Background/Purpose: Systemic Sclerosis (SSc) represents a group of heterogeneous autoimmune disorders of connective tissue characterized by progressive fibrosis of the skin and internal organs [1]. The disease spectrum ranges from localized form, which causes physical disability to more systemic forms affecting internal organs and the vasculature, which may result in life threatening crises. The SSc Quality of Life Questionnaire (SScQoL) [2] is a 29-item, needs-based tool, which measures the disease impact on health and well being of patients with SSc. The aim of this study was to adapt the English SScQoL into Spanish and to determine measurement equivalence between the two versions of the tool.

Methods: This was a cross-sectional tool validation study involving cross-cultural adaptation and cross-cultural validation phases. In the first phase, we adapted the English SScQoL into Spanish using a 5-stage process involving: forward translation, synthesis of translations, backward translation, expert committee revision and pilot testing with 30 patients [3]. This process ensured conceptual equivalence between the English and the Spanish version of the SScQoL. For the cross-cultural validation phase, we recruited adult patients with SSc who were willing and able to complete the questionnaire unaided. We collected the data from both versions of SScQoL and analysed them by comparing with the Rasch model using fit statistics. Fit to the Rasch model implies construct validity, unidimensionality, reliability and statistical sufficiency. To discount local dependency, we grouped items into the following subscales: function, emotion, sleep, social and pain and repeated Rasch analysis at a subscale level. Finally, we pooled both the UK and the Spanish data in order to test measurement equivalence between the two cultures and to calibrate the SScQoL into a logit-based interval scale.

Results: The SScQoL translated well from English into Spanish providing a conceptually equivalent version of the tool. For the validation sample comprised 121 patients from the UK (male/female = 15/106) and 106 from Spain (male/female = 16/87). Their mean (SD) age was 57.1(12.1) and 58.0 (13.9) for the UK and Spain respectively. Table 1 presents item parameters, where non-significant Chi-Square probabilities indicate that most items did not deviate from the Rasch model although local dependency was evident (data not shown). Items were grouped into the afore-mentioned 5 subscales and re-analysed, resulting in adequate fit to Rasch model. See table 2.  There was no differential item functioning by culture, thus allowing for a common scale, which can be converted into logit-based transformed scores for use in parametric analyses when required.

Conclusion:  The SScQoL was adapted successfully into Spanish and satisfied the strict requirements of the Rasch measurement model, thus establishing its cross-cultural validity.  The SScQoL data obtained from Spain and the UK are therefore comparable. Further research is required to determine cross-cultural validity of the SScQoL in the American Spanish populations.

Table 1: Results of item analysis using Rasch models

Fit residual within +/-2.5 and non-significant p-value (>0.001 with Bonferroni correction) suggest fit to Rasch model. Table 2: Summary fit statistics for the 5-subscale SScQoL

SD, Standard deviation; DF, Degrees of freedom; *P-value = Chi-Square probability, where  >0.05 (>0.01 for Bonferroni correction) suggest adequate fit to the model; PSI, Person separation index reliability; **Less than 5% significant independent t-tests suggest unidimensionality [4].