Background: low back pain (LBP) is the main cause of years lived with disability worldwide. Psychosocial factors have been shown to be good predictors of persistent LBP. Within these, unhelpful beliefs about the back seem to be important in the development and chronicity of the symptoms. The Back Pain Attitudes Questionnaire (Back-PAQ) is an instrument that explores beliefs about the back that has been validated for people with and without back pain and healthcare professionals. However, until now, it has not been translated and validated for the Argentine population.
Objective: translate into Spanish, cross-cultural adapt and validate the Back-PAQ for the Argentine population with and without back pain.
Study design: study of diagnostic accuracy/assessment scale.
Methods: the study was carried out in three consecutive phases: translation, cross-cultural adaptation and validation. We included Argentinians aged 18 years or more. We used the Back-PAQ, modified Fear Avoidance Beliefs Questionnaire (mFABQ) and the Global Rating of Change (GROC) scale to assess the psychometric properties.
Results: three hundred and seventy-two participants were included for the analysis. The time taken to answer and score the questionnaire was 5.6 and 1.6 min, respectively. Neither a ceiling nor a floor effect was observed. Internal consistency was 0.76. One hundred and eighty-six participants were considered stable. Test-retest reliability was 0.90. A weak correlation (0.33) was found between the Back-PAQ and the mFABQ.
Conclusion: the Argentine version of the Back-PAQ is a viable, reliable and valid tool for the assessment of the back beliefs of the Argentine population.
The Argentine Spanish version and validation of the Back-PAQ was conducted.
The Argentine version of the Back-PAQ is a feasible tool for clinical practice.
The internal consistency and test-retest reliability were considered acceptable.
The correlation between the total score of the Back-PAQ and the mFABQ was weak.
Low back pain (LBP) is the main cause of years lived with disability both in Argentina and worldwide ( ). The lifetime prevalence of LBP is between 13.8% and 84%, and around 73% of people will suffer recurrences following a first episode ( ; ). In Argentina, LBP is also the third-leading cause of employment-associated disability and an important contributor to work absenteeism, causing an enormous economic burden ( ). Acute LBP presents a self-limited course, with the majority of the episodes resolving within the first six weeks ( ). However, 10-40% of patients have persistent LBP, and those people incur most of the social and economic burden ( ). Psychosocial factors are more strongly linked to the transition from acute to chronic back pain and disability, than biomedical or biomechanical factors ( ; ; ).
Beliefs about back pain appear important in the development and chronicity of the symptoms ( ; ). They can be influenced by the media, internet, family and particularly by health professionals ( , ; ). Unhelpful beliefs such as “the back is fragile/needs to be protected/is easy to injure” may have a negative impact on the recovery of patients with LBP and can influence the generation of fear of movement, catastrophisic thoughts and low efficacy ( ). This highlights the importance of detecting these beliefs for informing subsequent management. In recent years self-reported questionnaires have taken great prominence at the time of assessment, however, very few of these are translated and validated for languages other than English in patients with LBP ( ). The Back Pain Attitudes Questionnaire (Back-PAQ), recently developed by Darlow et al., addresses themes identified through qualitative interviews with people who had LBP ( ). These themes were: the psychological influence on recovery; the prognosis of LBP; the relationship between LBP and spinal injury; activity and participation during LBP; the vulnerability of the lumbar spine; and the special nature of LBP (its impact, difference with other pains, and need to consult with health professionals). The questionnaire has shown acceptable internal consistency and test-retest reliability and has been validated in the general population (including people with and without LBP) and amongst health professionals ( ; ). It has been translated and validated for the French population and is currently being translated into other languages ( ).
At present, there is no validated tool that assesses beliefs about the back for the Argentine population. Thus, the aim of the study was to translate into Spanish, undertake cross-cultural adaptation and validate the Back-PAQ questionnaire for the Argentine population.
Translation, cross-cultural adaptation and validation study.
Before starting with the study we obtained the permission from the developer of the Back-PAQ (e-mail communication) ( Appendix A ) and it was approved by the Research Ethics Committee of Santojanni Hospital, Autonomous City of Buenos Aires, Argentina.
The present study consisted of three consecutive phases: i) translation of the English version of the Back-PAQ into Spanish; ii) cross-cultural adaptation for the Argentinian population; and iii) subsequent testing of the psychometric properties.
Translation and cross-cultural adaptation
The translation and cross-cultural adaptation were performed according to the guideline proposed by ( Fig. 1 ).
The forward translation of the English version of the Back-PAQ was performed by 2 native bilingual translators whose mother tongue was Spanish and who worked independently from each other (Translator 1 & Translator 2). One of them had a medical background and the other did not have a medical background.
Synthesis of the translations
Both versions were compared and synthetized in one preliminary version (Translation synthesis). A written report was made documenting the synthesis of the process.
Following this, 2 bilingual translators with English as their mother tongue who did not have medical backgrounds and who were not aware of the original questionnaire made 2 different back translations from the preliminary Spanish version. (Back Translator 1 & Back Translator 2).
Later, an expert committee formed by a methodologist, health professionals, language professionals and the translators (forward and back translators) reviewed all the translations and developed the prefinal version of the questionnaire for field testing in collaboration with the original developer of the questionnaire.
Test of the prefinal version
The prefinal version was tested in a pilot study in 30 participants with the aim of evaluating the understanding of the questionnaire ( ). Following the completion of the questionnaire the participants were given a survey asking about the level of difficulty. These pilot participants were also asked to indicate any item or items that were difficult to read or interpret and given the opportunity to provide qualitative feedback.
Submission of documentation to the developers
Finally, all written informs and the process of the translation was analyzed by the developer of the questionnaire.
After the expert committee analyzed the prefinal version of the questionnaire and the results of the pilot study the final version was developed ( Appendix B ). This questionnaire will be available on the www.otago.ac.nz/backpaq site.
The website www.surveymonkey.com was used to distribute the survey.
We included Argentinians aged 18 years or older with or without LBP who gave written informed consent. We used a non-probabilistic convenience sampling inviting people to participate in the survey through social networks (WhatsApp, Facebook, Twitter, E-mail). People from other nationalities were excluded.
The Back-PAQ is a questionnaire composed of 34 Likert-style items, each with 5 possible responses. Eleven items (1, 2, 3, 15, 16, 17, 27, 28, 29, 30, 31) are scored reversed compared with the normal direction of the survey. It was designed to evaluate unhelpful beliefs about the back that could interfere with recovery from LBP. Back-PAQ total scores range from 34 to 170, with higher scores indicating more unhelpful beliefs ( ).
Modified fear avoidance beliefs questionnaire – physical activity
The Fear avoidance belief questionnaire (FABQ) is a self-reported questionnaire which focuses on how patient’s fear avoidance beliefs about physical activity (FABQ-PA) and work (FABQ-W) may affect and contribute to their LBP and resulting disability ( ). The modified version (mFABQ-PA) was design to be used with people with and without pain ( ). It consists it 4 Likert-style items, each with 7 possible responses. The total score ranges from 0 to 24 with higher scores indicating stronger fear-avoidance beliefs. The mFABQ was found to be highly correlated with the five question of the FABQ-PA (r = 0.97) ( ).
Global rating of change
A Global rating of change (GROC) was used to measure participants self-perceived change in health status over time ( ). We used a scale of perception of change with 5 categories (“much better”, “better”, “same”, “worse”, “much worse”) that has been reported elsewhere ( ). A scale with a greater number of categories would allow a more precise discrimination ( ). However, the choice of tools with multiple categories can be difficult to interpret for patients who are not accustomed to the use of this type of tool. We decided to use this scale since it is recommended by researchers to qualify the self-perception of the change ( ).
During the initial evaluation (T1), the participants answered demographic questions and completed the following questionnaires: Back-PAQ and mFABQ. Within 48 h of completing T1, the participants received an e-mail inviting them to complete the second evaluation (T2), composed of the GROC and the Back-PAQ. During the period of time between T1 and T2, the participants did not receive any treatment in order to maintain clinical stability ( Fig. 2 ) . Researchers were trained in scoring of the questionnaires before the study began.
All data was entered into a Microsoft Excel ® spreadsheet.
Measures if the questionnaire is feasible to be used in the field you want to use it ( ). The time required to answered the scale by the participants and the time required to manually score it by the physician were timed in the pilot study. The understanding of the questionnaire was tested in the pilot study with a survey ( ).
Floor and ceiling effect
The floor and ceiling effect were present if more than 15% of the participants scored the minimum (34) and maximum (170) value in T1 ( ).
Internal consistency describes the extent to which all the items in a test measure the same concept or construct and hence it is connected to the inter-relatedness of the items within the test. Internal consistency should be determined before a test can be employed for research or examination purposes to ensure validity. It was assessed with Cronbach’s alpha coefficient of the T1 Back-PAQ scores. Acceptable values of Cronbach’s Alpha were those between 0.70 and 0.95 ( ; ).
Item-item and item-total correlation were assessed by Pearson or Spearman correlation coefficients as appropriate. Values that ranged from 0.15 to 0.85 were considered acceptable for item-item correlations ( ). Item-total correlations were controlled to avoid “overlapping”. Values ≥ 0.50 were considered acceptable for item-total correlations ( ; ).
Test Re-Test reliability
Test-retest reliability is the extent to which the same results are obtained on repeated administrations of the same self-reported outcome measure when no change in clinical status has occurred. The GROC was used to check for changes in perceived health status between the 2 test occasions. Participants who answered “same” in the GROC were considered stable and were included for the test-retest reliability analysis. This assured that there were not changes in the health status of participants ( ). The second administration of the questionnaire (T2) was done 48 h after the first administration (T1). The intraclass correlation coefficient (ICC) and 95% confidence intervals were calculated ( ). We used a two-way random effects model (ICC 2,1 ). An ICC >0.75 was considered acceptable ( ). We used a Bland-Altman plot to report limits of agreement ( ).
The validity determines the extent to which the questionnaire measures the construct(s) it purports to measure. Construct validity refers to the extent to which scores on a particular measure relate to other measures, consistent with theoretically derived hypotheses concerning the constructs that are being measured ( ). Concurrent validity was evaluated by comparing Back-PAQ baseline scores with the mFABQ-PA baseline scores. We used Pearson correlation coefficient and Spearman correlation coefficient as appropriate. The correlation was categorized as very weak (0–0.19), weak (0.20–0.39), moderate (0.40–0.69), strong (0.70–0.89) and very strong (0.90–1) ( ; ). We used criteria from De Boer et al. to establish construct validity ( ). It would be acceptable if the hypotheses are specified in advance and at least 75% of the results are in correspondence with these hypotheses ( ). We hypothesized weak positive correlation between the Back-PAQ and the mFABQ-PA.
According to the literature it is recommended to include 10 subjects for each item evaluated and at least 100 subjects. As the Back-PAQ has 34 items, the targeted sample was 340 subjects. This sample size is considered excellent for determining internal consistency by the factorial analysis method ( ).
All statistical analyses were performed with IBM SPSS Macintosh, version 24.0 (IBM Corp., Armonk, NY, USA). Continuous data that assumed a normal distribution were reported as mean and standard deviation (SD). Otherwise, the median and interquartile range (IQR) were used. Categorical variables were reported as presentation number and percentage (%). To determine the distribution of the sample, the Shapiro-wilk or Kolmogorov Smirnov test were used as appropriate.
The significance level was set at < 0.05.
Translation and cross-cultural adaptation
The preliminary version of the Spanish translation was compared with the original questionnaire. Minor discrepancies between the preliminary version and the original version, were found concerning wording, understanding, and phrasing. These discrepancies were found to be small and were discussed and solved with the developer of the questionnaire (BD) aiming for better understanding. For example, the word “twinge” in item #14 did not have an exact translation into Spanish, so we decieded to use “dolor punzante” since its meaning was the closest to the original word. During the pilot test two participants reported difficulty with question #26. This was modified for better understanding, after discussion with the original author and an expert committee, using “ejercicio intenso” instead of “ejercicio vigoroso”. Regarding the difficulty of the questionnaire 43% answered it was normal, 40% easy and 16% very easy. No one rated it as hard or very hard.
A total of 387 individuals answered the Back-PAQ, 15 were excluded based on non-Argentine nationality and 153 did not complete the second evaluation ( Fig. 3 ). The demographic and clinical data are shown in Table 1 . There were no important differences between those who did and did not complete the second evaluation, except those who did not respond were more likely to have current pain.