECTIVES: To cross-culturally adapt the Anterior Knee Pain Scale (AKPS), the Functional Index Questionnaire (FIQ), and the Pain Severity Scale (PSS) for patellofemoral pain syndrome (PFPS) into Brazilian Portuguese. This study also aimed to test the measurement properties of the AKPS, the FIQ, and the PSS, and the existing Brazilian Portuguese versions of the numeric pain rating scale (NPRS) and the Global Perceived Effect scale in a group with PFPS.

TT BACKGROUND: PFPS is a common condition. Therefore, translated, culturally adapted, and clini-metrically tested instruments for measuring PFPS are needed.

TT METHODS: The AKPS, FIQ, and PSS instru-ments were cross-culturally adapted into Brazilian Portuguese. The measurement properties of the AKPS, FIQ, PSS, NPRS, and Global Perceived Effect scale (internal consistency, ceiling and floor effects, and construct validity) were tested in 83 patients with PFPS. The reproducibility and re-sponsiveness were tested in 52 patients with PFPS in a test-retest design, with follow-up testing at 48

to 72 hours and at 4 weeks after baseline.

TT RESULTS: The AKPS, the FIQ, and the PSS yielded adequate internal consistency (Cronbach alpha ranging from .75 to .87) and excellent reli-ability (intraclass correlation coefficients [model 2,1] ranging from 0.90 to 0.97). The AKPS and the PSS yielded very good agreement (standard error of measurement, 2.9% and 3.5%, respectively). The highest correlations were observed among the AKPS, the FIQ, and the PSS (Pearson r>0.60, P<.05). No floor or ceiling effects were observed for any of the instruments. Effect sizes used for measuring internal responsiveness ranged from moderate to high for all measures. The NPRS and the AKPS were the measures with the highest external responsiveness.

TT CONCLUSION: The Brazilian Portuguese versions of the AKPS, FIQ, PSS, NPRS, and Global Perceived Effect scale have acceptable measurement properties. J Orthop Sports Phys Ther 2013;43(5):332-339. Epub 13 March 2013. doi:10.2519/jospt.2013.4228

TT KEY WORDS: anterior knee pain syndrome, knee, measurement properties, questionnaires

1Department of Physical Therapy, Sports Traumatology and Orthopedics Center, Universidade Federal de São Paulo, São Paulo, Brazil. 2São Paulo Running Injury Group, São Paulo, Brazil. 3Masters and Doctoral Programs in Physical Therapy, Universidade Cidade de São Paulo, São Paulo, Brazil. 4Musculoskeletal Division, The George Institute for Global Health, Sydney, Australia. 5Department of Health Sciences, University of Jyväskylä, Jyväskylä, Finland. This study was approved by the Research Ethics Committee of the Universidade Cidade de São Paulo. The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the manuscript. Address correspondence to Ronaldo Alves da Cunha, Department of Physical Therapy, Universidade Federal de São Paulo, Rua Estado de Israel 638, São Paulo, SP CEP 04022-001 Brazil. E-mail: ronaldoalvesdacunha@yahoo.com.br T Copyright ©2013 Journal of Orthopaedic & Sports Physical Therapy®

RONALDO ALVES DA CUNHA, PT1,2 • LEONARDO OLIVEIRA PENA COSTA, PT, PhD2-4 • LUIZ CARLOS HESPANHOL JUNIOR, PT, MS1,2

RAQUEL SIMONI PIRES, PT, PhD3 • URHO M. KUJALA, MD, PhD5 • ALEXANDRE DIAS LOPES, PT, PhD2,3

Translation, Cross-cultural Adaptation, and Clinimetric Testing of Instruments Used to Assess Patients With Patellofemoral Pain

Syndrome in the Brazilian Population

43-05 da Cunha.indd 332 4/17/2013 3:54:28 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | 333

above self-reported instruments has ac-ceptable measurement properties to be used in Brazilian patients with PFPS.1,14 Portuguese is spoken by approximately 240 million people worldwide, and the number of scientific articles published by Brazilian authors is increasing. Brazil is currently ranked 17th among the 146 most highly cited countries for scientific research.9 Growth of publications on clin-ical populations requires that self-report instruments be appropriately translated, cross-culturally adapted, and tested for their measurement properties. Therefore, the aims of this study were (1) to trans-late and cross-culturally adapt the AKPS, FIQ, and PSS into Brazilian Portuguese; and (2) to test the measurement proper-

ties of the Brazilian Portuguese versions of the AKPS, FIQ, and PSS, as well as the NPRS and GPE scale, which have already been translated and culturally adapted into Brazilian Portuguese.5,23

METHODS

This study was performed in 2 stages: (1) translation and cross-cul-tural adaptation, and (2) measure-

ment property assessment. The first stage was conducted according to published guidelines for translation and cross-cultural adaptation of health-related questionnaires,1,14 and the second stage employed the quality criteria for the mea-surement properties of questionnaires.25

All participants signed a consent form and were informed about the procedures of the study at all stages. This study was approved by the Research Ethics Com-mittee of the Universidade Cidade de São Paulo, São Paulo, Brazil.

Translation and Cross-cultural AdaptationTranslation and cross-cultural adaptation of the AKPS, FIQ, and PSS were per-formed in 5 stages: translation, synthesis, backward translation, expert committee review, and pretesting.Translation Items from the English ver-sion of the 3 instruments were initially translated into Brazilian Portuguese by 3 independent native Brazilian Portuguese

TABLE 1 Description of Instruments

Abbreviations: AKPS, Anterior Knee Pain Scale; FIQ, Functional Index Questionnaire; GPE, Global Perceived Effect scale; NPRS, numeric pain rating scale; PFPS, patellofemoral pain syndrome; PSS, Pain Severity Scale for PFPS.

Questionnaire ConstructOrigin Country (Language) Origin Population Language Versions Description

AKPS18 Disability Finland (English)18 Patients with anterior knee pain, patellar subluxation, and patellar dislocation18

Turkish,19 English (Australia)2,6

Specific questionnaire for anterior knee pain comprising 13 items with separate categories related to different levels of knee function. Categories inside each item are rated, and answers are added to result in a global index in which a score of 100 represents “no deficit” and a score of zero represents “the highest possible deficit.”18

FIQ4 Disability Canada (English)4 Patients with PFPS4 English (Australia)2,6 Multiple-choice, PFPS-specific questionnaire comprising 8 questions representing everyday life activities. There are 3 answer options for each activity: (1) unable to do (0 points), (2) can do with a problem (1 point), and (3) no difficulty (2 points). The total score is obtained by adding the points from each question; a total score of 0 indicates “complete inability to perform everyday life activities,” and a score of 16 indicates “no problems performing everyday life activities.”4

PSS20 Pain intensity during different activities

Canada (English)20 Patients with PFPS20 ... Specific scale to assess pain intensity in PFPS, comprising 10 questions on different activities. For each question, a numerical scale rates pain from zero (no pain) to 10 (the worst possible pain). The maximum score is 100. The higher the score, the higher the pain intensity.20

NPRS24 Pain intensity Australia and New Zealand (English)24

Patients with low back pain24

Portuguese (Brazil)5 An 11-point scale that assesses pain intensity, with scores ranging from zero (no pain) to 10 (strongest possible pain).5

GPE21 Global measurement of change

United States (English)21 Patients with lateral elbow tendinopathy21

Portuguese (Brazil)5 Numerical 11-point scale with scores ranging from –5 (“much worse”) to +5 (“completely recovered”) through zero (“no change”). In this study, for all measurements, participants were asked the following question: “How do you describe your PFPS today compared to the first episode?” Higher scores indicate higher clinical recovery.5

43-05 da Cunha.indd 333 4/17/2013 3:54:29 PM

334  |  may 2013  |  volume 43  |  number 5  |  journal of orthopaedic & sports physical therapy

[ research report ]speakers who were also fluent in the Eng-lish language. One of the translators was aware of the outcomes analyzed by the questionnaires (pain and function), but the other 2 translators were not. This strategy was used to obtain 1 conceptual translation of the outcome measures and 2 translations that accurately reflected the standard linguistic practice of the population not influenced by scholarly concerns.1

Synthesis These 3 versions of each ques-tionnaire were compared and synthesized by the authors of this study together with the 3 translators who produced the Bra-zilian Portuguese versions, and an initial consensual Brazilian Portuguese–lan-guage version of each instrument was developed.1

Backward Translation The initial con-sensual Brazilian Portuguese version of each instrument was back translated into English by 3 bilingual professional translators who were unaware of the purpose of the instruments. Next, these backward-translated versions were com-pared with the original English versions.1

Expert Committee A committee of 3 bilingual rehabilitation specialists was established. They were assisted by all translators whenever necessary. Each committee member independently ana-lyzed the semantic, idiomatic, experien-tial, and conceptual equivalence of each item in all 3 instruments. During this process, committee members had the original English version, the Brazilian Portuguese forward-translated version, and the English backward-translated version available. Whenever any item was identified as nonequivalent by any committee expert, it was reviewed and discussed by the committee members un-til an agreement was reached to produce a final version culturally adapted for the Brazilian population.1

Pretesting The Brazilian Portuguese–language version of each instrument adapted for the Brazilian population was tested for cultural equivalence. During this stage, a “not applicable” option was added to all Brazilian Portuguese–version

items to identify questions that the Bra-zilian population would not comprehend, as well as activities that would not be regularly performed by Brazilians.14 This option was only used during the pretest-ing stage and was not retained in the final version of the questionnaires.

At this stage, the instruments were completed by 31 patients who were re-ceiving treatment in physical therapy clinics in the city of São Paulo, Brazil and were diagnosed with PFPS by a medi-cal doctor. Exclusion criteria for these patients included history of surgery on the affected lower limb; less than 3 years of schooling; and a history of any neu-rological, musculoskeletal (other than PFPS), or systemic condition that could influence the assessment of knee pain and function. After the completion of the questionnaires, the volunteers were asked about the difficulty they might have had comprehending and answering ques-tions, and any unanswered or “not appli-cable” items were discussed. To establish the final Brazilian Portuguese–language version of all 3 instruments, an accept-able upper limit of 15% was established for the percentage of items for which difficulties were reported, including the number of unanswered and “not appli-cable” answers.1,14

Measurement Property TestingTo assess measurement properties, a group of Brazilian patients seeking care for PFPS was recruited from 6 different physical therapy clinics in São Paulo, Bra-zil. All patients had a medical diagnosis of PFPS. Exclusion criteria were the same as those used in the pretesting phase. The total sample consisted of 83 patients with PFPS who completed all instruments at baseline. From this group, 52 were invited to complete the instruments again at 48 to 72 hours from baseline and at 4 weeks from baseline, to calculate reproducibility and responsiveness, respectively.

The participants filled out the Brazil-ian Portuguese–language version of the AKPS, the FIQ, and the PSS adapted for the Brazilian population, as well as

2 other instruments previously adapted for the Brazilian population,5 the NPRS and GPE scale. The characteristics of all instruments used in this study are de-scribed in TABLE 1. The choice of the in-tervals between testing sessions (48 to 72 hours for reproducibility and 4 weeks for responsiveness) was based on a Coch-rane systematic review of patients with PFPS, which indicated that this clinical condition did not show improvement af-ter 48 to 72 hours of treatment but did exhibit improvement after 4 weeks.15 All self-report instruments were completed by patients who were receiving physi-cal therapy treatment for PFPS in 1 of 6 physical therapy clinics.

Data AnalysisThe measurement properties tested in this study for the 5 instruments were internal consistency, reproducibility (ie, reliability and agreement), construct va-lidity, ceiling and floor effects, and inter-nal and external responsiveness.Internal Consistency Internal consisten-cy (homogeneity) of the AKPS, FIQ, and PSS was measured using the Cronbach alpha index and “alpha if item deleted” statistics. Estimates ranging from .70 to .95 were considered to be adequate.25 The NPRS and the GPE scale were not tested for internal consistency, as these instru-ments are composed of a single item.Reproducibility Reproducibility was as-sessed by determining reliability (rela-tive measurement error) and agreement (absolute measurement error). Reliability was calculated with intraclass correla-tion coefficients (ICC2,1), using the scores from the baseline assessment and the follow-up assessment at 48 to 72 hours. ICCs were interpreted as poor if less than 0.40, moderate if between 0.40 and 0.75, substantial if between 0.75 and 0.90, or excellent if greater than 0.90.25 Agree-ment was measured by calculating the standard error of measurement (SEM), using data from the baseline assessment and the assessment taken 48 to 72 hours later. The SEM is expressed in the same units as those used in each instrument.

43-05 da Cunha.indd 334 4/17/2013 3:54:30 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | 335

The SEM was calculated as the standard deviation of differences between scores from the 2 testing sessions divided by the square root of 2.8 The percentage of the SEM relative to the total score range is considered an important indicator of agreement and was interpreted as very good if 5% or less, good if greater than 5% and less than or equal to 10%, doubt-ful if greater than 10% and less than or equal to 20%, or negative if greater than 20%.22

Construct Validity Construct validity was measured by calculating the level of asso-ciation, using Pearson r, for the scores ob-tained at baseline among the translated, cross-culturally adapted instruments. We also calculated the correlations among the instruments using the change scores between baseline and 48 to 72 hours on the AKPS, PSS, NPRS, and FIQ. For same-construct instruments, correla-tion coefficients equal to or greater than 0.70 are recommended; when comparing similar constructs, correlations ranging from 0.70 to 0.40 are considered to be moderate.25

Floor and Ceiling Effects Floor and ceil-ing effects were measured by calculat-ing the percentage of respondents who reached the lowest or highest possible scores in any given instrument.25 Floor or ceiling effects were considered to occur when more than 15% of all respondents obtained the lowest or highest possible total score.25

Internal Responsiveness The internal responsiveness of each instrument was measured using effect-size estimations with a corresponding 84% confidence interval (CI). Effect size was estimated using the mean differences between the scores at baseline and the 4-week follow-up, divided by the standard deviation at baseline. For direct comparisons of the effect sizes of different instruments, we calculated the 84% CI.16 Nonoverlapping 84% CIs are equivalent to a 0.05 z-score and can be used to compare the effect siz-es of different instruments.16 Effect sizes indicate the questionnaire’s sensitivity in measuring clinical changes in patients,

with a score of less than 0.20 indicating slight change, of 0.50 moderate change, and of 0.80 or greater large change.16 Only the 17 patients who had a change of at least 3 points on the GPE scale were included in this analysis.External Responsiveness The external responsiveness of the AKPS, FIQ, PSS, and NPRS questionnaires was assessed as follows: (1) by correlating the score on the GPE at 4 weeks with the change score (score at 4 weeks minus baseline score) of each instrument (only the 17 patients who had a change of at least 3 points on the GPE scale were included in the analysis), and (2) by constructing receiver-operat-ing-characteristic16 curves using the GPE dichotomized score to classify the partici-pants as “improved” or “not improved.” A GPE cutoff score of 2 points or greater was considered as improved, based on the number of subjects in each category. We used the same cutoff in a previous study in patients with acute low back pain.5 We also performed a sensitivity analysis us-ing 3 and 4 points as possible cutoffs and

observed similar results. This analysis was based on the area under the curve (AUC), and values of 0.70 or higher were considered to be responsive.25 A critical review on the use of the GPE scale pro-vided evidence of the validity of this scale in measuring change.17

RESULTS

Translation and Cross-cultural Adaptation

No semantic, linguistic, or cul-tural difficulties were encountered during the translation process of

the AKPS, FIQ, and PSS instruments. All discrepancies could be clearly eluci-dated and resolved by the expert com-mittee. During the cultural equivalence pretesting, all questions and options were answered and were determined to be satisfactorily comprehensible and applicable by all 31 participants. None selected the “not applicable” option that was added in the pretesting version of the 3 instruments. Therefore, we decided to

TABLE 2 Characteristics of the Study Participants*

Abbreviations: AKPS, Anterior Knee Pain Scale; FIQ, Functional Index Questionnaire; GPE, Global Perceived Effect; NPRS, numeric pain rating scale; PSS, Pain Severity Scale for patellofemoral pain syndrome.*Values are mean SD unless otherwise indicated.†Some patients had bilateral pain; in these cases, we asked the patients to fill out the measures based on the knee with the most severe pain.

Variable Baseline (n = 83) 48-72 h (n = 52) 4 wk (n = 52)

Gender, n (%)

Female 59 (71%) 34 (65%) 34 (65%)

Male 24 (29%) 18 (35%) 18 (35%)

Age, y 31.3 11.2 31.5 11.4 31.5 11.4

Weight, kg 69.1 10.8 69.8 11.2 69.8 11.2

Height, m 1.68 0.07 1.68 0.08 1.68 0.08

Affected knee, n (%)†

Right 30 (36%) 19 (37%) 19 (37%)

Left 16 (19%) 12 (23%) 12 (23%)

Both 37 (45%) 21 (40%) 21 (40%)

AKPS (0-100) 70.5 14.1 72.7 14.7 75.3 15.8

FIQ (0-16) 11.3 2.9 11.0 3.3 11.9 3.1

PSS (0-100) 32.1 21.5 32.7 21.7 29.9 21.3

NPRS (0-10) 4.3 2.7 4.2 2.8 3.3 2.6

GPE (–5 to +5) 0.2 2.4 0.2 2.4 1.8 2.4

43-05 da Cunha.indd 335 4/17/2013 3:54:32 PM

336  |  may 2013  |  volume 43  |  number 5  |  journal of orthopaedic & sports physical therapy

[ research report ]

use the data from these participants in the baseline clinimetric testing stage (in-ternal consistency, construct validity, and ceiling and floor effects). Consequently, the Portuguese-language versions of the AKPS, the FIQ, and the PSS did not re-quire additional modifications (ie, the numbers of items from the original Eng-lish versions were maintained in the Por-tuguese versions) and were used to test measurement properties.

Measurement Properties TestingEighty-three volunteers completed all 5 instruments at baseline. Fifty-two par-ticipants were invited to complete the instruments again 48 to 72 hours later and a third time 4 weeks later. The mean SD age of the participants at baseline was 31.3 11.2 years. The majority of participants were women (71%, n = 59), and nearly half had bilateral knee pain (45%, n = 37). The demographic charac-teristics of the sample and total scores on the instruments at baseline and at follow-ups are provided in TABLE 2.Internal Consistency We used the base-line data of 83 patients with PFPS for this analysis. The internal consistency of the Brazilian Portuguese–language versions of the AKPS, FIQ, and PSS was rated as

adequate (Cronbach alpha index ranging from .75 to .87). When assessing alpha if item deleted, no item was identified as contributing more to the construct than any other (alpha if item deleted ranged from .70 to .87). The results from the in-ternal consistency assessments for each instrument and the comparisons with internal consistency reported in other studies are presented in TABLE 3.Reproducibility We used data from 52 patients with PFPS in this analysis. The reliability analysis indicated that the AKPS, FIQ, PSS, and NPRS instruments exhibited excellent reliability, with ICC2,1 values ranging from 0.90 to 0.97. The GPE exhibited substantial reliability, with an ICC2,1 of 0.78. The percentage of

the SEM to the total score was classified as very good for the AKPS and PSS, good for the FIQ and NPRS, and doubtful for the GPE. TABLE 3 describes the reproduc-ibility of the instruments and compari-sons with previous studies.Construct Validity This analysis was per-formed using the baseline data of 83 pa-tients with PFPS. All associations among the instruments indicated that they mea-sure similar constructs (r>0.70, P<.05). The correlations among the AKPS, FIQ, and PSS were statistically significant (P<.05) and higher than 0.60 (TABLE 4). The correlation values among instru-ments for the change scores ranged from weak to moderate (TABLE 5).Floor and Ceiling Effects We used the

TABLE 3Internal Consistency (Cronbach Alpha) and Reproducibility  

(Reliability and Agreement) of the Instruments

Abbreviations: AKPS, Anterior Knee Pain Scale; FIQ, Functional Index Questionnaire; GPE, Global Perceived Effect scale; ICC, intraclass correlation coefficient; NA, not available; NPRS, numeric pain rating scale; PSS, Pain Severity Scale for patellofemoral pain syndrome.*Values in parentheses are the range of the alpha if item deleted.†Values are ICC2,1 (95% confidence interval).‡Values are standard error of measurement (percent of total score).§Values are standard error of measurement.

This Study (n = 83)* Other Studies This Study (n = 52)† Other Studies This Study (n = 52)‡ Other Studies§

AKPS (0-100) .75 (.70-.77) .84 (n = 40)19 0.95 (0.91, 0.97) ICC3,1 = 0.90 (n = 50)2 2.98 (2.9%) 4.72

ICC3,1 = 0.81 (n = 20)6 3.16

FIQ (0-16) .76 (.70-.76) … 0.90 (0.83, 0.94) ICC3,1 = 0.94 (n = 50)2 0.99 (6.2%) 1.02

ICC3,1 = 0.49 (n = 20)6 1.26

ICC = 0.48 (n = 18)4

PSS (0-100) .87 (.85-.87) … 0.97 (0.94, 0.98) Spearman rho = 0.95 (n = 24)20

3.50 (3.5%) …

NPRS (0-10) NA NA 0.92 (0.87, 0.95) … 0.75 (7.5%) …

GPE (–5 to +5) NA NA 0.78 (0.65, 0.87) … 1.10 (11%) …

Cronbach Alpha Index Reliability Agreement

TABLE 4Pearson Correlation Between   

Measurements at Baseline

Abbreviations: AKPS, Anterior Knee Pain Scale; FIQ, Functional Index Questionnaire; NPRS, numeric pain rating scale; PSS, Pain Severity Scale for patellofemoral pain syndrome.*P<.01.

AKPS FIQ PSS NPRS

AKPS 1.00 ... ... ...

FIQ 0.66* 1.00 ... ...

PSS –0.63* –0.67* 1.00 ...

NPRS –0.26 –0.26 0.37* 1.00

43-05 da Cunha.indd 336 4/17/2013 3:54:33 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | 337

answers for all completed instruments at baseline, 48 to 72 hours after baseline, and 4 weeks after baseline to perform this analysis. No respondent reached the highest or the lowest possible score on any instrument examined in this study. Therefore, no floor or ceiling effects were observed at any of the time points.Internal Responsiveness Analyses of re-sponsiveness were performed using data from 17 participants who completed the instruments at baseline and 4 weeks after baseline (TABLE 6).

The effect sizes from all measures ranged from moderate (0.60 for the FIQ) to high (–4.61 for the GPE). We observed overlaps in most of the 84% CIs, with few exceptions (GPE versus AKPS, GPE ver-sus FIQ, and GPE versus PSS), meaning

that most of these measures were equally responsive and that the GPE was more internally responsive than the AKPS, FIQ, and PSS.External Responsiveness We observed an absence of correlations between the change scores for each of the question-naires and the GPE score at 4 weeks, with the exception of the NPRS (r = –0.78, P = .01) and FIQ (r = 0.51, P = .04). With regard to the AUC, only the AKPS and the NPRS can be considered responsive (AUC, 0.70 or greater). This means that the AKPS and the NPRS are the only in-struments that are able to discriminate patients who improved from those who did not improve after 4 weeks of inter-vention. It is important to highlight that the results of the AUC are highly depen-

dent on the GPE scale, which is a subjec-tive measure, whereas the AKPS, FIQ, and PSS are more objective measures.

DISCUSSION

The Brazilian Portuguese ver-sions of the AKPS, the FIQ, and the PSS (APPENDIX, available online)

did not need any specific cross-cultural changes. These adaptations might not have been necessary because the ac-tivities evaluated by these instruments were common in both the source (Eng-lish-speaking) and target (Portuguese-speaking) populations. In addition, the questions from the original versions of these 3 instruments were described clear-ly and directly, making them easy to un-derstand and to translate into Brazilian Portuguese. The AKPS, the FIQ, and the PSS have adequate internal consistency, which is similar to the results observed in the Turkish version of the AKPS19 (TABLE

3). To our knowledge, there are no studies assessing the internal consistency of the FIQ and the PSS in the literature.

The Brazilian Portuguese versions of the AKPS, FIQ, PSS, and NPRS have ex-cellent reliability, whereas the GPE has substantial reliability. Previous studies of different AKPS versions yielded reliabili-ty results ranging from substantial6 to ex-cellent.2 Similar estimates were observed for the FIQ,2,6 the PSS,20 the NPRS,5 and the GPE.5 The differences in reliability results among studies may be explained by the different time points established for retest, differences in populations, and different statistical analyses used, but, in general, we found similar results to those reported in the literature.2,5,6,20 Even though the test-retest interval of 48 to 72 hours might allow memorization of answers by the participants, this short interval helps to ensure that treatments do not cause major changes in the clinical conditions.

The agreement measured by the per-centage of the SEM related to the total score range was rated as very good for the AKPS and the PSS and good for the FIQ

TABLE 5Pearson Correlation Between Change

Scores of the AKPS, FIQ, PSS, and NPRS

Abbreviations: AKPS, Anterior Knee Pain Scale; FIQ, Functional Index Questionnaire; NPRS, numeric pain rating scale; PSS, Pain Severity Scale for patellofemoral pain syndrome.*P<.01.

AKPS Change FIQ Change PSS Change NPRS Change

AKPS change 1.00 ... ... ...

FIQ change 0.41* 1.00 ... ...

PSS change –0.52* –0.36* 1.00 ...

NPRS change –0.40* –0.14 0.37* 1.00

TABLE 6Internal and External Responsiveness

of the Instruments

Effect Size, n = 17* Score Change Correlation, n = 17† AUC, n = 52‡

AKPS 1.96 (2.85, 1.07) 0.27 0.70

FIQ 0.60 (1.55, –0.34) 0.51§ 0.60

PSS –0.82 (0.23, –1.88) –0.33 0.67

NPRS –2.53 (–1.27, –3.79) –0.78§ 0.85

GPE –4.61 (–6.38, –2.84) NA NA

Abbreviations: AKPS, Anterior Knee Pain Scale; AUC, area under the curve; FIQ, Functional Index Questionnaire; GPE, Global Perceived Effect scale; NA, not applicable; NPRS, numeric pain rating scale; PSS, Pain Severity Scale for patellofemoral pain syndrome.*Values are effect size (84% confidence interval). These analyses included only patients who changed at least 3 points on the GPE scale.†Versus GPE at 4 weeks. Change scores were calculated as follows: 4 weeks minus baseline for AKPS, FIQ, PSS, and NPRS. These analyses included only patients who changed at least 3 points on the GPE scale.‡Improvement of 2 or greater on the GPE scale.§P<.05.

Internal Responsiveness External Responsiveness

43-05 da Cunha.indd 337 4/17/2013 3:54:34 PM

338  |  may 2013  |  volume 43  |  number 5  |  journal of orthopaedic & sports physical therapy

[ research report ]and the NPRS. These results are consis-tent with other studies that found very good agreement for the AKPS and the FIQ.2,6 The agreement of the GPE was classified as doubtful. We could not iden-tify any previous study that had assessed GPE agreement.

The AKPS and FIQ correlated mod-erately with each other, indicating ac-ceptable construct validity. The different characteristics of the questions in these 2 instruments might explain this moder-ate relationship. Correlations among the AKPS, the FIQ, and the PSS were also moderate and statistically significant. Although these instruments measure similar features, there are likely some differences in what each tool measures in patients with PFPS. The correlation between the PSS and the NPRS was low. Although both instruments assess pain intensity, the PSS assesses pain intensity related to specific activities, whereas the NPRS provides a more general assess-ment of pain intensity. Similar results were observed in previous measurement properties studies in individuals with PFPS.2,6,20

No ceiling and floor effects were de-tected for any of the measures. Therefore, the 5 instruments analyzed by this study are able to identify participants whose condition may improve or worsen, with-out hampering the reliability and respon-siveness of the instruments.25

We observed moderate to large effect sizes in the analysis of internal respon-siveness, with the GPE as the measure that had the highest effect size. Both gen-eral measures (the GPE and the NPRS) had the highest effect sizes, which may indicate that patients with PFPS who participated in the study showed general improvement instead of activity-specific improvement. Although these estimates provide evidence of good responsiveness of these measures, it is important to high-light that the analysis was performed in only 17 of 52 participants (ie, the patients who experienced substantial changes), and therefore the results have to be in-terpreted with caution.

External responsiveness, analyzed by the AUC, showed that the AKPS and the NPRS were the best instruments to distinguish between patients who had improved and those who had not. Simi-lar findings were observed in a study that measured the responsiveness of the AKPS and the visual analog pain scale, using the global rating of change (GROC) as a comparator for measuring improve-ment.6 The GPE measures clinical change from the onset of symptoms, whereas the GROC measures changes in status within a specific period of time,6 similar to the AKPS, FIQ, PSS, and NPRS instruments. Nevertheless, the comparison of respon-siveness of the AKPS and the FIQ using the GPE as a standard instrument to measure improvement in this study was similar to that of another study using the GROC scale.6

The GPE as a global measure of change has been criticized by some au-thors, because it assesses change from the onset of the condition, which may af-fect the interpretation of our results and be considered a limitation of the study. However, the GPE is the only global mea-sure of change available in Brazil5 and, therefore, the only available instrument to calculate the external responsiveness of these measures. Future studies using dif-ferent comparators, such as the GROC6 scale, may help to validate our results.

The results from this study are im-portant for clinicians and researchers because they provide robust evidence supporting the use of these measures to assess Brazilian patients with PFPS in the clinical setting and for clinical research by Brazilian researchers. Finally, they also provide evidence to support the pooling of data from Brazilian Portuguese–speak-ing and English-speaking populations in multisite clinical trials and systematic literature reviews.

CONCLUSION

The  Brazilian  Portuguese  ver-sions of the AKPS, FIQ, PSS, NPRS, and GPE instruments have accept-

able measurement properties and may be used to assess Brazilian patients with PFPS. The PSS had the highest inter-nal consistency and reliability, and the AKPS was the measure with the high-est agreement based on absolute error. In addition, the AKPS, the PSS, and the FIQ measure similar constructs; the GPE and the NPRS are the measures with the highest internal and external responsive-ness, respectively. t

KEY POINTSFINDINGS: The Brazilian Portuguese ver-sions of the AKPS, FIQ, PSS, NPRS, and GPE instruments have acceptable measurement properties and may be used to assess Brazilian patients with PFPS.IMPLICATIONS: In addition to providing a Brazilian Portuguese version of the instruments to be used for clinicians, this study adds evidence supporting the measurement properties of instruments for assessing patients with PFPS.CAUTION: All results presented in this study are only generalizable to Brazilian patients with PFPS.

REFERENCES

1. 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;25:3186-3191.

2. Bennell K, Bartrim S, Crossley K, Green S. Out-come measures in patellofemoral pain syndrome: test-retest reliability and inter-relationships. Phys Ther Sport. 2000;1:32-41.

3. Boling M, Padua D, Marshall S, Guskiewicz K, Pyne S, Beutler A. Gender differences in the incidence and prevalence of patel-lofemoral pain syndrome. Scand J Med Sci Sports. 2010;20:725-730. http://dx.doi.org/10.1111/j.1600-0838.2009.00996.x

4. Chesworth BM, Culham E, Tata GE, Peat M. Valida-tion of outcome measures in patients with patel-lofemoral syndrome. J Orthop Sports Phys Ther. 1989;10:302-308.

5. Costa LO, Maher CG, Latimer J, et al. Clinimetric testing of three self-report outcome measures for low back pain patients in Brazil: which one is the best? Spine (Phila Pa 1976). 2008;33:2459-2463. http://dx.doi.org/10.1097/BRS.0b013e3181849dbe

6. Crossley KM, Bennell KL, Cowan SM, Green S. Analysis of outcome measures for persons with

43-05 da Cunha.indd 338 4/17/2013 3:54:35 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | 339

MORE INFORMATIONWWW.JOSPT.ORG@

patellofemoral pain: which are reliable and valid? Arch Phys Med Rehabil. 2004;85:815-822.

7. Davis IS, Powers CM. Patellofemoral pain syn-drome: proximal, distal, and local factors, an international retreat, April 30-May 2, 2009, Fells Point, Baltimore, MD. J Orthop Sports Phys Ther. 2010;40:A1-A16. http://dx.doi.org/10.2519/jospt.2010.0302

8. de Vet HC, Terwee CB, Knol DL, Bouter LM. When to use agreement versus reliability measures. J Clin Epidemiol. 2006;59:1033-1039. http://dx.doi.org/10.1016/j.jclinepi.2005.10.015

9. Essential Science Indicators. The year 2007: top 20 country rankings in all fields. Available at: http://www.in-cites.com/countries/2007allfields.html. Accessed February 25, 2013.

10. Fredericson M, Powers CM. Practical manage-ment of patellofemoral pain. Clin J Sport Med. 2002;12:36-38.

11. Fukuda TY, Rossetto FM, Magalhães E, Bryk FF, Lucareli PR, de Almeida Aparecida Carvalho N. Short-term effects of hip abductors and lateral rotators strengthening in females with patel-lofemoral pain syndrome: a randomized con-trolled clinical trial. J Orthop Sports Phys Ther. 2010;40:736-742. http://dx.doi.org/10.2519/jospt.2010.3246

12. Fulkerson JP, Arendt EA. Anterior knee pain in females. Clin Orthop Relat Res. 2000:69-73.

13. Fulkerson JP, Shea KP. Disorders of patel-lofemoral alignment. J Bone Joint Surg Am.

1990;72:1424-1429. 14. Guillemin F, Bombardier C, Beaton D. Cross-

cultural adaptation of health-related quality of life measures: literature review and proposed guide-lines. J Clin Epidemiol. 1993;46:1417-1432.

15. Heintjes E, Berger MY, Bierma-Zeinstra SM, Bern-sen RM, Verhaar JA, Koes BW. Exercise therapy for patellofemoral pain syndrome. Cochrane Da-tabase Syst Rev. 2003:CD003472. http://dx.doi.org/10.1002/14651858.CD003472

16. Husted JA, Cook RJ, Farewell VT, Gladman DD. Methods for assessing responsiveness: a critical review and recommendations. J Clin Epidemiol. 2000;53:459-468.

17. Kamper SJ, Maher CG, Mackay G. Global rat-ing of change scales: a review of strengths and weaknesses and considerations for design. J Man Manip Ther. 2009;17:163-170.

18. Kujala UM, Jaakkola LH, Koskinen SK, Taimela S, Hurme M, Nelimarkka O. Scoring of patellofemoral disorders. Arthroscopy. 1993;9:159-163.

19. Kuru T, Dereli EE, Yaliman A. Validity of the Turkish version of the Kujala patellofemoral score in patel-lofemoral pain syndrome. Acta Orthop Traumatol Turc. 2010;44:152-156. http://dx.doi.org/10.3944/AOTT.2010.2252

20. Laprade JA, Culham EG. A self-administered pain severity scale for patellofemoral pain syndrome. Clin Rehabil. 2002;16:780-788. http://dx.doi.org/10.1191/0269215502cr553oa

21. Nourbakhsh MR, Fearon FJ. The effect of

oscillating-energy manual therapy on lateral epi-condylitis: a randomized, placebo-control, double-blinded study. J Hand Ther. 2008;21:4-14. http://dx.doi.org/10.1197/j.jht.2007.09.005

22. Ostelo RW, de Vet HC, Knol DL, van den Brandt PA. 24-item Roland-Morris Disability Question-naire was preferred out of six functional status questionnaires for post-lumbar disc surgery. J Clin Epidemiol. 2004;57:268-276. http://dx.doi.org/10.1016/j.jclinepi.2003.09.005

23. Peccin MS, Ciconelli R, Cohen M. Questionário específico para sintomas do joelho “Lysholm Knee Scoring Scale” – tradução e validação para a língua portuguesa. Acta Ortop Bras. 2006;14:268-272. http://dx.doi.org/10.1590/S1413-78522006000500008

24. Pengel LH, Refshauge KM, Maher CG, Nicholas MK, Herbert RD, McNair P. Physiotherapist-directed exercise, advice, or both for subacute low back pain: a randomized trial. Ann Intern Med. 2007;146:787-796.

25. Terwee CB, Bot SD, de Boer MR, et al. Quality cri-teria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol. 2007;60:34-42. http://dx.doi.org/10.1016/j.jclinepi.2006.03.012

EARN CEUs With JOSPT’s Read for Credit Program

JOSPT’s Read for Credit (RFC) program invites Journal readers to study and analyze selected JOSPT articles and successfully complete online quizzes about them for continuing education credit. To participate in the program:

1. Go to www.jospt.org and click on “Read for Credit” in the left-hand navigation column that runs throughout the site or on the link in the “Read for Credit” box in the right-hand column of the home page. 2. Choose an article to study and when ready, click “Take Exam” for that article. 3. Log in and pay for the quiz by credit card. 4. Take the quiz. 5. Evaluate the RFC experience and receive a personalized certificate of continuing education credits.

The RFC program o�ers you 2 opportunities to pass the quiz. You may review all of your answers—including the questions you missed. You receive 0.2 CEUs, or 2 contact hours, for each quiz passed. The Journal website maintains a history of the quizzes you have taken and the credits and certificates you have been awarded in the “My CEUs” section of your “My JOSPT” account.

43-05 da Cunha.indd 339 4/17/2013 3:54:36 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | B1

ESCALA PARA DOR ANTERIOR DO JOELHO (EDAJ – AKPS)Em cada questão, circule a letra que melhor descreve os atuais sintomas relacionados ao seu joelho.

1. Você caminha mancando?a. Nãob. Levemente ou de vez em quandoc. Constantemente

2. O seu joelho suporta o seu peso?a. Apóio totalmente, sem dorb. Apóio, mas sinto dorc. É impossível suportar o peso

3. Ao caminhara. Não tenho limites para caminharb. Caminho mais que 2 kmc. Caminho entre 1 e 2 kmd. Não consigo

4. Ao subir / descer escadasa. Não tenho dificuldadeb. Sinto um pouco de dor ao descerc. Sinto dor ao descer e ao subird. Não consigo

5. Ao agachara. Não tenho dificuldadeb. Sinto dor após agachamentos repetidosc. Sinto dor a cada agachamentod. Somente agacho com diminuição de meu peso (me apoiando)e. Não consigo

6. Ao correra. Não tenho dificuldadeb. Sinto dor após correr mais do que 2 kmc. Sinto dor leve desde o começod. Sinto dor intensae. Não consigo

7. Ao pular/saltara. Não tenho dificuldadeb. Tenho um pouco de dificuldadec. Sinto dor constanted. Não consigo

8. Ao sentar com os joelhos flexionados/dobrados por período prolongado

a. Não tenho dificuldadeb. Sinto dor para me manter sentado após ter realizado exercíciosc. Sinto dor constanted. A dor faz com que necessite estender (esticar) os joelhos de tempos

em tempose. Não consigo

9. Dora. Nenhumab. Leve e ocasionalc. A dor atrapalha o sonod. De vez em quando é intensae. Constante e intensa

10. Inchaço (edema)a. Nenhumb. Após esforço intensoc. Após atividades diáriasd. Toda noitee. Constante

11. Movimentos anormais (subluxação) e doloridos da rótula (patela)a. Não ocorreb. Ocorre ocasionalmente durante atividades esportivasc. Ocorre ocasionalmente durante atividades diáriasd. Já tive pelo menos um deslocamentoe. Já tive mais que dois deslocamentos

12. Atrofia da coxa (tamanho da coxa)a. Nenhuma alteração do tamanho da coxab. Leve alteração do tamanho da coxac. Severa alteração do tamanho da coxa

13. Sente dificuldade para flexionar/dobrar o joelho?a. Nenhumab. Levec. Muita

APPENDIX

BRAZILIAN PORTUGUESE VERSIONS OF THE INSTRUMENTS

43-05 da Cunha.indd 1 4/30/2013 1:39:49 PM

B2 | may 2013 | volume 43 | number 5 | journal of orthopaedic & sports physical therapy

[ research report ]

ESCALA DE INTENSIDADE DA SÍNDROME DA DOR FEMOROPATELAR (EISDF – PSS)Para cada atividade, gostaria que você desse uma nota para o quanto de dor no joelho você sentiu na semana passada numa escala de 0 a 10, onde 0 seria nenhuma dor e 10 seria pior dor possível. Caso alguma das questões não se aplique, como, por exemplo, não conseguir executar a tarefa por muita dor, ou que não esteja relacionado ao seu dia-a-dia, marcar a opção “não se aplica.”

1. Ao subir escadas

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica2. Ao agachar

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica3. Ao caminhar

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica4. Ao correr moderadamente (trotar)

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica5. Ao correr muito rápido

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica6. Ao praticar uma atividade esportiva

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica7. Ao sentar com os joelhos dobrados/flexionados (por 20 minutos)

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica8. Ao ajoelhar-se (independente da duração)

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica9. Pior dor em repouso/dormindo

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica10. Pior dor ao descansar após atividade

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível ( ) não se aplica

APPENDIX

43-05 da Cunha.indd 2 4/30/2013 1:39:49 PM

journal of orthopaedic & sports physical therapy | volume 43 | number 5 | may 2013 | B3

QUESTIONÁRIO DO ÍNDICE DE FUNÇÃO (QIF – FIQ)Você apresenta atualmente algum problema relacionado com seu joelho:( ) direito ( ) esquerdo

1. Caminhar cerca de 1.600 metrosNão consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

2. Subir dois lances de escadas (aproximadamente 16 degraus)Não consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

3. AgacharNão consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

4. AjoelharNão consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

5. Sentar por longos períodos com seus joelhos dobrados/flexionados na mesma posiçãoNão consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

6. Subir quatro lances de escada (aproximadamente 32 degraus)Não consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

7. Correr uma distância curta, cerca de 100 metros (aproximadamente a distância de um campo de futebol)Não consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

8. Caminhar por uma distância curta (cerca de um quarteirão)Não consigo ( ) Consigo com dificuldade ( ) Nenhuma dificuldade ( )

ESCALA DE AVALIAÇÃO NUMÉRICA DA DOR (NPRS)Eu gostaria que você desse uma nota para sua dor numa escala de 0 a 10, onde 0 seria nenhuma dor, e 10 seria a pior dor possível. Por favor, dê um numero para descrever sua média de dor.

0 1 2 3 4 5 6 7 8 9 10

nenhuma dor pior dor possível

ESCALA DA PERCEPÇÃO DO EFEITO GLOBAL (EPEG – GPE)Comparado quando esta dor no joelho começou, como você descreveria seu joelho nestes dias?

–5 –4 –3 –2 –1 0 1 2 3 4 5

extremamente pior sem modificação Completamente recuperada

APPENDIX

43-05 da Cunha.indd 3 4/30/2013 1:39:49 PM