The present study was a methodological study which was conducted in two stages. The first stage was the translation and cultural adaptation of the English version of the PCAS-9. The second stage of validation (face validity, content validity, construct validity, and reliability) dealt with the Persian version of PCAS-9. The samples in the translation stage were two fluent translators in Persian and English for forward translation and two translators for backward translation. In the psychometric phase, the study samples were 10 patients in need of palliative care were selected for face validation, 10 specialists in palliative care and psychometrics for content validation, 510 patients admitted to different wards of hospitals affiliated to Shahid Sadoughi University of Medical Sciences in Yazd/Iran for assessing the construct validity, and 30 patients for evaluating the reliability. Criteria for entering construct validity were having literacy, willingness to participate in the study, having a chronic illness requiring palliative care; exclusion criteria included having psychological problems such as communication and perceptual disorders and unwillingness to participate in the study.
Translation and Cultural Adaptation
At first, the instrument was translated and culturally adapted. After obtaining permission from the designer of the PCAS-9, Laura Perry, the translation was performed based on the approach proposed by Polit and Yang . Two experienced translators fluent in English with native Persian language first translated the questionnaire into Persian. Then, the expert committee compared and combined both versions. The final translation of the PCAS-9 was translated back into English by two different translators. A final reconciliation was done with a committee of experts and the original designer.
In the next step, face validity (qualitative), content validity (qualitative) and construct validity were examined . The translated instrument was distributed among 10 randomly selected patients to express their opinions on the difficulty and ambiguity about each item (qualitative face validity). In order to assess the content validity, 10 experts in the fields of palliative care and development of instruments were asked to present their corrective opinions in terms of grammar, use of appropriate words, and placement of items in their proper place in detail and in writing . After carefully studying their views, appropriate corrections were made by the present research group. The comments were assessed and 9 items was revised.
The construct validity phase was performed by exploratory and confirmatory factor analysis on a sample size of 510 people. For large-scale studies, the sample size should have 10-20 participants per instrument phrase. In other words, for large-scale studies, the sample size should be between 100 and 250 people . Thus, 300 patients were randomly selected for the exploratory factor analysis (EFA). The fit of the samples was evaluated using the Kaiser–Meyer–Olkin (KMO) index, which was equal to 0.805. The KMO rate varies from 0 to 1. The higher the rate, the more appropriate the factor analysis. Values above 0.9 are excellent and 0.8 are good [27, 28]. Then, Bartlett's test was performed which was significant with chi-square (χ2) 1295.04 (P<0.001). Therefore, the samples had the necessary appropriateness and minimum requirements for performing EFA, and there was sufficient correlation between items to perform factor analysis. In EFA, the principal components method has been used to extract the factors and the Varimax method with Kaiser normalization has been used for factor rotation. Thus, if the value of the extracted share of each item is less than 0.5, we exclude that item from factor analysis. Also, the criterion for deciding on the classification of items is special values higher than 1 and factor scores higher than 0.4 .
In the next step, the final number of factors was confirmed using CFA. In so doing, 210 participants were examined in confirmatory factor analysis (CFA). To judge the fit of the model, from indices of χ2/degree of freedom (df), Goodness of Fit Index (GFI), Adjusted Goodness of Fit Index (AGFI), Root Mean square Residual (RMR), Normed Fit Index (NFI), Non-Normed Fit Index (NNFI), Incremental Fit Index (IFI), Comparative Fit Index (CFI) and Root Mean Square Error of Approximation (RMSEA) were used .
In the present study, reliability was investigated using internal consistency and stability. The instrument was completed by 30 patients and re-completed at 2-week intervals. The most acceptable statistical test to calculate the reliability of stability is the intra-class coefficient correlation (ICC) . If this index is higher than 0.7, the stability rate is desirable .
Internal consistency is the most widely used method of assessing reliability. This type of reliability is used to test the correlation of different items in the tool. The most common method for determining internal consistency is the Cronbach’s alpha coefficient. The optimal Cronbach's alpha coefficient is between 0 and 1, wherein a high score indicates a higher internal consistency. For interpretation, alpha coefficient values above 0.75 are considered acceptable .
Data collection was performed during July-November 2021. Data were analyzed using SPSS 20 and LISERL8.5.