Keywords
patellar dislocation - quality of life - surveys and questionnaires - transcultural
adaptation - translation
Introduction
Patellar instability is an important knee condition not only because of its incidence
– in the United States, annual estimates are 29/100 thousand people in the general
population,[1] reaching 77/100 thousand people in some risk groups –[2] but also for recurring in most patients,[1] in rates ranging from 17% to 70% in selected groups.[3]
[4] In addition, it essentially affects the youngest, most active members of society,
with a peak incidence between the ages of 15 to 19 years,[3] resulting in an economic impact regardless of the proposed treatment method.[5]
Although common during sports activities,[4] atraumatic mechanisms are reported in individuals with predisposing conditions.[6] Patellar instability is sometimes accompanied by limitations regarding recreational
or sports activities, and reduced quality of life.[7]
[8] Since patellar instability is a multifactorial condition,[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9] it can be managed with several therapeutic options according to the patient's anatomical
features and individual presentation.[10]
The assessment of the therapeutic outcomes using clinical and radiographic criteria
alone may underestimate the impact of the disease on the daily life of the patient.
The health status should take into consideration the influence of the clinical condition
in different daily life, work, recreation, sports and social scenarios.[11]
[12] The assessment instruments to address the therapeutic effectiveness and impact on
the quality of life have been designed to broaden our understanding regarding health
care outcomes.[13]
Questionnaires such as the Kujala[14] and International Knee Documentation Committee Subjective Knee Evaluation Form (IDKC)[15] are already established in the literature as tools for this kind of clinical application.
While the IKDC assesses a wide variety of knee conditions, the Kujala questionnaire
is more specific to patellofemoral joint disorders, since it specifically documents
patellofemoral pain. Even so, the daily subjective limitations experienced by these
patients might not be fully understood, hindering the evaluation of the clinical interventions.[16]
[17]
Hiemstra et al.[18] developed the Banff Questionnaire for Patellar Instability[7]
[18] to assesses the quality of life of these patients in terms of symptoms and functional,
social, and economic activities. The present study aimed to translate and cross-culturally
adapt the Banff Questionnaire into Brazilian Portuguese.
Methodology
The present study was initiated after approval by the Ethics in Research Committee
under number CAAE 70103717.3.0000.5505. Data was collected from the Knee Group outpatient
clinic of our institution. Literate patients with recurrent patellar instability,
older than 12 years of age, who signed the informed consent form or had it signed
by a legal guardian were included in the present study. Patients with neurological
or systemic comorbidities were excluded from the study.
Sample size was calculated based on the number of variables analyzed, as recommended
in the literature, and set at a minimum of 62 patients, that is, twice the number
of questions contained in the Banff instrument.[19]
In total, 62 patients with recurrent patellar instability participated in the present
study. The diagnosis was established by a history of at least two episodes of patellar
dislocation observed by the patient or third parties, along with the findings of the
clinical examination and imaging tests, as described by Brattstroem.[20]
The translation and cross-cultural adaptation of the Banff Questionnaire for Patellar
Instability into Brazilian Portuguese followed the internationally accepted linguistic
validation process described by Guillemin et al.[21] and modified by Beaton et al.[22]
[23] Linguistic validation aims to generate a translation that is both equivalent to
the original text and comprehensible by the target population. The method used in
the present study is described below ([Figure 1]).
-
Translation: the translation started after obtaining the authorization to use the
questionnaire by the authors of the original article, the review of the questionnaire
items and organization of the material and the data collection flow, that is, the
“Preparation” process. The 32 questions, instructions, answer options and other items
from the original questionnaire in English were independently translated into Portuguese
by 2 Brazilian orthopedic surgeons fluent in both languages, resulting in 2 translated
texts (Banff VT1 and Banff VT2).
-
Synthesis: both versions were compared by an expert committee, resulting in “Banff
Version T12” (Banff VT12).
-
Back translation: the Banff VT12” was back translated by two native English speakers
who were also fluent in Portuguese, and blinded as to the original questionnaire and
with no knowledge of the subject. The aim of this stage was to find conceptual translation
errors and gross inconsistencies from the previous steps and to generate two “back-translated
versions” (Banff VRT1 and Banff VRT2).
-
Expert committee review: a new meeting including the third and fourth translators
and the same expert committee was held to search for inconsistencies and check correspondences
between the back-translated texts (Banff VRT1 and Banff VRT2), the initial translation
(Banff VT12), and the original questionnaire. The harmonization of discrepancies,
seeking semantic, idiomatic, experiential, and conceptual equivalence between the
texts, resulted in a “Banff Brazilian Prefinal Version” (in Portuguese, “Banff Versão
Brasileira Pré-Final”, VBPF). This text consolidated all information produced so far
in an easily understandable instrument used at the pretest with the sample from the
study.
-
Pretest: this step was carried out at the Knee Group's outpatient clinic at Hospital
São Paulo, Orthopedics and Traumatology Department, Escola Paulista de Medicina, Universidade
Federal de São Paulo (UNIFESP), using the “Banff VBPF” and a probing technique:[21] after the application of the questionnaire, each patient was individually surveyed
for clarity, understanding and acceptability of each item from the instrument. Doubts
or suggestions would require an item reformulation to be discussed with the committee;
otherwise, we would proceed to the last step.
-
Presentation of the final version to the authors of the original questionnaire: reports
prepared during all steps of the process were submitted along with discrepancies and
the committee consensus on each Banff VBPF item. The main author of the original questionnaire
approved this version with no suggestions or changes. The Banff VBPF was then renamed
“Banff Brazilian Final Version” (in Portuguese, “Banff Versão Brasileira Final”, VBF).
Fig. 1 Flowchart of the linguistic validation of the Brazilian Portuguese version of the
banff questuionnaire. Abbreviations: VT1, translated version 1; VT2, translated version
2; VT12, translated version 12; VRT1, back-translated version 1; VRT2, back-translated
version 2; VBPF, Brazilian Portuguese prefinal version; VBF, Brazilian Portuguese
final version.
Results
The pretest occurred from June 2018 to August 2019 at the aforementioned Knee Group's
outpatient clinic. The study included a total of 62 patients with a mean age of 29.2
years (standard deviation: 11.6 years), ranging from 12 to 57 years old. Most patients
(44 subjects; 70.69%) were female, whereas 18 (29%) were male. The Banff scores are
summarized in [Table 1].
Table 1
|
Banff Score
|
Mean
|
Standard deviation
|
Minimum
|
Maximum
|
95% confidence interval
|
|
General
|
30.33
|
15.7
|
4.68
|
86.5
|
26.71–33.96
|
|
Male gender
|
33.09
|
19.04
|
|
|
24.65–41.54
|
|
Female gender
|
29.19
|
14.21
|
|
|
25.27–33.11
|
During the reconciliation of VT1 and VT2, the expert committee found a total of 22
discrepant items in 32 questions, 19 discrepant items in 32 answers, and 18 discrepant
items in 21 information, title or instruction sentences from the original questionnaire.
Emphasizing the need to maintain the equivalence, as proposed by Guillemin et al.,[21] each item was individually analyzed by the committee in an attempt to reach a consensus
at the synthesis stage and maintain the features from the original version.
The VT12 questionnaire was back translated, resulting in the two English texts shown
in [Figure 1], called VRT1 and VRT2. During the harmonization, the semantic, idiomatic, experiential,
and conceptual cross-cultural equivalences[22] of each item regarding its original version were analyzed; no reformulation by the
committee was required when the concordance index among its 5 members was higher than
80%.[24]
Only one analysis was required by the expert committee. The back-translated versions
resulted in 26 discrepant items in 32 questions, 29 discrepant items in 32 answers,
and 20 discrepant items in 21 information, title, or instruction sentences. All discrepancies
were resolved, resulting in the VBPF. The committee pointed out the need for some
adaptations in the VBPF to maintain both the equivalence to the original version and
the instrument comprehension by the Brazilian population with no change in the analyzed
measures. A summary of this process is exemplified in [Table 2].
Table 2
|
Question
|
Original Text (BPII)
|
Translation (VT12)
|
Changes for VBPF
|
Comment
|
|
1a
|
“…severity of the 'giving way' episodes?”
|
“…episódios em que a rótula (patela) sai do lugar? (Gravidade)”
|
“…episódios em que a rótula sai do lugar? Qual a gravidade dos episódios de deslocamento?"
|
Expression with no idiomatic equivalent in Portuguese. To keep it understandable to
the Brazilian population and preserve its conceptual equivalence, the term “dislocation”
was avoided, using “kneecap out of place” (“rótula sai do lugar”) and “displacement”
(“deslocamento”)
|
|
3
|
“…stiffness”
|
“joelho duro”
|
“joelho duro”
|
Although the committee believed that the most appropriate semantic equivalence would
be “rigidez” (Portuguese for “stiffness”), the expression “joelho duro” (“hard knee”)
had the same conceptual equivalence but greater understanding. So, it was kept in
the VBPF
|
|
9
|
“…sudden twisting and pivoting movements or changes in direction”
|
“…movimentos de giro/rotação repentinos ou movimentos de mudanças de direção repentinos"
|
“…movimentos rotacionais ou de mudanças bruscas de direção”
|
Attempt to simplify the terms from translation VT12, maintaining its conceptual equivalence
|
|
14
|
“…knee 'giving way'"
|
“…joelho 'sair do lugar'"
|
“…rótula (patela) sair do lugar”
|
Expression with no idiomatic equivalence in Portuguese. “Joelho” (“knee”) from translation
VT12 was replaced by “rótula (patela)” (“patella”) because, according to the committee,
the translated term suggested the idea of the entire knee dislocating, instead of
the patella alone
|
|
28
|
“…to psychologically 'come to grips'?”
|
“…superar psicologicamente”
|
“…superar psicologicamente”
|
Expression with no idiomatic equivalence in Portuguese, translated into an expression
of similar conceptual equivalence, which was kept in the VBPF
|
There was no record of questions, answers or other items which were not understood
by the evaluated patients regarding the linguistic validation during the pretest.
Therefore, the VBPF was submitted, with no need for reformulation, to the author of
the original questionnaire together with a report of the adaptations made throughout
the process. The author of the original questionnaire approved this version with no
suggested modifications; as such, the VBF became the Banff Questionnaire for Patellar
Instability – Brazilian Version, available in Annex 1.
Discussion
There is an increased concern in the literature not only to improve patient satisfaction
but to develop outcome measures for a specific population or clinical condition.[25] Since patellar instability is a multifactorial condition,[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[26]
[27]
[28] it requires appropriate tools to compare different treatment strategies. [Figure 2] summarizes the instruments for the assessment of patellar instability as reported
by Hiemstra et al.[25]
Fig. 2 Instruments evaluating patellofemoral disorders, except for the Banff questionnaire.
Abbreviations: IKDC, International Knee Documentation Committee; KOOS, Knee Injury
and Osteoarthritis Outcome Score; NPI, Norwich Patellar Instability.
These data reveal that some questionnaires used for many years to assess patellofemoral
joint disorders focus on characterizing other knee conditions, often emphasizing items
not necessarily observed in this type of injury. In total, 60 of the 100 points of
the Lysholm score, for instance, which was translated and validated for the Portuguese
language,[29] measure pain and instability, making it unsuitable for an estimate of anterior pain.[14] Although widely used in patients with a previous history of patellar dislocation,
the Kujala questionnaire, “Scoring of Patellofemoral Disorders”, also translated and
culturally adapted into Portuguese,[30] has only 1 in 13 questions directed specifically at patellar instability.
The Banff Patella Instability Instrument (BPII) was first published in Canada, in
2013, by Hiemstra et al.,[18] in an attempt to fill the void[31] of the lack of a specific questionnaire to assess patellar instability in the templates
of “Patient-Reported Outcome Measures”. The 32 questions belonging to 5 different
domains were listed by a modified Ebel method, performed by an international group
of experts, to identify which specific outcome measures were most relevant for patellar
instability.[7]
Since the weight attributed to each answer is similar among different items, the final
score consists of the average value of all answers from all five domains, and a higher
final score reflects a better quality of life.[25] As such, the Banff questionnaire can assess the quality of life of patients with
patellar instability in a more comprehensive way.[7]
Even though the Norwich Patella Instability Score[32] was introduced because of a similar demand and context to the BPII to analyze outcome
measures in patients with patellar instability, it focuses on characterizing physical
symptoms generated by the clinical presentation. The weight attributed to its 19 items
obeys a complex algorithm: the maximum score for items referring to activities that
most commonly generate instability symptoms is lower, whereas activities that do not
generate these same symptoms for most patients, except those with more severe instabilities,
receive more points. Thus, a high final score indicates greater degrees of instability,
and, therefore, worse function.
Only a fraction of the patellar instability questionnaires available in the literature
was validated at any of the nine possible spheres as recommended by the Consensus-based
Standards for the Selection of Health Status Measurement Instruments (COSMIN).[33]
[34] The BPII was analyzed per several psychometric properties, including content validity,
internal consistency and reliability,[18] and construct and criterion validity.[7]
In 2016, the BPII underwent a factor analysis and item reduction,[34] resulting in the BPII 2.0, with 23 items divided into the same 5 domains as its
first version. This reduction was partly due to the fact that many patients did not
answer all the questions; in addition, it constitutes an attempt to adapt these questions
to the pediatric population. The author of the original questionnaire reported, via
e-mail, that the BPII 2.0 is associated with a smaller number of unanswered questions.
Several psychometric properties of the BPII 2.0 were tested and added to its validation
process, including a multicenter study of the validation of the BPII 2.0 to the Pedi-IKDC,[35] with moderate correlation, and a cross-cultural validation for the German language
targeting the German, Austrian and Swiss populations.[36] In addition, the BPII 2.0 is being validated into Dutch, Spanish, Finnish and French.[25]
Even after establishing a consensus, the committee had doubts in some questions from
the final Portuguese translation regarding the acceptability of certain grammatical
constructions (such as, “quanto medo”, regarding the intensity of fear) by the target
population. Nevertheless, all questions were understood by all participants during
the pretest, with no suggestions of changes. Another limitation of the present study
was to carry out only the translation and cross-cultural adaptation of the Banff Questionnaire
for Patellar Instability. Since the validation is a complex, iterative process, further
studies are required to increase the representative sample of the Brazilian population,
and our group is working on that.
Conclusion
The BPII has been successfully translated and cross-culturally adapted into Brazilian
Portuguese, enabling its application to assess the quality of life of patients with
patellar instability in Brazil.
