Reliability and Validity of the Persian Version of the Quality of Life in Autism Questionnaire (QoLA) Parent Version in Iranian Mothers

Background: Parents of children with autism spectrum disorder (ASD) are known to poorer quality of life. The Quality of Life in Autism Questionnaire (QoLA) is a commonly used instrument for measuring the quality of life in parents of children with ASD. The aim of this study was to evaluate the reliability and validity of the QoLA in Iranian mothers of children with ASD. Methods: The sample of this methodological study consisted of 88 mothers of children with ASD in Arak, Iran. The data were collected using convenience sampling method between September 2019 and January 2020. A battery of questionnaires was administered to mothers which included the QoLA, Patient Health Questionnaire-9 (PHQ-9), Generalized Anxiety Disorder-7 (GAD-7), and Perceived Stress Scale-4 (PSS-4). Factor structure and internal consistency of the QoLA were examined via conrmatory factor analysis (CFA) and Cronbach’s alpha, respectively. Convergent validity was examined by relationship with WHOQOL-BREF, PHQ-9, GAD-7 and PSS-4. Results: The mean total scores of QoLA Part A and Part B were 86.50 (SD=13.89) and 61.41 (SD=18.21), respectively. Both subscales exhibited good internal validity (with Cronbach’s alpha of 0.899 and 0.950 for Part A and Part B, respectively). The convergent validity of both subscales of QoLA was proved via moderate to strong correlations with measure of the WHOQOL-BREF. In addition, both QoLA Part A and Part B scores were negatively correlated with measures of PHQ-9, GAD-7, and PSS-4. The conrmatory factor analyses provided evidence for unidimensionality of both subscales of QoLA. Conclusion: The Persian version of QoLA displays satisfactory reliability and validity in Iranian mothers of children with ASD.

de cit hyperactivity disorder and medical conditions such as cystic brosis [2,6]. These levels of stress and depression in families of children with ASD can have a negative impact on their quality of life (QoL) and it is important to assess QoL as part of the comprehensive assessment of families of children with ASD [7] The term "QoL" rst appeared in around 1920, but it was not until the middle of the century that the concept became a part of universal discourse [2]. QoL is a multidimensional construct and involve health and non-health-related domains of functioning. Different models of QoL are available with differing foci, including some that concentrate on health-related aspects of QoL and comprising eight main areas including personal development, self-determination, interpersonal relations, social inclusion, rights, emotional wellbeing, physical wellbeing, and material wellbeing [2]. Therefore, QoL is a useful indicator of adaptation of parents of children with ASD and also an important part of their experiences [2]. Recent studies have used measurements of general quality of life to assess the QoL of parents of children with ASD, including the World Health Organization Quality Of Life (WHOQOL) (2), Family Quality of Life (FQOL) [4] Short-Form Health Surveys [8,9]; and EuroQol Five Dimensional Questionnaire (EQ-5D-3L) [10]. However, in order to provide appropriate support services for parents of ASD children, it is essential to understand the unique impacts of ASD related symptoms on parental QoL using an ASD speci c measure such as the Quality of Life in Autism (QoLA) [2]. Most evaluation scales however are designed in developed countries and since there is no suitable tool in Iran, this study was designed to investigate the following two objectives: translating the parental version of QoLA into Persian and evaluating its psychometric properties.

Participants and study design
The sample of this cross-sectional study consisted of mothers of children with ASD who were resident in Arak, Iran. The data were collected using convenience sampling method between September 2019 and January 2020. The eligibility criteria for the present study included (1) mother having a child with ASD; (2) age over 18 years; (3) ability to read and write in Persian; and (4) willingness to participate in the study.
Mothers were asked to complete the questionnaire following informed consent and a total of 88 mothers agreed to take part and returned the completed questionnaire.

Translation of the QoLA into Persian
The standard "forward-backward" translation method was used to translate the QoLA from English into Persian and back into English.

Questionnaires
Demographic variables Participants provided demographic information including mother's age, child's age, child's sex, mother's education, mother's occupation, and marital duration.

Quality of Life in Autism Questionnaire (QoLA)
The QoLA is a is a self-administered instrument that measures the QoL of parents of children with ASD [2]. The QoLA is composed of two modules: the QoLA Part A and the QoLA Part B. The QoLA Part A consists of 28 items that measures parents' overall perception of their QoL. The items are rated on 5point Likert scale ranging from 1 (not very much) to 5 (very much). The total QoLA Part A scores can range from 28 to 140; higher scores indicate better QoL. The QoLA Part B consists of 20 items that measures parents' perception of how problematic their child's autism-speci c di culties are for them. In this part, parent rate items on a 5-point Likert scale, ranging from 5 (not much of a problem for me) to 1 (very much of a problem for me). The total QoLA Part B scores can range from 20 to 100; higher scores indicate fewer problems for parents regarding their child's ASD-related behaviors.

The World Health Organization Quality of Life-brief version (WHOQOL-BREF)
The WHOQOL-BREF is an abbreviated version of the WHOQOL-100 which contains 26 of the original 100 items [11]. This scale yields four domains of QOL (i.e., Physical Health, Psychological, Social Relationship and Environment). Respondents rate items on a 5-point Likert scale, ranging from 1 to 5. Raw domain scores were scaled to range from 0 to 20, with higher score representing better QoL. The Persian version of the WHOQOL-BREF has been proven to be a reliable and valid instrument to assess QoL [12]. In this study, Cronbach alpha of the WHOQOL-BREF subscales were ranging from 0.627 to 0.805. Patient Health Questionnaire-9 (PHQ-9) The PHQ-9 is a 9-item self-report instrument designed to measure depression based on the DSM-IV criteria for major depressive episode [13]. Respondents rate items on a 4-point Likert scale, ranging from 0 (not at all) to 3 (nearly every day). Total scores can range from 0 to 27; higher scores indicate worse depressive symptoms. The Persian version of the PHQ-9 has been proven to be a reliable instrument to assess depression in different population [14,15]. In the present study, the Cronbach's alpha was 0.862.

Generalized Anxiety Disorder-7 (GAD-7)
The GAD-7 is a 7-item self-report instrument designed to measure anxiety based on the DSM-IV criteria [16]. Respondents rate items on a 4-point Likert scale, ranging from 0 (not at all) to 3 (nearly every day). Total scores can range from 0 to 21; higher scores indicate more anxiety symptoms. The Persian version of the GAD-7 has been proven to be a reliable instrument to assess anxiety [17]. In the present study, the Cronbach's alpha was 0.877.

Perceived Stress Scale-4 item (PSS-4)
The PSS-4 is an ultra-brief version of the PSS-14 which contains 4 of the original 14 items. This scale measures the "degree to which situations in one's life over the last month are appraised as unpredictable, uncontrollable, and overloaded". Respondents rate items on a 5-point Likert-type scale ranging from 0 (never) to 4 (very often). The total score can range from 0 to 16, higher scores indicate more stress. The Persian version of the PSS has been proven to be a reliable instrument to assess stress in different population [18,19]. In the present study, the Cronbach's alpha was 0.565.

Statistical analysis
The con rmatory factor analysis (CFA), with maximum likelihood estimation method, was carried out in order to investigate the factor structure of the QoLA Part A and Part B. Overall model t was examined using multiple t criteria, as suggested in the literature. Speci cally, four goodness-of-t indices were used, including chi-square/degree of freedom (χ 2 /df), comparative t index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). Values of χ 2 /df < 3, CFI > 0.90, and RMSEA and SRMR < 0.08 are indicative of a good t with the data [20][21][22][23]. In order to evaluate the internal consistency of the QoLA, Cronbach's alpha and the corrected-item total correlation were calculated. The convergent validity of the scale was assessed by examining the relationship between QoLA scores and scores on the WHOQOL-BREF domains. Statistical analyses were undertaken using SPSS for windows, version 16.0 (SPSS Inc., Chicago, IL, USA) and LISREL 8.80 (Scienti c Software International, Inc., Lincolnwood, IL, USA).

Participant characteristics
The demographic characteristics of the mothers and their children are summarized in Table 1. The mean age and marital duration of the mothers were 35.00 (SD = 5.34) and 12.89 (SD = 5.10) years, respectively. Of the mothers, 79.5% were housewives and 44.3% had university education. The mean age of their children was 6.44 (SD = 2.81) years, and 69.3% of children were male. Descriptive statistics and internal consistency of the QoLA The descriptive statistics and internal consistency of the QoLA-Part A and QoLA-Part B are displayed in Table 2. The mean total score for QoLA Part A and QoLA Part B were 86.50 (SD = 13.89) and 61.41 (SD = 18.21), respectively. For QoLA Part A, item 22 ("I feel guilty") achieved the highest mean scores of 3.76 and items 23 ("I am part of a community") and 24 ("I can get the support that I need from the community") had the lowest mean scores of 1.83 and 1.89, respectively (Table 2). For QoLA Part B, item 18 ("Engaging in reckless or tactless behaviours") recorded the highest mean score (3.53), followed by item 7 ("Saying things that are socially embarrassing"; mean, 3.48). The lowest score was for item 4 ("Holding a conversation"; mean, 2.72) ( Table 3).

Convergent validity
As presented in Table 3, QoLA Part A scores were positively correlated with the WHOQOL-BREF domains.
The magnitude of all correlations were strong except for the Social Relationships domain. QoLA Part B scores were positively and moderately correlated with the Physical Health and Psychological domains. There were also positive but non-signi cant correlation between QoLA Part B and Social Relationships and Environment domains.
Relationship of QoLA with measures of depression, anxiety, and stress As presented in Table 3, QoLA Part A scores were negatively correlated with measures of PHQ-9, GAD-7, and PSS-4. The magnitude of these correlations were strong. QoLA Part B scores were positively and moderately correlated with measures of PHQ-9, GAD-7, and PSS-4.

Discussion
The aim of the present study was to translate the QOLA into Persian and examine its psychometric properties in mothers of children with ASD. Improving the QoL of parents of children with ASD is one of the main goals in rehabilitation of these children. For this purpose, having a validated disease-speci c scale is essential to assess the QoL of parents of ASD children. The great advantage of using QoLA in clinical practice is that it measures more precisely the true impact of ASD-and not of other stressful life events-on parents' QoL, which cannot be achieved via generic scales. Our results demonstrate the same pattern of mean scores on both Parts of QoLA as was found in the development study of QoLA by Eapen et al. [2], although QoLA Part B scores in our Iranian population tended to be slightly lower. The internal consistency of both Parts of QoLA was found to be good as the Cronbach's alpha values were above the recommended value of 0.7 [24]. Similar ndings have been observed in previous studies validating both parts of the QoLA [2,25,26]. However, according to the corrected item-total correlations, some modi cations for items 2, 17, and 23 in QoLA Part A might be needed in the scale to yield better internal consistency. In Turkish version of QoLA Part B, corrected item-total correlations for items 2, 17, and 22 was lower than 0.2 [25]. It is likely that cross-cultural differences might explain these ndings.
The convergent validity was examined by looking at whether self-reported WHOQOL correlated with QoLA scores. In addition, this study con rms the expected negative relationship between QoLA Part A and measures of anxiety, depression, and stress. Mothers of ASD children with a high QoLA had lower levels of anxiety, depression or stress, and vice versa. This negative relationship is in line with the results of Eapen et al. study [2] and other studies assessing QoL for parents of ASD children [27,28].
Also, ndings indicated that the convergent validity was stronger between the QoLA Part A scores and Psychological and Environment domains of the WHOQOL-BREF compared to the correlations with Physical Health and Social Relationships domains of the WHOQOL-BREF.
This is the rst study, to the best of our knowledge, examining the factor structure of QoLA by CFI. The CFA results indicated that one-factor model of both parts of QoLA in mothers of ASD children had relatively adequate t to the data. However, after allowing for covariances between several items, a better t was achieved.
The ndings of this study should be considered in the light of several limitations. First, the sample size was relatively small. Second, because of practical reasons, we only asked mothers to complete the QoLA questionnaire. Third, we did not evaluate test-retest reliability in this research. Fourth, the present results are based on a cross-sectional design that prevents inferences regarding the direction of causality.

Conclusion
In summary, despite some limitations, this study has replicated the ndings of the original study. It has also indicated that the psychometric properties of the Persian version of QoLA are comparable to those of the original English version. The ndings suggest that it is a valid and reliable instrument for measuring the QoL in mothers of children with ASD but requires further psychometric studies (including test-retest reliability) in the Iranian population, especially in fathers of children with ASD.