original article

New version of the Pediatric Evaluation of Disability

Inventory (PEDI-CAT): translation, cultural adaptation to
Brazil and analyses of psychometric properties
Marisa C. Mancini1, Wendy J. Coster2, Maíra F. Amaral1,3,
Bruna S. Avelar1, Raphael Freitas1, Rosana F. Sampaio1

ABSTRACT | Background: The Pediatric Evaluation of Disability Inventory-Computer Adaptive Test (PEDI-CAT),
developed with innovative measurement methodologies, evaluates functioning of children and youth, from 0 to 21 years,
with different health conditions. It is a revision of an earlier instrument (PEDI) that has been used in national and
international clinical practice and research. It was felt to be necessary to make this new version (PEDI-CAT) available
in Brazil. Objectives: Translate and culturally adapt the PEDI-CAT to the Brazilian-Portuguese language and test its
psychometric properties. Method: This methodological study was developed through the following stages: (1) translation,
(2) synthesis, (3) back-translation, (4) revision by an expert committee, (5) testing of the pre-final version, and (6) evaluation
of the psychometric properties. The 276 translated PEDI-CAT items were divided into three age groups (0-7, 8-14, and
15-21 years). Results: The PEDI-CAT translation followed all six stages. The adaptations incorporated cultural and
socioeconomic class specificities. The PEDI-CAT/Brazil showed good indices of inter-examiner (intraclass correlation
coefficient-ICC=0.83-0.89) and test-retest (ICC=0.96-0.97) reliability, good internal consistency (0.99) and small standard
error of measurement in all three age groups (0.12-0.17). Factor analyses grouped the items from the three functional
skills domains into one factor, and items from the responsibility scale into three factors, supporting the adequacy of
these factor solutions to the conceptual structure of the instrument and the developmental model. Conclusion: The
PEDI-CAT/Brazil is a theoretically consistent, culturally appropriate, and reliable instrument. Its availability in Brazil
will contribute to the evaluation and measurement of functional outcomes from clinical interventions, longitudinal
follow-up, and rehabilitation research.
Keywords: assessment; functioning; translation, cultural adaptation; psychometric properties; rehabilitation.

BULLET POINTS

• PEDI-CAT item translation incorporated Brazilian-specific cultural adaptations.

• PEDI-CAT/Brazil has strong reliability, internal consistency and small SEM estimates.
• Factor solutions support its content adequacy to proposed conceptual structure.

HOW TO CITE THIS ARTICLE

Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF. New version of the Pediatric Evaluation of Disability
Inventory (PEDI-CAT): translation, cultural adaptation to Brazil and analyses of psychometric properties. Braz J Phys Ther.
2016 Nov-Dec; 20(6):561-570.  http://dx.doi.org/10.1590/bjpt-rbf.2014.0166

Introduction
Procedures for the translation and cultural adaptation stages and procedures9-12, it needs to be preceded by
of foreign functional measures are applied in Brazil to two important questions: (1) Is the instrument to be
provide professionals and services with standardized translated into another language relevant to clinical
instruments that are often used in other countries and practice and scientific research? (2) Does the instrument
cited in the international literature1-8. Because this is a have appropriate content and format for use in the
complex and expensive process, with many structured country for which it is being made available?13.

1
Programa de Pós-graduação em Ciências da Reabilitação, Escola de Educação Física, Fisioterapia e Terapia Ocupacional, Universidade Federal de
Minas Gerais (UFMG), Belo Horizonte, MG, Brazil
2
Department of Occupational Therapy, College of Health and Rehabilitation Sciences: Sargent College, Boston University, Boston, USA
3
Departamento de Terapia Ocupacional de Lagarto, Universidade Federal de Sergipe (UFS), Lagarto, SE, Brazil
Received: Nov. 25, 2015 Revised: Feb. 08, 2016 Accepted: Feb. 23, 2016

http://dx.doi.org/10.1590/bjpt-rbf.2014.0166 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570 561 

 Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF

The Pediatric Evaluation of Disability Inventory- a new, more challenging item (i.e., one with greater
Computer Adaptive Test (PEDI-CAT) affirmatively relative difficulty) is displayed next, and so on, to
answers these two questions. This is the new version obtain an accurate estimate of the individual’s score
of the Pediatric Evaluation of Disability Inventory along the functional continuum14. This technology
(PEDI). It incorporates innovative measurement results in less administration time, as there is no need
methodologies, substantially extends the age range, and to respond to all items on a scale.
offers new items and a new format for the functional Strong levels of reliability and validity for the
evaluation of children and youth, from 0 to 21 years PEDI-CAT21 are documented in the literature, and the
of age, with various health conditions14. instrument is used internationally in clinical practice and
The PEDI-CAT is a functional assessment that is research22,23. In Brazil, the PEDI in its original version,
theoretically grouded on the International Classification is an outcome measure widely used in rehabilitation
of Functioning (ICF), Disability and Health15 and centers and clinical research. The availability of this
the ICF-Children and Youth (ICF-CY)16. Based on new version (PEDI-CAT) will be very useful for
the biopsychosocial and developmental models, it professionals and managers of rehabilitation services,
incorporates the sociocultural perspective17. In the who will be able to identify the functional level of
biopsychosocial model, functioning reflects the a child or youth, document changes occurring as a
interaction between individuals with a health condition result of services/interventions, and accurately guide
and the opportunities or barriers present in the setting specific targets for functional treatment with greater
in which they live, including internal (personal) and time efficiency.
external (environmental) factors. This interaction has a This study describes the procedures used for
bidirectional influence on body structures and functions, translation and cultural adaptation of the PEDI-CAT
activities and participation, which together represent into Brazilian Portuguese and tested its psychometric
the components of function. Based on the ICF-CY properties (i.e. reliability and validity).
model, the PEDI-CAT contents provide information
about the activities and participation component. In  a Method
socio-cultural approach, child’s learning process of
daily activities is grounded in guided participation and PEDI-CAT description
the transfer of responsibilities. Engagement of the child The PEDI-CAT24 consists of four domains: (1)
(i.e., the apprentice) with his/her parents and family Daily Activities (DA), (2) Mobility (MB), (3) Social/
members (i.e., experts) in the daily routine provides a Cognitive (SC), and (4) Responsibility (RS). It aims
context that triggers a collaborative process in which to provide a detailed description of the individual’s
the child is guided by the expert to become engaged function and document individual changes and
and gradually take responsibility for the performance the progress of functional skills acquired after an
of activities and tasks, while the caregiver decreases intervention. The PEDI-CAT is not a performance‑based
the assistance given17,18. This approach is used to guide “test”, but rather, it consists of a large item bank of
the content and scoring criteria of the items on one 276 functional activities acquired during childhood,
of the PEDI‑CAT test scales (i.e., Responsibility). adolescence, and early adulthood. Its application
The theoretical-methodological foundation that requires a computer with the instrument’s software
grounded the construction of the PEDI-CAT was installed and can either be self-administered (i.e. filled
derived from the Rasch model. This is a probabilistic out by the child’s parents), or a professional may be
model, from Item Response Theory (IRT), which present with the parents to ensure understanding of
transforms ordinal information into interval measures19. the information for each item14.
In this model, individuals and items are calibrated In the DA, MB, and SC domains, scoring is based
hierarchically, in order of relative difficulty, making on a four-point ordinal scale with different levels
it possible to identify more and less difficult items of difficulty. The RS domain scores the items on a
as well as more and less skilled subjects on the same five‑point scale, describing the sharing of responsibility
one-dimensional continuum, based on the calibrated between caregiver and child/youth in performing each
items and subjects (logits)20. The computer adaptive item (Table 1). For the four domains, the respondent
testing method uses an algorithm to select the is asked to choose the option that best describes the
(calibrated) items to be administered while evaluating child’s function for each item. If the respondent is
a particular individual in real time. For example, if unsure, there is an option for answering, “I do not
the individual obtains a high score on the initial item, know”14.

  562 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570

 PEDI-CAT: translation and adaptation

Table 1. A brief description of the content of each domain and the scoring scale of the PEDI-CAT items.

DOMAIN CONTENT* ITEM SCORING SCALE

Getting Dressed, Keeping Clean, Home
Daily Activities 1) Unable = Cannot do, does not know how or is too young.
Tasks, and Eating and Mealtime
2) Hard = Performs with much help, extra time, or effort.
Basic Movement and Transfers, Standing 3) A little hard = Performs with a little help, extra time, or
Mobility and Walking, Steps and Inclines, effort.
Running and Playing, and Wheelchair 4) Easy = Performs with no help, extra time, or effort or the
child’s skills are past this level.
Interaction, Communication, Everyday I Don’t Know.
Social/Cognitive
Cognition, and Self Management

1) Adult/caregiver has full responsibility; the child does not

take any responsibility.
2) Adult/caregiver has most responsibility, and the child takes
a little responsibility.
Organization and Planning, Taking Care 3) Adult/caregiver and child share responsibility
Responsibility of Daily Needs, Health Management, approximately equally.
and Staying Safe 4) Child has most responsibility with a little direction,
supervision, or guidance from an adult/caregiver.
5) Child takes full responsibility without any direction,
supervision, or guidance from an adult/caregiver.
I Don’t Know

*Information available at: http://pedicat.com/category/domains/.

Translation and cultural adaptation In the next stage, the T123 version was back
The PEDI-CAT translation process followed the translated into English independently by three other
methodology proposed by specific guidelines9-11 bilingual translators (BT1, BT2, and BT3) who had
for this type of study, taking into consideration no knowledge of the original version of the instrument
and who were not involved in the previous stage.
information from the PEDI-CAT translation guide sent
The translators were familiar with English and
by the authors after receiving authorization for the
Portuguese, and one was an English teacher.
translation. This study was approved by the Research
In the fourth stage, an expert committee reviewed
Ethics Committee of the Universidade Federal de
the versions (original, T123, BT1, BT2, and BT3) and
Minas Gerais (UFMG), Belo Horizonte, MG, Brazil
discussed each item, searching for the best solution
(CAAE: 20466614.0.0000.5149).
to solve the discrepancies and different alternatives
The translation and cultural adaptation followed
in translation. Rather than focusing on indices of
six stages: (1) translation (2) synthesis, (3) back
agreement, the committee attempted to make the best
translation, (4) review by an expert committee, use of the language expertise of its members, solving the
(5) testing of the pre-final version, and (6) testing of following types of disagreements: conceptual (referring
the psychometric properties. to the conceptualization of the evaluated phenomenon),
Three pairs of independent bilingual translators, idiomatic (different linguistic expressions), semantic
whose native language was Portuguese but who were (differences related to the test content), and experiential
fluent in English, translated the original English (related to cultural differences). The expert committee
version of the PEDI-CAT items into Portuguese. was composed of two physical therapists and four
The translators worked in independent pairs to occupational therapists who had not participated in
foster the use of best consensus terminology in each the previous stages. One of these participants had
translated version. They were either physical therapists extensive experience in psychometrics, one was the
or occupational therapists with significant experience author of the original PEDI-CAT, and another was
in child development. This process resulted in three also an English teacher. The other three experts had
translated versions: T1, T2, and T3, which were participated in the translation and/or back-translation of
analyzed in detail at a meeting with the translators other functional instruments. At this stage, additional
and coordinators, resulting in a synthesized translated reviews were conducted by professional from two
version, T123. rehabilitation centers in different regions of Brazil

Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570 563 

 Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF

(i.e., southeast and northeast). These reviews aimed administrators were deemed fit to administer the
to determine whether the functional activities and the instrument.
language used in the wording of the translated items The PEDI-CAT/Brazil was administered at a date,
were consistent with regional idiosyncrasies and could time, and place most convenient for the respondent.
easily be understood by health professionals of various An additional strategy regarding the items’ scoring
backgrounds13. Professionals from the Associação scales was adopted for all administrations. A supplement
Mineira de Reabilitação (AMR) in Belo Horizonte, with the criteria from each response option from
MG, and the Early Treatment and Stimulation Center the three functional skills domains as well as from
(Núcleo de Tratamento e Estimulação Precoce‑NUTEP) the responsibility scale was printed on cards that
in Fortaleza, CE, Brazil contributed to the reviews. remained with the respondents while they completed
This judicious review process generated the pre‑final the assessment. However, the score values (i.e., 1,
version of the instrument, hereafter called the 2, 3, and 4) were excluded from the cards to avoid
PEDI-CAT/Brazil. influencing each respondent’s answer. This method
helped each respondent remember the response options
Testing the pre-final version for each item and thus facilitate and standardize data
collection.
Participants
After completing the PEDI-CAT/Brazil, each
The pre-final version of the PEDI-CAT/Brazil was
respondent was asked to evaluate the adequacy
completed by a sample of 810 parents and/or guardians
of the instrument translation by answering the
of children and youth from 0 to 21 years of age with
following questions: (1) What is your opinion of
normal development. Participants were recruited by
the translation of the PEDI-CAT? (2) Do the tasks
convenience, informed about the study objectives,
listed in this questionnaire describe your child’s
and asked to sign an informed consent form.
function?11. According to the guidelines adopted in
this study, a negative response percentage above 15%
Procedures
would indicate the need to revise the instrument’s
Following the authors’ instructions, the 276 translation26,27.
PEDI‑CAT/Brazil items were distributed to the
three age groups (0-7, 8-14, and 15-21 years of age) Psychometric properties
by experienced professionals in child development
The psychometric and measurement properties
based on the age appropriateness of each item.
These professionals did not include the translators or evaluated included inter-examiner and test-retest
the panel of experts. It is important to stress that this reliability, internal consistency, standard error of
division of items by age group was performed for the measurement (SEM) and factor analysis. Reliability
sole purpose of facilitating the subsequent collection estimates were calculated for each age group and
of normative data, and the final Brazilian version of for the 20 administrators, totaling 60 measures.
the instrument comprised all of the items related to The test‑retest reliability evaluated the consistency
all ages covered by the instrument. of parents’ responses across two occasions when the
After the three age-group versions were created, PEDI-CAT items were administered with an interval of
20 administrators received detailed training on 7 to 15 days between the two tests. The inter‑examiner
instrument content and data collection procedures. reliability was evaluated by the stability of information
The standardized training lasted 30 hours. It included when groups of examiners administered the PEDI-CAT
the following steps: (1) explanation of the instrument items to the same parents. Internal consistency was
goals, (2) detailed explanation of each item and its examined using Cronbach’s alpha, and a reference value
scoring scale, (3) review of the use of the Economic above 0.70 was considered acceptable28. Data from
Classification Criteria Brazil 2012, proposed by the the reliability analyses were used to calculate the
Brazilian Association of Research Companies (Associação SEM. Exploratory factor analysis (EFA) evaluated the
Brasileira de Empresas de Pesquisa‑ABEP)25 to organization of items from the three functional skills
characterize the socioeconomic status of respondents, domains (i.e., Daily Activities, Mobility and Social/
(4) scoring of training videos for all three age groups, Cogntive ) and from the Responsibility domain into
and (5) evaluation of inter‑examiner and test-retest dimensions or concepts (i.e., factors) as well as the
reliability. After completing all of these steps, the explanatory value of the factors solutions.

  564 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570

 PEDI-CAT: translation and adaptation

Statistical analyses equivalence. The type and frequency of strategies

The participants were characterized using adopted in each test domain are described in Table 2.
descriptive statistics. Intra-class correlation coefficients Table 3 shows some examples of discrepancies found
(ICC) and confidence intervals (CI) of the mean during the translation process and the strategies used
coefficients (95%) informed the inter-examiner and by the expert committee to solve them.
test-retest reliability. The criteria adopted for the
ICC interpretation was: ICC<0.40-weak agreement; Testing the pre-final PEDI-CAT/Brazil
ICC≤0.75-moderate agreement and ICC> 0.75 indicated version
excellent agreement29. The SEM was calculated as The pre-final PEDI-CAT/Brazil version was
the square root of the mean square error obtained in administered to the three age groups, with the greatest
the analysis of variance30. Factor analysis adequacy proportion in the 0 to 7-year age group (45%, n=367).
was tested using the Kaiser-Meyer-Olkin (KMO) The mean age was 9.37 years (SD=5.93), there was a
test and Bartlett’s test of sphericity. Correlations greater percentage of females (57%), and individuals
above 0.50 in the KMO test and values of p <0.05 in classified as socioeconomic level B (43%). Most
Bartlett’s test of sphericity indicated adequacy of the respondents reported having no trouble understanding
data for factor analysis. Measures of central tendency the items (95%) and that the content of the PEDI-CAT/
(i.e., mean) and variability (i.e., standard deviation Brazil adequately represented the child’s function in
[SD], variance, and covariance) were used to define the family’s daily life (98%) (Table 4).
the factor loadings and the eigenvalues assigned to
the items. To define the number of factors in the final PEDI-CAT/Brazil reliability, internal
solution, the Kaiser Rule (i.e., eigenvalues above one) consistency and SEM
was used in addition to Scree plot analysis. Oblique The inter-examiner and test-retest reliabilities,
rotation method (i.e., Oblimin rotation) optimized internal consistency, and SEM were good to excellent,
interpretation, as correlation among factors was as shown in Table 5.
assumed31.
The statistical analyses used the Statistical Package Factor analysis
for the Social Sciences (SPSS), version 19.0. The KMO test showed correlations higher than 0.50,
and the results of Bartlett’s test of sphericity indicated
Results p-values lower than 0.05, which indicate adequacy of
the data for factor analysis. Initially, factor analysis
PEDI-CAT translation and cultural of the items of the three functional skills domains
adaptation identified four factors with eigenvalues greater than
Some differences in the versions analyzed by the one. However, plot analysis showed that after the
expert committee during the PEDI-CAT translation first factor, the variance curve became horizontal.
process were observed and resolved using strategies Data were further analyzed by forcing solutions of
such as addition, omission, or substitution of words three, two and one factor. In the three factors solution
and the provision of examples, in an attempt to reach only one item (MB079-Walks down a flight of stairs
semantic, conceptual, idiomatic, and experiential without holding onto handrail) loaded on the second

Table 2. Number and type of strategies recommended by the Expert Committee in each domain of the PEDI-CAT.

PEDI-CAT PEDI-CAT STRATEGIES USED IN THE TRANSLATION**

TOTAL
DOMAIN* ITEMS EX AD OM SUB EXC
DA 68 7 5 4 0 0 16
MB 97 0 9 5 1 1 16
SC 60 0 5 0 1 0 6
RS 51 7 3 0 3 0 13
TOTAL 276 14 22 9 5 1 51
*Domains: DA: Daily Activities; MB: Mobility; SC: Social/Cognitive; RS: Responsibility. **Strategies: EX: example; AD: addition, OM:  omission;
SUB: substitution; EXC: exclusion.

Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570 565 

 Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF

Table 3. Examples of translation strategies proposed by the Expert Committee.

ORIGINAL
ITEM TRANSLATED VERSION STRATEGY
VERSION

Pours liquid from a

Despeja o líquido de uma caixa (por
DA013 large carton into a Example
exemplo, de suco ou leite)* em um copo
glass

Cuts with scissors Abre uma embalagem de plástico duro (por

DA021 to open hard plastic exemplo, de brinquedo ou eletrônico)* Example
packaging usando tesoura

Climbs on and off a Sobe e desce um brinquedo de escalar

MB095 Addition
climbing structure (trepa-trepa)*

Walks 3 miles*/5
MB128 Caminha por 5 quilômetros Omission
kilometers

Plays peek-a-boo or
SC019 Brinca de “achou” ou jogos simples de bater palmas com as mãos* Substitution
pat-a-cake*
Following a recipe or
cooking instructions
that includes 3-4 Segue uma receita ou instruções de culinária que incluem 3-4
RS015 Substitution
ingredients and steps, ingredientes e passos, tais como massa de bolo pronta, miojo, tapioca*
such as macaroni and
cheese or brownies
Identifying correct
polling location;
Identificar o local correto de votação; Compreender o processo e os
Understanding the
direitos/deveres de votar; Requisitar formulário de justificativa quando
RS055 voting process and Addition
necessário; Compreender os procedimentos necessários de uso da
rights; Requesting
urna eletrônica*
absentee ballots as
needed
*Parts in bold illustrate what was added, substituted and/or omitted in relation to the orginal item.

factor and three items (DA039-Fastens hairclips or 15 to 21-year age group were grouped into the first
barrettes, DA040-Puts hair up in a ponytail and factor. Items from the 0 to 7 year and 8 to 14-year groups
DA074-Puts on bra and fastens in front or back) on loaded on the second and third factors, respectively.
the third factor. For the two factors solution, only one This three factors solution explained approximately
item (MB079) loaded on the second factor. Those 82% of the item variance.
analyses confirmed the plot information. Thus, the one
factor solution best represented the conceptual (i.e.
Discussion
latent) structure underlying these items, explaining
89% of the variance. The translation and cultural adaptation of measurement
Factor analysis of the Responsibility domain grouped instruments utilizes standardized methods and
items into three factors. Responsibility items from the judicious criteria to develop appropriate versions of

  566 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570

 PEDI-CAT: translation and adaptation

Table 4. Testing of the pre-final version of the PEDI-CAT/Brazil.

Sample Description and Information 0 to 7 years 8 to 14 years 15 to 21 years TOTAL

provided about the PEDI-CAT (n=367) (n=240) (n=203) (n=810)
Age* 3.78 (2.21) 11.17 (1.90) 17.36 (1.65) 9.37 (5.93)
Children/ Youth F 205 (56%) 131 (55%) 128 (63%) 464 (57%)
Sex**
M 162 (44%) 109 (45%) 75 (37%) 346 (43%)
A 52 (14%) 27 (11%) 31 (15%) 110 (14%)
Family socioeconomic Socioeconomic B 169 (46%) 94 (39%) 87 (43%) 350 (43%)
information*** classification C 122 (33%) 96 (40%) 73 (36%) 291 (36%)
D 24 (7%) 23 (10%) 12 (6%) 59 (7%)

Difficulty None 342 (93%) 233 (97%) 193 (95%) 768 (95%)
demonstrated in A little 22 (6%) 7 (3%) 10 (5%) 39 (5%)
About the PEDI-CAT understanding items A lot 3 (1%) 0 (0%) 0 (0%) 3 (0%)
Do items represent Yes 359 (98%) 237 (99%) 195 (96%) 791 (98%)
functioning? No 8 (2%) 3 (1%) 8 (4%) 19 (2%)
*Age: mean value (standard deviation). **Sex: F: Female, M: Male, the values determine the frequency (percentage). ***Family socioeconomic
classification. The categories represent socioeconomic levels. They are defined from a standardized questionnaire that assigns points to items
related to the presence and amount of certain home appliances, number of cars owned, and the level of formal education of the main family
member (provider). The points are summed, and specific ranges are translated into categories, in which higher total scores refer to higher
socioeconomic levels. Level A=46 to 35 points; B=23 to 34 points; C=14 to 22 points; and D=8 to 13 points.

Table 5. Inter-examiner and test-retest reliability indices, internal consistency, and standard error of measurement values of the PEDI-CAT/
Brazil.

0-7 years 8-14 years 15-21 years

Inter-examiner Reliability 0.86 (0.81-0.90) 0.83 (0.76-0.88) 0.89 (0.85-0.92)
(95% CI)
Test-retest Reliability 0.96 (0.95-0.97) 0.97 (0.97-0.98) 0.97 (0.96-0.98)
(95% CI)
Internal Consistency 0.99 0.99 0.99
Standard error of measurement 0.12 0.17 0.13

The values reported in this table represent the lower reliability indices obtained from the analyses.

foreign instruments for use in countries with different instrument6. An example illustrates the translation of
languages13. This study rigorously followed the stages items from the Mobility domain. For item MB012 “Sits
of internationally recognized guidelines9-11, resulting on floor with pillow for support”, it was necessary to
in the development of a translated version of the determine whether it referred to the child in a static
PEDI-CAT that incorporates the specifics of Brazilian position or dynamically positioning him/herself. Thus,
culture and reality. It could be understood by health the term “sits” was translated as “remains seated”
professionals and by parents of children and youth rather than “sits”, as the latter refers to the action of
from different socioeconomic classes. In addition, sitting and does not reflect a true translation of the
procedures included a review stage of the translated situation measured by the original item.
version by professionals from rehabilitation centers Producing a valid translation into a language other
in the southeast and northeast regions of the country. than the one in which the original version was developed
Strategies such as word addition, omission, substitution often presents difficult challenges, such as experiential
and provision of examples were used to account for differences between cultures. The expert committee
the varied socio-cultural and educational realities involved in the development of the PEDI-CAT/Brazil
of the Brazilian population, while ensuring that the version decided to remove item MB032 “gets in and
translated words and expressions remained faithful out of the bathtub” and to omit or replace the terms
to each specific situation measured by the original “bath” and “tub” from items DA051 “Cleans up

Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570 567 

 Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF

thoroughly in bath or shower” and RS020 “Cleaning original version14, supporting the PEDI-CAT/Brazil
spills and wiping up food crumbs; Scrubbing sink as a reliable instrument for the functional evaluation
and tub; Emptying trash; Replacing or repairing of Brazilian children and youth. The SEM values
broken fixtures or objects” because bath tubs are not observed in the PEDI-CAT/Brazil, which reflect
common in Brazil. This adaptation had the consent the degree of inaccuracy (i.e., measurement error)
of the PEDI-CAT authors, without prejudice to the of the final score, were small (0.12 to 0.17 points).
functional evaluation. Previous Brazilian adaptation These values were also similar to those provided in
of the content of the original PEDI contributed to the the original PEDI-CAT manual14, suggesting that
development of items for the PEDI-CAT. The mobility if the magnitude of a change in scores (e.g., after
item from the Portuguese version32 MB033 “gets in intervention or development) is higher than the SEM
and out of the shower” was incorporated as an item value, such a change should be attributed to the effect
in the new PEDI-CAT version. of changes resulting from the investigated process.
A necessary precaution in the translation and Factor analysis showed that much of the variance in
cultural adaptation of instruments into Portuguese is the PEDI-CAT/Brazil items could be explained by one and
appropriateness of the language to the characteristics three factors in the Functional Skills and Responsibility
and peculiarities of different Brazilian regions. Given domains, respectively. The Responsibility items were
the extensive size of the country, cultural and linguistic grouped into factors related to age groups, indicating
differences among Brazilian regions should be considered the suitability of the instrument for the development
in a translated instrument that will be used across of children and youth’s responsibilities.
the country13. In item RS035 “Following a recipe or The collection of normative PEDI-CAT/Brazil
cooking instructions that includes 3 to 4 ingredients data is the next stage of this project, which has the
and steps such as macaroni and cheese or brownies”, support of a second center in Fortaleza, CE, Brazil.
for example, the examples “macaroni and cheese” and The availability of a theoretically consistent functional
“brownies” were replaced by “massa de bolo pronta” instrument that is culturally appropriate and reliable,
(i.e. cake mix), “miojo” (i.e. ramen noodles), and which requires a shorter administration time without
“tapioca”; the latter is a typical snack in the northern losing the accuracy of estimates of the functional
and northeastern regions (Table 3). levels of infants, children, adolescents, and young
The pre-final version revealed that this PEDI-CAT/ adults, may contribute to improved evaluation and
Brazil version was understood and evaluated as an measurement of clinical intervention outcomes and
adequate measure of function by most participants. longitudinal monitoring, further supporting advances
Only 5% of respondents reported having difficulty in rehabilitation research.
understanding some items. It is only necessary to
review a translation when the doubt index exceeds
15%26. However, it was observed that the respondents
Acknowledgements
who reported having some difficulty in understanding We would like to thank the parents of the children,
cited some items in common, especially in the adolescents, and young adults who agreed to voluntarily
Responsibility domain. This is a complex domain that participate in this study. Acknowledgements should
measures the transfer of responsibility from parents also be given to the group of professionals who
or other caregivers to children that occurs during collaborated in the translation and cultural adaptation of
learning relationships18. The items in this domain the PEDI-CAT/Brazil: Adriana de França Drummond,
provide a detailed explanation of information that the Amanda Aquino, Ana Paula B. Gontijo, Bruna Maria
respondent should consider for scoring. A possible Soares Pereira, Camila Mendes, Priscilla Rezende
strategy could, therefore, be to provide a small booklet Figueiredo, Rachel Helena, Rafaelle Gracine, Rejane
in Portuguese containing very detailed criteria and Vale Gonçalves, and Solange Figueiredo Nogueira.
detailed examples to assist in the choice of options We thank the therapists from Associação Mineira de
for responses to some items of that domain. Such an Reabilitação (led by Marina de Brito Brandão) and
initiative would require the approval of the original Núcleo de Tratamento e Estimulação Precoce (led by
PEDI-CAT test authors. Rita Brasil), who carefully reviewed the pre‑final
High levels of reliability and internal consistency translated version. Finally, financial support for
were found across all age groups in this study29. this study was granted by the Brazilian government
These results are similar to those reported for the agencies Conselho Nacional de Desenvolvimento

  568 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570

 PEDI-CAT: translation and adaptation

Científico e Tecnológico (CNPq), Coordenação http://dx.doi.org/10.1097/00007632-200012150-00014 .

de Aperfeiçoamento de Pessoal de Nível Superior PMid:11124735.
(CAPES), the Fundação de Amparo a Pesquisa do 10. Guillemin F, Bombardier C, Beaton D. Cross-cultural
adaptation of health-related quality of life measures:
Estado de Minas Gerais (FAPEMIG), and Laboratory literature review and proposed guidelines. J Clin Epidemiol.
of Medical Investigation from the Clinics Hospital, 1993;46(12):1417-32. http://dx.doi.org/10.1016/0895-
FMUSP - LIM 34 (Rehabilitation Sciences). The Office 4356(93)90142-N. PMid:8263569.
of the Provost for Research from Universidade Federal 11. Van Widenfelt BM, Treffers PDA, Beurs E, Siebelink BM,
de Minas Gerais, Brazil, provided funds for the English Koudijs E. Translation and cross-cultural adaptation of
translation and research assistantship. assessment instruments used in psychological research
with children and families. Clin Child Fam Psychol Rev.
2005;8(2):135-47. http://dx.doi.org/10.1007/s10567-005-
References 4752-1. PMid:15981582.
12. Mokkink LB, Prinsen CA, Bouter LM, Vet HC, Terwee
1. Pilz B, Vasconcelos RA, Marcondes FB, Lodovichi SS, CB. The COnsensus-based Standards for the selection of
Mello W, Grossi DB. The Brazilian version of STarT Back health Measurements INstruments (COSMIN) and how to
Screening Tool - translation, cross-cultural adaptation and select an outcome measure instrument. Braz J Phys Ther.
reliability. Braz J Phys Ther. 2014;18(5):453-61. http://dx.doi. 2016. In press. http://dx.doi.org/10.1590/bjpt-rbf.2014.0143.
org/10.1590/bjpt-rbf.2014.0028. PMid:25372008. PMid:26786084.
2. Lamarão AM, Costa LCM, Comper MLC, Padula RS. 13. Coster WJ, Mancini MC. Recommendations for translations
Translation, cross-cultural adaptation to Brazilian-Portuguese and cross-cultural adaptation of instruments for occupational
and reliability analysis of the instrument Rapid Entire Body therapy research and practice. Rev Ter Ocup Univ São Paulo.
Assessment-REBA. Braz J Phys Ther. 2014;18(3):211-7. 2015;26(1):50-7. http://dx.doi.org/10.11606/issn.2238-6149.
http://dx.doi.org/10.1590/bjpt-rbf.2014.0035. PMid:25003273.
v26i1p50-57.
3. Furtado SRC, Sampaio RF, Vaz DV, Pinho BAS, Nascimento
14. Haley SM, Coster WJ, Dumas HM, Fragala-Pinkham MA,
IO, Mancini MC. Brazilian version of the instrument of
Moed R. PEDI-CAT: development, standardization and
environmental assessment Craig Hospital Inventory of
administration manual. Boston: Boston University; 2012.
Environmental Factors (CHIEF): translation, cross-cultural
adaptation and reliability. Braz J Phys Ther. 2014;18(3):259-67. 15. World Health Organization – WHO. International classification
http://dx.doi.org/10.1590/bjpt-rbf.2014.0036. PMid:25003279. of functioning, disability and health. Genebra: WHO; 2001.
4. Lopes AR, Trelha CS. Translation, cultural adaptation and 16. World Health Organization – WHO. International classification
evaluation of the psychometric properties of the Falls Risk of functioning, disability and health children and youth version.
Awareness Questionnaire (FRAQ): FRAQ-Brazil. Braz J Genebra: WHO; 2007.
Phys Ther. 2013;17(6):593-602. http://dx.doi.org/10.1590/ 17. Rogoff B. Developmental transitions in children’s participation
S1413-35552012005000128. PMid:24346294. in sociocultural activities. In: Sameroff A, Haith MM. The
5. Pereira LM, Dias JM, Mazuquin BF, Castanhas LG, Menacho five to seven year shift: the age of reason and responsibility.
MO, Cardoso JR. Translation, cross-cultural adaptation Chicago: The University of Chicago; 1996.
and analysis of the psychometric properties of the lower 18. Rogoff B. Apprenticeship in thinking: cognitive development
extremity functional scale (LEFS): LEFS- BRAZIL. Braz in social context. New York: Oxford University Press; 1990.
J Phys Ther. 2013;17(3):272-80. http://dx.doi.org/10.1590/
19. Ayala RJ. The theory and practice of item response theory.
S1413-35552012005000091. PMid:23966144.
New York: The Guilford Press; 2009.
6. Amaral MF, Paula RL, Drummond A, Dunn L, Mancini MC.
Translation of the Children Helping Out: Responsibilities, 20. Chang W, Chan C. Rasch analysis for outcome measures:
Expectations and Supports (CHORES) questionnaire into some methodological considerations. Arch Phys Med
Brazilian-Portuguese: semantic, idiomatic, conceptual and Rehabil. 1995;76(10):934-9. http://dx.doi.org/10.1016/
experiential equivalences and application in normal children S0003-9993(95)80070-0. PMid:7487434.
and adolescents and in children with cerebral palsy. Braz 21. Dumas HM, Fragala-Pinkham MA, Haley SM, Ni P, Coster
J Phys Ther. 2012;16(6):515-22. http://dx.doi.org/10.1590/ W, Kramer JM, et al. Computer adaptive test performance
S1413-35552012000600011. PMid:23348598. in children with and without disabilities: prospective field
7. Puga VOO, Lopes AD, Costa LOP. Assessment of study of the PEDI-CAT. Disabil Rehabil. 2012;34(5):393-
cross-cultural adaptations and measurement properties 401. http://dx.doi.org/10.3109/09638288.2011.607217.
of self-report outcome measures relevant to shoulder PMid:21988750.
disability in Portuguese: a systematic review. Braz J Phys 22. Kao YC, Kramer JM, Liljenquist K, Tian F, Coster WJ.
Ther. 2012;16(2):85-93. http://dx.doi.org/10.1590/S1413- Comparing the functional performance of children and
35552012005000012. PMid:22588123. youths with autism, developmental disabilities, and no
8. Maher CG, Latimer J, Costa LOP. The relevance of cross- disability using the revised pediatric evaluation of disability
cultural adaptation and clinimetrics for Physical Therapy inventory item banks. Am J Occup Ther. 2012;66(5):607-16.
instruments. Braz J Phys Ther. 2007;11(4):245-52. http:// http://dx.doi.org/10.5014/ajot.2012.004218. PMid:22917127.
dx.doi.org/10.1590/S1413-35552007000400002. 23. Kramer JM, Coster WJ, Kao YC, Snow A, Orsmond GI. A
9. Beaton DE, Bombardier C, Guillemin F, Ferraz MB. Guidelines new approach to the measurement of adaptive behavior:
for the process of cross-cultural adaptation of self-report development of the PEDI-CAT for children and youth
measures. Spine (Phila Pa 1976). 2000;25(24):3186-91. with autism spectrum disorders. Phys Occup Ther Pediatr.

Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570 569 

 Mancini MC, Coster WJ, Amaral MF, Avelar BS, Freitas R, Sampaio RF

2012;32(1):34-47. http://dx.doi.org/10.3109/01942638.201 30. Lexell JE, Downham DY. How to assess the reliability
1.606260. PMid:21846290. of measurements in rehabilitation. Am J Phys Med
24. Pediatric Evaluation of Disability Inventory Computer Adaptive Rehabil. 2005;84(9):719-23. http://dx.doi.org/10.1097/01.
Test – PEDI-CAT. Information about the PEDI-CAT (English phm.0000176452.17771.20. PMid:16141752.
version) [Internet]. 2015 [cited 2015 Nov 8]. Available from:
31. Hair JR, Black WC, Babin BJ, Anderson RE, Tatham RL.
http://pedicat.com/category/home/
Multivariate data analysis. 6th ed. New York: Pearson Prentice
25. Associação Brasileira de Empresas de Pesquisa – ABEP. A
Hall; 2006.
ABEP: Critério Brasil [Brazilian Market Research Association]
[Internet]. São Paulo: ABEP; 2015 [cited 2015 Nov 8]. 32. Mancini MC. Pediatric Evaluation of Disability Inventory
Available from: http://www.abep.org/criterio-brasil (PEDI): manual of the Brazilian adapted version. Belo
26. Ciconelli RM, Ferraz MB, Santos W, Meinão I, Quaresma MR. Horizonte: Editora UFMG; 2005. In Portuguese.
Brazilian-Portuguese version of the SF-36: a reliable and
valid quality of life outcome measure. Rev Bras Reumatol.
1999;39(3):143-50.
27. Nusbaum L, Natour J, Ferraz MB, Goldenberg J. Translation,
adaptation and validation of the Roland-Morris questionnaire. Correspondence
Braz J Med Biol Res. 2001;34(2):203-10. http://dx.doi. Marisa C. Mancini
org/10.1590/S0100-879X2001000200007. PMid:11175495. Universidade Federal de Minas Gerais
28. Nunnally JC, Bernstein IH. The assessment of reliability. Escola de Educação Física, Fisioterapia e Terapia Ocupacional
Psychometric Theory. 1994;3:248-92. Programa de Pós-graduação em Ciências da Reabilitação
29. Portney LG, Watkins MP. Foundations of clinical research: Avenida Antônio Carlos, 6627, Pampulha
applications to practice. Upper Saddle River: Prentice Hall; CEP 31270-010, Belo Horizonte, MG, Brazil
2000. v. 2. e-mail: [email protected], [email protected]

  570 Braz J Phys Ther. 2016 Nov-Dec; 20(6):561-570