Cross-cultural adaptation of the Nordic musculoskeletal questionnaire
(Parte 1 de 4)
Correspondence address: Prof. Neusa M. C. Alexandre, Department of Nursing, College of Medical Sciences, State University of Campinas (UNICAMP), Cidade Universitatria ‘Zeferino Vaz’, Dist. Barao Geraldo, Cep. 13083-970, Campinas, SP, Brazil Tel.: 5 19 37888820 Fax: 5 19 37888822 E-mail: firstname.lastname@example.org or email@example.com
Cross-cultural adaptation of the Nordic musculoskeletal questionnaire
E. N. C. de Barros 1 RN
& N. M. C. Alexandre 2 RN , PhD
1 Nurse, Department of Nursing, College of Medical Sciences, State University of Campinas (UNICAMP), Campinas, Sao Paulo,
2 Nurse, Department of Nursing, College of Medical Sciences, State University of Campinas (UNICAMP), Campinas, Sao Paulo,
Abstract Background: Reports in the literature have identified a need for internationally standardized and reliable measurements to analyse musculoskeletal symptoms. Screening of musculoskeletal disorders may serve as a diagnostic tool to evaluate the work environment. The Nordic general questionnaire is a standardized instrument used to analyse musculoskeletal symptoms in an ergonomic or occupational health context. Purpose: To translate and adapt a version of the Nordic general questionnaire into Brazilian Portuguese and evaluate its reliability. Method: The cross-cultural adaptation was performed according to internationally recommended methodology, using the following guidelines: translation; back-translation; committee review; and pretesting. First, the questionnaire was independently translated into Portuguese by two teachers and one doctor, and a consensus version was generated. Second, two other translators performed a back-translation independently from one another. This version was then submitted to a committee, consisting of six specialists in the area of knowledge of the instrument, to evaluate its equivalence to the original instrument. The final version was pretested on 20 subjects randomly selected in an outpatient clinic. Reliability was assessed by a test–retest procedure at 1-day intervals using the Kappa coefficient in a group of 40 subjects. The Kappa agreement values were calculated for each one of the four questions of the questionnaire. The agreement among the same observers was substantial, varying from 0.8 to 1, according to the Kappa values. Results: these demonstrated strong agreement of the instrument, suggesting that the Brazilian version of the ‘Standardized Nordic Questionnaire’ offers substantial reliability.
Keywords Cross-Cultural Adaptation, Ergonomics, Musculoskeletal Symptoms, Research Instrument
Musculoskeletal disorders represent a serious public health problem, being one of the most important causes of disability and absenteeism in workers, as well as having various personal, social and economic impacts (Courtney & Webster 1999; Lidgren 1998; Santos Filho & Barreto 2001; Yelin & Calahan
102E. N. C. de Barros & N. M. C. Alexandre
© 2003 International Council of Nurses, International Nursing Review , 50 , 101–108
1995). Included among these musculoskeletal disorders is low back pain, the most frequent cause of activity limitation in workers of productive age (Alexandre & Benatti 1998; Alexandre et al. 1996; 2001; Guo et al. 1999; Hagen & Thune 1998; Lee et al. 2001; Nordin et al. 1997).
It is currently known that the occurrence of musculoskeletal disorders encompasses several interrelated variables; for this reason, new approaches to identify and cope with the problem are increasingly being sought. The international literature describes questionnaires and specific scales for using to evaluate various aspects of musculoskeletal disorders, including: frequency, location and intensity of symptoms; functional capability and quality of life of the patients; and social and emotional aspects (Alonso et al. 1990; Deyo et al. 1998; Macdowell & Newell 1996; Nusbaum et al. 2001). These instruments are very important as they provide standardized data, which can be used in clinical practice, research and public health programmes. Some authors have recommended the adaptation of scales and questionnaires already validated in another language, defending the idea that this will facilitate the exchange of information in the scientific community (Hutchinson et al. 1996).
Several researchers have directed their studies to the development of methods for assessing musculoskeletal disorders, particularly to help define the relationship of the symptoms with the type of work and the areas of the body most affected. The ‘Standardized Nordic Questionnaire’ is an internationally respected instrument designed to standardize studies on evaluation of musculoskeletal complaints (Kuorinka et al. 1987). Its objective is to evaluate musculoskeletal problems in an ergonomic approach. The first part of this instrument consists of questions regarding parts of the human body that correspond to nine anatomical areas (three on the upper limbs, three on the lower limbs and three on the torso), marked on an illustration of the human body seen from behind. The instrument utilized in this study was adapted to the English language and the questions arranged in the following order: annual and weekly prevalence; functional incapacity; and search for medical assistance in the last 12 months (Bork et al. 1996; Dichinson et al. 1992;
Rosecrance et al. 1996). This instrument may be applied through an interview or answered by the person being evaluated.
Purpose of the study
The present study had the objective of culturally adapting the general content of the ‘Standardized Nordic Questionnaire’ to the Portuguese language and assessing the reliability of its Brazilian version.
In order to ensure the quality of adapted instruments, international norms should be followed when carrying out the adaptation. The employment of a methodology accepted by the scientific literature permits the comparison of data obtained in different services (Riberto et al. 2001). This study aimed to follow the essential steps that should be taken in the adaptation of instruments, based on specialized literature (Alexandre & Guirardello 2002; Guillemin 1995; Hutchinson et al. 1996). The phases carried out were: first, translation into the Portuguese language from the English version and back translation into English; second, analysis by a committee of specialists; and third, pretest and reliability testing (test–retest).
Initial translation into the Portuguese language
Translation of the general questionnaire that is part of the ‘Standardized Nordic Questionnaire’, was carried out by three Brazilian individuals who worked independently; two were English teachers and the third was a physician in the residency programme of Yale University (USA). The three translation versions were compared by the author and analysed until there was a consensus regarding the initial translation.
The initial translation into Portuguese was back translated into English by two different bilingual independent translators, whose mother tongue was English and who did not participate in the previous phase. This phase had the purpose of checking for
Nordic musculoskeletal questionnaire adaptation103
© 2003 International Council of Nurses, International Nursing Review , 50 , 101–108 discrepancies between content and meaning of the original version and the translated instrument. All the versions were analysed and compared by the author, and a final version was obtained.
The final translated version was submitted to a committee consisting of six specialists in the area of knowledge of the instrument, who were informed of the measures and concepts involved. This multidisciplinary team comprised an orthopaedist, a rheumatologist, a neurologist, an occupational health nurse, an undergraduate nursing student and a physician. Each of the committee members was asked to evaluate the equivalency of the original instrument to the final translated version of the ‘Standardized Nordic Questionnaire’. This evaluation was directed by the use of an instrument specifically developed for this study with the theoretical support of another study about cultural adaptation (Kimura 1999). This instrument consisted of nine items divided into two parts – A and B – with part A corresponding to the original version of the instrument being studied and part B being the version translated into Portuguese. Each of the professionals on this committee was instructed to evaluate each item regarding the equivalency of the transla- tions presented, following the scale: ‘ + 1’ when equivalent and ‘ - 1’ when not. On the basis of other research (Kimura 1999), the authors decided that in the event of obtaining three ‘ - 1’ answers it would be necessary to re-evaluate the translation for its alteration.
The questionnaire was applied to 20 randomly selected individuals, while they were waiting to be examined at the Health Unit of the University where the study was carried out. This group included employees, students and professors of the institution.
In order to simplify recording of doubts and suggestions concerning the questionnaire being studied, an instrument was specifically prepared for this research phase. This instrument at first contained general information on the interviewee, such as gender, age, civil status and occupation. It was divided in three main parts: ‘instructions to fill out questionnaire’, ‘questions’ and ‘body areas’. A place to write down doubts and suggestions was provided for each one of these three parts. All of the participants were informed of the purpose of the research study and the need to advise whenever a doubt about the written information and structure of the instrument emerged.
Evaluation of the reliability of the questionnaire
The reliability was assessed by using the test–retest method. This test consists of the application of the questionnaire to the same subjects, under a similar condition, in two or more situations (LoBiondo- Wood & Haber 1998; Polit & Hungler 1995).
The questionnaire was applied to 40 individuals from the university staff, chosen at random by the systematic choice method. These individuals comprised four distinct groups, as follows: 10 undergraduate nursing students; 10 professors from the Nursing Department; 10 administrative employees from the Medical Sciences School of the institution; and 10 patients assisted by the physiotherapy service at the Health Unit of the University. In order to be part of the sample, certain criteria had to be met, such as being adult, being literate, and being able to understand verbal expressions and answers by marking the chosen answers. The selected subjects were informed of the purpose of the research. Each received two copies of the instrument, and they were orientated to answer the first copy on the day when the questionnaire was handed in and the second copy on the next day, at the same time and under similar conditions, without referring to the previous questionnaire. General data on the interviewee were also collected, such as age, gender and educational status.
The data were entered and analysed by means of a database (Microsoft Excel 2000 Professional; Pro- gram Microsoft Map ” 1995–98. MapInfo Cor- poration, Troy, NY). The participants were characterized by descriptive analysis. The Kappa coefficient (k) was utilized for evaluation of test– retest reliability. This coefficient has values ranging
104E. N. C. de Barros & N. M. C. Alexandre
© 2003 International Council of Nurses,
International Nursing Review , 50 , 101–108 from - 1 to + 1. Values close to + 1 indicate total agreement between the two moments, while values close to - 1 indicate total discrepancy. Values higher than 0.75 represent strong agreement and those below 0.40 indicate poor agreement. Kappa coefficient values of 0.40–0.75 represent moderate agreement (Fleiss 1981).
At first, agreement of all the answers of participants was evaluated according to the nine body areas affected. Kappa coefficients concerning the answers to the four questions were also evaluated, without considering pain location.
The protocol received the approval of the Ethics Committee for Research of the Medical Sciences School of UNICAMP with the number 073/2001. All the individuals who agreed to participate in each phase of this study signed the informed consent form.
Analysis of the judges’ evaluations regarding equivalency of the instrument demonstrated that there was no need to modify its form and content. Therefore, the instrument was maintained in the form presented to this committee.
After analysing the records of participants’ doubts and suggestions, it was observed that there were two ways of making it easier to understand how to fill out the instrument. The first was to move the question about weekly prevalence of the problem to the end of the questionnaire, to facilitate the temporal situation, so that the first three questions refer to the last 12 months and the fourth question refers to the last 7 days. The second was that the word ‘doctor’ should be replaced by ‘health care professional (doctor, physician)’, increasing the options of the kind of assistance searched for in view of the problem presented. These suggestions were carefully analysed, and there was consensus that they were viable and would make it easier to understand the instrument. The final version is shown in Fig. 1.
Analysis of instrument reliability
(Parte 1 de 4)