Clin Rheumatol (2014) 33:1751–1757 DOI 10.1007/s10067-013-2429-6
ORIGINAL ARTICLE
Translation into Brazilian Portuguese, cross-cultural adaptation and validation of the Stanford presenteeism scale-6 and work instability scale for ankylosing spondylitis Renata Frauendorf & Marcelo de Medeiros Pinheiro & Rozana Mesquita Ciconelli
Received: 11 March 2013 / Revised: 16 September 2013 / Accepted: 24 October 2013 / Published online: 13 November 2013 # Clinical Rheumatology 2013
Abstract Loss of productivity at work, as a result of health problems, is becoming an issue of interest due to the high burden it represents in society. The measurement of such phenomenon can be made using generic and specific scales for certain diseases such as the Stanford Presenteeism Scale (SPS-6) and the Work Instability Scale for Ankylosing Spondylitis (AS-WIS), specific for patients with ankylosing spondylitis (AS). The aim of this study was to translate and perform a cross-cultural adaptation of SPS-6 and AS-WIS into Portuguese and check their psychometric properties. The study also aimed to evaluate the relationship between the general scores of the scales and the main sociodemographic and clinical data, lifestyles, and absenteeism in patients with AS and correlate these variables with SPS-6 and AS-WIS scales. A sample of 120 patients with AS and 80 workers at a university hospital was evaluated. The processes for the translation and cross-cultural adaptation of the instruments followed preestablished steps and rules presented in the literature. For the evaluation of measurement properties and correlations between scales, intra-class correlation coefficient (reproducibility analysis), Cronbach alpha (internal consistency), and Pearson correlation coefficient (validity) were R. Frauendorf Translational Medicine, UNIFESP/EPM, São Paulo, São Paulo, Brazil M. de Medeiros Pinheiro (*) Spondyloarthritis Outpatient Clinic, Department of Rheumatology, UNIFESP/EPM, São Paulo, São Paulo, Brazil e-mail: [email protected] M. de Medeiros Pinheiro Avenida Dr. Altino Arantes, 669, Ap. 105 Vila Clementino, 04042-033 São Paulo, São Paulo, Brazil R. M. Ciconelli Department of Rheumatology, UNIFESP/EPM, São Paulo, SP, Brazil
employed. The inter-observer (0.986) and intra-observer (0.992) reproducibilities of the AS-WIS were shown to be high as well as the internal consistency (0.995). Similarly, the inter-observer reliability of SPS-6 was considered good (0.890), although it showed a poorer performance when considering the same observer (Pearson correlation coefficient= 0.675 and intra-class correlation=0.656). Internal consistency, for the total number of items, as measured by Cronbach alpha, was 0.889. The validity of the scales was evaluated thru the comparison of the achieved scores with the results of the WLQ, SF-36, ASQoL, BASFI, BASDAI, HAQ-S, and SRQ-20 instruments. Correlations between loss of productivity at work, worse quality of life, presence of emotional disturbances, and worse health conditions were positive. The process of translation, cross-cultural adaptation, and validation of the SPS-6 as a generic measurement for the loss of productivity at work and of the AS-WIS as a specific measurement for patients with AS are valid, reproducible, and specific instruments to be used in Brazil. In both scales, productivity at work was associated to advanced age, higher rate of absenteeism in the last month and year, presence of peripheral arthritis, and a larger number of comorbidities in patients with AS. The AS-WIS and SPS-6 showed a good correlation among them although they are not mutually exclusive but supplementary. Keywords Ankylosing spondylitis . Cost efficiency analysis . Validation studies . Worker's health
Introduction Loss of productivity at work as a result of health problems is becoming an issue of great interest due to the high burden it represents in the society [1]. Costs of disease can be direct or indirect. Direct costs usually result from medical interventions,
1752
diagnostic procedures, hospitalization, and expenditures made by the patients with their treatment. The main indirect costs are related to reduction of the ability to work and involve phenomena known as absenteeism and presenteeism [2]. Presenteeism is considered as being the main factor related to loss of productivity at work (LPW) and consists in the experience of the worker who cannot perform the tasks in their entirety, due to physical and psychological reasons, thus characterizing functional absence although physically present [3, 4]. The generic measurement of this phenomenon can be carried out by scales and self-informed questionnaires that capture the subjective perception that workers have on their health and how much it interferes on their productivity [5]. In Brazil, there are only two instruments that have been translated, cross-culturally adapted, and validated to measure presenteeism: Work Productivity and Activity Impairment Questionnaire-General Health (WPAI-GH) [6] and Work Limitations Questionnaire (WLQ) [1]. To date, according to the literature, there are only a few studies that have performed the translation, cross-cultural adaptation, and validation of these scales as well as have not assessed its psychometric properties for the Brazilian population [7]. The aim of the present study was to translate and perform a cross-cultural adaptation of the Standard Presenteeism Scale-6 (SPS-6) and Work Instability Scale for Ankylosing Spondylitis (AS-WIS) to Portuguese language and also to assess their psychometric properties. The research also evaluated the relationship between the general scores of the scales and the main sociodemographic, clinical, lifestyle, and absenteeism data in patients with ankylosing spondylitis (AS) and correlated these variables with SPS-6 and AS-WIS.
Patients and methods The study was authorized by Stanford University, Leeds University, and the Ethics Committee of the Federal University of São Paulo-Paulista School of Medicine (UNIFESP-EPM). All research volunteers read, agreed to, and signed the informed consent form. A total of 80 employees of the São Paulo Hospital, a facility belonging to UNIFESP-EPM, and 120 AS patients were evaluated in this study [8]. The study included subjects over 18 years of age who were working in the month that preceded the research. Patients with other autoimmune or disabling diseases who have overcome AS were excluded. The patients included in the study were active in the formal or informal labor market during the evaluation period. Patients with AS were evaluated at the Rheumatology Outpatient Clinic. Employees answered the questions in their own working environment. The questionnaires were applied
Clin Rheumatol (2014) 33:1751–1757
individually. The forms were self-administered to individuals who had at least finished high school. An interviewer was present only in the cases in which the instruments were not self-administered. SPS-6 and AS-WIS The Stanford Presenteeism Scale-6 or SPS-6 is a generic scale that evaluates presenteeism, and it is an abridged version of the Stanford Presenteeism Scale-32. Its original version had been developed and validated by Koopman et al. in 2002 [9]. It intends to measure the ability of the individual to concentrate and to perform work activities in spite of health problems, based on questions that involve cognitive, emotional, and behavioral aspects. The structure of the scale integrates two dimensions of presenteeism, with three items each. The first one is associated to the construct of “completed work” and relates to the amount of work performed when the worker is under the influence of presenteeism causes. The second one relates to the “concentration ability” also under the influence of presenteeism causes. The total sum of the six items allows us to achieve a presenteeism score that may vary from 6 to 30 points. Low scores indicate low presenteeism and reduced productivity, while high scores indicate high presenteeism and productivity [9]. The Work Instability Scale for Ankylosing Spondylitis or AS-WIS is the only specific instrument to evaluate the productivity of patients with AS [10]. This is a chronic inflammatory disease that affects 0.1 to 1.4 % of the general population and is characterized by the preferential involvement of the entheses, mainly in the spine [10, 11]. In general, it affects young adults in productive age, which represents a heavy economic burden since it causes functional disability, chronic pain, and quality of life impairment [12]. The available data for the AS-WIS were associated to the items of yet another scale developed by the same group of researchers, the Work Instability Scale for Rheumatoid Arthritis (RA-WIS) [13]. Translation and cross-cultural adaptation The process of translation and cross-cultural adaptation was carried out according to the rules established in the literature by Beaton et al. [14]. For each one of the scales, two translations were carried out by independent Brazilian certified translators with great proficiency in the English language. These versions were compared, and the discrepancies were discussed with the translators. At the end of this process, versions T1 and T2 of the initial translation of SPS-6 and AS-WIS were achieved. These versions were discussed by the researchers together with the two translators. The consensus of this discussion led to versions T12 of the two scales.
Clin Rheumatol (2014) 33:1751–1757
The T12 versions, in Brazilian Portuguese, were sent to other two independent certified translators who had English as their mother language and Brazilian citizenship. Both translators had no access to the original scales. Each one translated the scales back to English, therefore leading to back translation 1 (BT1) and back translation 2 (BT2) of SPS-6 and ASWIS. The questions that presented any kind of disagreement were sent to the original authors of the scales for clarifications. Based on yet another discussion involving translators and researchers, BT12 versions of the scales were produced. The committee of experts was composed of two medical researchers, a Portuguese language teacher, an English teacher, and the translators involved in the process. Semantic, idiomatic, conceptual, and cross-cultural equivalences of the items of each scale were evaluated, leading to the final version (FV) of each one of the scales in Brazilian Portuguese. The FVof ASWIS was applied to a group of 30 patients with AS. The FVof SPS-6 was applied to a group of 15 patients with the same characteristics and to 15 employees of UNIFESP-EPM.
1753
Statistical analysis Descriptive statistics methods were used to characterize the sample, and the properties of the scales were measured using intra-class correlation coefficients and their respective p values (significance levels) and confidence intervals (CI) (for reproducibility analysis), the Cronbach alpha coefficient (internal consistency), and the Pearson correlation coefficient (validity). In order to achieve the first specific goal, an analysis of variance (ANOVA) was carried out on the numeric variables and a chi-square test on the category variables. The results were interpreted using a significance level of 5 %. Correlation coefficients ranging from 0.800 to 1.00 were considered as very high; 0.600 to 0.800, high; 0.400 to 0.600, moderate; and
ORIGINAL ARTICLE
Translation into Brazilian Portuguese, cross-cultural adaptation and validation of the Stanford presenteeism scale-6 and work instability scale for ankylosing spondylitis Renata Frauendorf & Marcelo de Medeiros Pinheiro & Rozana Mesquita Ciconelli
Received: 11 March 2013 / Revised: 16 September 2013 / Accepted: 24 October 2013 / Published online: 13 November 2013 # Clinical Rheumatology 2013
Abstract Loss of productivity at work, as a result of health problems, is becoming an issue of interest due to the high burden it represents in society. The measurement of such phenomenon can be made using generic and specific scales for certain diseases such as the Stanford Presenteeism Scale (SPS-6) and the Work Instability Scale for Ankylosing Spondylitis (AS-WIS), specific for patients with ankylosing spondylitis (AS). The aim of this study was to translate and perform a cross-cultural adaptation of SPS-6 and AS-WIS into Portuguese and check their psychometric properties. The study also aimed to evaluate the relationship between the general scores of the scales and the main sociodemographic and clinical data, lifestyles, and absenteeism in patients with AS and correlate these variables with SPS-6 and AS-WIS scales. A sample of 120 patients with AS and 80 workers at a university hospital was evaluated. The processes for the translation and cross-cultural adaptation of the instruments followed preestablished steps and rules presented in the literature. For the evaluation of measurement properties and correlations between scales, intra-class correlation coefficient (reproducibility analysis), Cronbach alpha (internal consistency), and Pearson correlation coefficient (validity) were R. Frauendorf Translational Medicine, UNIFESP/EPM, São Paulo, São Paulo, Brazil M. de Medeiros Pinheiro (*) Spondyloarthritis Outpatient Clinic, Department of Rheumatology, UNIFESP/EPM, São Paulo, São Paulo, Brazil e-mail: [email protected] M. de Medeiros Pinheiro Avenida Dr. Altino Arantes, 669, Ap. 105 Vila Clementino, 04042-033 São Paulo, São Paulo, Brazil R. M. Ciconelli Department of Rheumatology, UNIFESP/EPM, São Paulo, SP, Brazil
employed. The inter-observer (0.986) and intra-observer (0.992) reproducibilities of the AS-WIS were shown to be high as well as the internal consistency (0.995). Similarly, the inter-observer reliability of SPS-6 was considered good (0.890), although it showed a poorer performance when considering the same observer (Pearson correlation coefficient= 0.675 and intra-class correlation=0.656). Internal consistency, for the total number of items, as measured by Cronbach alpha, was 0.889. The validity of the scales was evaluated thru the comparison of the achieved scores with the results of the WLQ, SF-36, ASQoL, BASFI, BASDAI, HAQ-S, and SRQ-20 instruments. Correlations between loss of productivity at work, worse quality of life, presence of emotional disturbances, and worse health conditions were positive. The process of translation, cross-cultural adaptation, and validation of the SPS-6 as a generic measurement for the loss of productivity at work and of the AS-WIS as a specific measurement for patients with AS are valid, reproducible, and specific instruments to be used in Brazil. In both scales, productivity at work was associated to advanced age, higher rate of absenteeism in the last month and year, presence of peripheral arthritis, and a larger number of comorbidities in patients with AS. The AS-WIS and SPS-6 showed a good correlation among them although they are not mutually exclusive but supplementary. Keywords Ankylosing spondylitis . Cost efficiency analysis . Validation studies . Worker's health
Introduction Loss of productivity at work as a result of health problems is becoming an issue of great interest due to the high burden it represents in the society [1]. Costs of disease can be direct or indirect. Direct costs usually result from medical interventions,
1752
diagnostic procedures, hospitalization, and expenditures made by the patients with their treatment. The main indirect costs are related to reduction of the ability to work and involve phenomena known as absenteeism and presenteeism [2]. Presenteeism is considered as being the main factor related to loss of productivity at work (LPW) and consists in the experience of the worker who cannot perform the tasks in their entirety, due to physical and psychological reasons, thus characterizing functional absence although physically present [3, 4]. The generic measurement of this phenomenon can be carried out by scales and self-informed questionnaires that capture the subjective perception that workers have on their health and how much it interferes on their productivity [5]. In Brazil, there are only two instruments that have been translated, cross-culturally adapted, and validated to measure presenteeism: Work Productivity and Activity Impairment Questionnaire-General Health (WPAI-GH) [6] and Work Limitations Questionnaire (WLQ) [1]. To date, according to the literature, there are only a few studies that have performed the translation, cross-cultural adaptation, and validation of these scales as well as have not assessed its psychometric properties for the Brazilian population [7]. The aim of the present study was to translate and perform a cross-cultural adaptation of the Standard Presenteeism Scale-6 (SPS-6) and Work Instability Scale for Ankylosing Spondylitis (AS-WIS) to Portuguese language and also to assess their psychometric properties. The research also evaluated the relationship between the general scores of the scales and the main sociodemographic, clinical, lifestyle, and absenteeism data in patients with ankylosing spondylitis (AS) and correlated these variables with SPS-6 and AS-WIS.
Patients and methods The study was authorized by Stanford University, Leeds University, and the Ethics Committee of the Federal University of São Paulo-Paulista School of Medicine (UNIFESP-EPM). All research volunteers read, agreed to, and signed the informed consent form. A total of 80 employees of the São Paulo Hospital, a facility belonging to UNIFESP-EPM, and 120 AS patients were evaluated in this study [8]. The study included subjects over 18 years of age who were working in the month that preceded the research. Patients with other autoimmune or disabling diseases who have overcome AS were excluded. The patients included in the study were active in the formal or informal labor market during the evaluation period. Patients with AS were evaluated at the Rheumatology Outpatient Clinic. Employees answered the questions in their own working environment. The questionnaires were applied
Clin Rheumatol (2014) 33:1751–1757
individually. The forms were self-administered to individuals who had at least finished high school. An interviewer was present only in the cases in which the instruments were not self-administered. SPS-6 and AS-WIS The Stanford Presenteeism Scale-6 or SPS-6 is a generic scale that evaluates presenteeism, and it is an abridged version of the Stanford Presenteeism Scale-32. Its original version had been developed and validated by Koopman et al. in 2002 [9]. It intends to measure the ability of the individual to concentrate and to perform work activities in spite of health problems, based on questions that involve cognitive, emotional, and behavioral aspects. The structure of the scale integrates two dimensions of presenteeism, with three items each. The first one is associated to the construct of “completed work” and relates to the amount of work performed when the worker is under the influence of presenteeism causes. The second one relates to the “concentration ability” also under the influence of presenteeism causes. The total sum of the six items allows us to achieve a presenteeism score that may vary from 6 to 30 points. Low scores indicate low presenteeism and reduced productivity, while high scores indicate high presenteeism and productivity [9]. The Work Instability Scale for Ankylosing Spondylitis or AS-WIS is the only specific instrument to evaluate the productivity of patients with AS [10]. This is a chronic inflammatory disease that affects 0.1 to 1.4 % of the general population and is characterized by the preferential involvement of the entheses, mainly in the spine [10, 11]. In general, it affects young adults in productive age, which represents a heavy economic burden since it causes functional disability, chronic pain, and quality of life impairment [12]. The available data for the AS-WIS were associated to the items of yet another scale developed by the same group of researchers, the Work Instability Scale for Rheumatoid Arthritis (RA-WIS) [13]. Translation and cross-cultural adaptation The process of translation and cross-cultural adaptation was carried out according to the rules established in the literature by Beaton et al. [14]. For each one of the scales, two translations were carried out by independent Brazilian certified translators with great proficiency in the English language. These versions were compared, and the discrepancies were discussed with the translators. At the end of this process, versions T1 and T2 of the initial translation of SPS-6 and AS-WIS were achieved. These versions were discussed by the researchers together with the two translators. The consensus of this discussion led to versions T12 of the two scales.
Clin Rheumatol (2014) 33:1751–1757
The T12 versions, in Brazilian Portuguese, were sent to other two independent certified translators who had English as their mother language and Brazilian citizenship. Both translators had no access to the original scales. Each one translated the scales back to English, therefore leading to back translation 1 (BT1) and back translation 2 (BT2) of SPS-6 and ASWIS. The questions that presented any kind of disagreement were sent to the original authors of the scales for clarifications. Based on yet another discussion involving translators and researchers, BT12 versions of the scales were produced. The committee of experts was composed of two medical researchers, a Portuguese language teacher, an English teacher, and the translators involved in the process. Semantic, idiomatic, conceptual, and cross-cultural equivalences of the items of each scale were evaluated, leading to the final version (FV) of each one of the scales in Brazilian Portuguese. The FVof ASWIS was applied to a group of 30 patients with AS. The FVof SPS-6 was applied to a group of 15 patients with the same characteristics and to 15 employees of UNIFESP-EPM.
1753
Statistical analysis Descriptive statistics methods were used to characterize the sample, and the properties of the scales were measured using intra-class correlation coefficients and their respective p values (significance levels) and confidence intervals (CI) (for reproducibility analysis), the Cronbach alpha coefficient (internal consistency), and the Pearson correlation coefficient (validity). In order to achieve the first specific goal, an analysis of variance (ANOVA) was carried out on the numeric variables and a chi-square test on the category variables. The results were interpreted using a significance level of 5 %. Correlation coefficients ranging from 0.800 to 1.00 were considered as very high; 0.600 to 0.800, high; 0.400 to 0.600, moderate; and
