A pilot study on the Spanish version of the Psychosocial Adjustment to Illness Scale (PAIS‐SR) with carers of people with Parkinson's disease

Abstract Aim To report the cross‐cultural adaptation and pilot study of the ongoing validation of the Spanish version of the Psychosocial Adjustment to Illness Scale with carers of people with Parkinson's disease. Design Cross‐cultural adaptation and pilot study with a cross‐sectional validation design of the Spanish version of the Psychosocial Adjustment to Illness Scale – Carers. Methods Twenty‐one carers of people with Parkinson's disease from a Primary Care practice in Spain were recruited and completed the PAIS‐Carers, the SF‐36 Health Survey, the Brief COPE Inventory and an assessment form. SPSS 23.0 was used to determine viability/acceptability and preliminary aspects of internal consistency of the instrument. Results Five of the seven domains presented floor effect (71.42%), and only one presented ceiling effect (14.28%). The internal consistency of the scale and domains showed acceptable values (over 0.7). The content validity of the Spanish version seemed satisfactory with positive comments in general from participants.


| BACKG ROU N D
Several studies have concluded that, although illness-related symptoms or stress can influence the psychosocial adjustment to a longterm condition (LTC) like PD, most of the times are factors related to family and social support, personality, expectations of illness management, financial status or coping skills that become paramount and transversal across conditions (Stanton, Revenson, & Tennen, 2007;Wang et al., 2014). Interestingly, these non-illness-related factors also gain importance for family carers and Navarta-Sánchez et al. (2016) concluded that both people with PD and carers' quality of life is clearly influenced by the psychosocial adjustment to illness, which significantly depends on coping skills, regardless the disabling and degenerative nature of PD (Navarta-Sanchez et al., 2017). Consequently, assessments and interventions should not only focus on the patient but also on the family, since it has been established that their experience of the adaptation process is comparable and that similar factors could influence this adjustment (Årestedt, Benzein, & Persson, 2015;Årestedt, Benzein, Persson, & Rämgård, 2016;Årestedt, Persson, & Benzein, 2014;Golics, Basra, Salek, & Finlay, 2013).
In this regard, the Psychosocial Adjustment to Illness Scale (PAIS-SR) (Derogatis, 1986;Derogatis & Derogatis, 1990) evaluates the psychosocial adjustment process of a person with a health condition and/or its consequences and has versions for patients and carers.
The scale is worldwide known and has been validated in several languages with people with different LTCs. Nevertheless, the scale has not been validated in a PD population and there is not a carer's version available in Spanish, despite the high relevance of this scale for carers of people living with PD and its implications for clinical use in Spanish-speaking populations. This paper aims to present a brief report of the cross-cultural adaptation and pilot study of the ongoing validation of the Spanish version of the PAIS-SR with carers of people with PD.

| Design
We present results from a pilot study with a cross-sectional observational design reporting the psychometric properties of the Spanish version (self-report) of the PAIS-SR with carers of people with PD.

| Cross-cultural adaptation process
After obtaining written permission from the author of the original PAIS-SR, the translation of the English original version of the scale into Spanish was performed by a panel of four experts following the standard protocols used for transcultural adaptation of psychology questionnaires (Bonomi et al., 1996;Eremenco, 1998;Wild et al., 2005) (Table 1).

| Setting, sampling and sample
The recruitment of participants took place in a Primary Health Centre of the northern region in Spain. Carers of people with PD living in the community and meeting the inclusion criteria in Table 2 were consecutively selected (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996;Stebbing, 2012).

| Data collection and instruments
The pilot study was completed in 2016, and the main validation study is under analysis. Data collection was planned as self-reported.
However, at all times, researchers were available to support participants should they have any question.

TA B L E 1 Cross-cultural adaptation of psychology questionnaires. Steps applied in this study
Steps Process Outcomes of the process Step 1. Two forward translations from English ("ENGLISH 1") into Spanish Two independent translators Translations "SPANISH 1" and "SPANISH 2" Step 2. Reconciliation of the two Spanish versions.
Same independent translators compare the two versions (differences and wording refining) Reconciled version "SPANISH 3" Step 3. Back translation Other bilingual translator with no previous contact translates to English The resulting English version is "ENGLISH 2" Step 4. Comparison of all versions in English and Spanish All versions used previously (ENGLISH 1, SPANISH 2, SPANISH 3, and ENGLISH 2) were studied to find inaccuracies in the forward reconciled translation At this point, discrepancies individually found between the two versions were discussed A more refined version (SPANISH 4) Step 5. SPANISH 4 version of the scale reviewed Spanish native expert naïve to the original version -but familiar to psychological scales-, reviewed the version to ensure natural wording Two of the translators previously involved also reviewed the Spanish 4 version for a natural wording Minor changes to the previous version were made, obtaining version SPANISH 5 Step 6. Pilot study of SPANISH 5 version

Research study team
The SPANISH 5 version was tested in the pilot study as explained in this paper

Carer's self-report version of the PAIS-SR
The scale has 46 items with Likert-type answers grouped under a total of seven domains (Health Care Orientation, Vocational Environment,

The 36-item short form Health Survey (SF-36)
The scale has 36 items with Likert-type answers looking at positive and negative health aspects (Alonso, Prieto, & Anto, 1995;Vilagut et al., 2005) happening over the last 4 weeks in relation to physical functioning, role limitations due to physical health, role limitations due to emotional problems, energy/fatigue, emotional well-being, social functioning, pain and general health.

Brief COPE scale (self-report)
This is a multidimensional instrument with Likert-type answers looking at different responses to stress (Carver, 1997). It contains 24 items under 12 subscales, which are self-distraction, active coping, denial, substance use, use of emotional support, behavioural disengagement, venting, positive reframing, planning, humour, acceptance and religion.

Evaluation and sociodemographic forms
Apart from a sociodemographic form, participants completed an evaluation form to determine whether they had understood all the items, had found anything irrelevant or offensive and whether they had any comment or suggestion for additional questions to be included.

| Data analysis
Data did not follow a normal distribution, and the following nonparametric statistics were applied to test the indicated attributes using SPSS 23.0. To determine the viability and acceptability of the cross-culturally adapted scale, we analysed data quality registering missing data (accepting more than 95% of computable data). The limit for missing data was <5% in our study (Smith et al., 2005).
Furthermore, the distribution of the punctuations with parameters inter-item correlation (criterion value r ≥ 0.20 and ≤0.75) (Piedmont, 2014;Smith et al., 2005). Spearman rank correlation coefficient (rS) was used for testing these associations.
Utility and content validity. The percentage of responses completed, time taken for completion and perceptions of carers that participated in the pilot study were registered. Furthermore, the content validity was enhanced by following the cross-cultural adaptation process of the scale where also experts from neurology and psychosocial adjustment to illness fields were involved.

| Ethical considerations
This study was approved by the Ethics Committee of the University of Navarra in Spain (reference 111/2013). Participants were accessed through the healthcare professionals in charge of their medical assistance in the centre under study. An informative letter and a consent form were provided.

| RE SULTS
Out of the 29 family carers invited to the study, 21 accepted to participate (see sociodemographic characteristics in Tables 3 and 4) There were no missing data, and data of items and domains were 100% computable. The mean scores in the scales were 46.1 (SD 8.9) for the Brief COPE; 69 (SD 25.2) for the SF-36; and 32.7 (SD 17.1) for the PAIS-Carers (Tables 4 and 5).
Looking further at the PAIS-SR acceptability (Table 5) As shown in Table 5    Although the cross-cultural adaptation process was developed rigorously according to international standards and the participants did not report any difficulty in understanding the Spanish version, Time for reflection is needed as for its cost-effectiveness and its use in practice nowadays. Finally, the PAIS-SR could be considered more adequate for other type of LTCs with no degenerative progression, especially when it comes to overburdened carers. These aspects remained open, and at this stage, we need to wait for the results of the full validation study.

| LI M ITATI O N S
This is the first study which has adapted the carers' version of the PAIS-SR to Spanish with carers of people with PD. This is a pilot study taking place in a small geographical locality of Spain and does not represent the whole population of carers of people with PD. However, this is considered sufficient for a pilot testing and the main validation study is taking place at a national level, overcoming this limitation.

ACK N OWLED G EM ENTS
Thanks are given to the Ministry of Economy and Competitiveness in Spain for the financial support in the development of this project.
Special considerations are in place for Dr Marta García Granero for her statistical advice and work within this project.

CO N FLI C T O F I NTE R E S T
No conflict of interest is declared.