Translation, cultural adaptation and psychometric properties of the Persian version of Occupational Balance Questionnaire in healthy adults

Abstract

Introduction:

Occupational balance is a fundamental concept that plays an influential role in enhancing health and well-being.

Method:

The Persian Occupational Balance Questionnaire-11 was administered to 157 healthy individuals in the test phase and 80 healthy individuals in the retest phase. Floor or ceiling effects, construct validity, discriminant validity, internal consistency reliability, test–retest reliability, standard error of the measurement and smallest detectable change were examined.

Results:

The Persian Occupational Balance Questionnaire-11 showed no missing responses. There were no floor or ceiling effects for the total score. The independent t-test did not show a statistically significant difference between groups (based on age category, marital status and job status) that didn’t support discriminant validity. Factor analysis demonstrated a three-factor solution for the Persian Occupational Balance Questionnaire-11 which jointly accounted for 57.62% of the total variance. There was high internal consistency for test (α = 0.813) and retest (α = 0.853). The test–retest reliability for Persian Occupational Balance Questionnaire-11 with a 1-week interval was excellent (ICC_agreement = 0.880). The standard error of the measurement and smallest detectable change were 2.02 and 5.60, respectively.

Conclusion:

Persian Occupational Balance Questionnaire-11 is a valid and reliable instrument for assessing occupational balance in healthy Persian-speaking participants.

Introduction

Occupational therapists consider occupational balance (OB) as an essential factor in enhancing health and well-being (Meyer, 1977). OB is a subjective concept and occupational therapists have developed this concept in different perspectives (Dür et al., 2014; Eakman, 2015; Håkansson et al., 2020; Matuska, 2012; Wagman and Håkansson, 2014). The initial definition of OB was having balance between work, rest, leisure and sleep developed by Adolf Meyer (Meyer, 1977). Different assessment tools for OB were conceptualized and developed (Wagman et al., 2012). Life balance inventory (LBI) (2012) originated from life balance model (LBM) developed by Matuska and Christiansen. In LBM, OB is defined as a “satisfactory pattern of daily activities that are meaningful to the individual and not mentally, physically or psychologically harmful” (Matuska and Christiansen, 2008). The LBI measures OB in the form of 53 activities, based on the LBM (Matuska, 2012). Another assessment tool is OB-Quest designed by Dür et al. (2014). They discovered that OB contains three components: balance between different types of occupations, the balance between abilities and resources and the balance between a person’s own and others’ demands (Dür et al., 2014). Meaningful Activities Wants and Needs Assessment (MAWNA) is another assessment tool developed by Eakman (2015). He believed that quality of life refers to a situation in which a person is so involved in meaningful daily activities that the person no longer feels the need for another meaningful activity. MAWNA compares the balance between actual and desired engagement in meaningful occupations (Eakman, 2015).

Wagman et al. (2012: 325) also analyzed the concept of OB. According to their findings, OB is defined as an “individual’s subjective experience of having the ‘right mix’ (i.e., amount and variation) of occupations in his/her occupational pattern.” Related to the definition, three perspectives of OB appeared: occupational areas, occupations with different characteristics and time use. According to the theoretical definition of Wagman et al., the “right mix” of occupations (OB) is a subjective concept; therefore OB provides person-centred content and is different for everyone. Wagman et al. also determined some disturbing elements which damage OB. One of these factors is dissatisfaction with the fit between a person’s occupational patterns and his or her role as a parent or child (lack of time). Another factor is the absence of the right amount (neither too little nor too much) and enough variation in occupational patterns (Wagman et al., 2012). Accordingly, Wagman and Håkansson (2014) first developed a 13-item questionnaire that covered different dimensions of OB. Afterward, they revised the questionnaire in their next study; 2 of the 13 questions were deleted and an 11-item questionnaire (OBQ11) was achieved (Håkansson et al., 2020).

Due to the impact of OB on health and individuals’ satisfaction with life, any imbalance in occupation can affect health (Backman, 2004; Dür et al., 2022; Parnell, 2022; To-Miles et al., 2022; Wilcock et al., 1997); therefore, instruments measuring OB are useful and fundamental. The OBQ11 assesses a comprehensive concept of OB and it can determine the level of individuals’ OB in a short period of time in comparison to similar assessment tools. The OBQ11 has not been translated into Persian, and its validity and reliability have not been investigated in the Persian-speaking population. Therefore, the current study aimed to translate, culturally adapt and investigate the psychometric properties of OBQ11-P. Floor or ceiling effects, construct validity, discriminant validity, internal consistency, test–retest reliability, standard error of the measurement (SEM) and smallest detectable change (SDC) were investigated.

Method

Study design

A cross-sectional design was adopted to translate and cross-culturally adapt the OBQ11 into the Persian language and to test reliability and validity in healthy adults. This study was reviewed and approved by the Research Ethics Committee of Tehran University of Medical Sciences with the code IR.TUMS.FNM.REC.1399.239.

Stage I: Translation and cultural adaptation

The process of translation and cultural adaptation was followed according to Beaton et al. (2000) guideline. The English version of the questionnaire was first forward translated by two native Persian speakers. Two translated versions were then reviewed by the expert committee (translators, three professors of occupational therapy, one professor of physiotherapy, one methodologist and one master’s student of occupational therapy), to synthesize the translations. This synthesized version was back-translated into English by another two translators who were blinded to the original version. Finally, the expert committee reviewed all the translations considering the original version and a pre-final version of the OBQ11-P was obtained. Then, the pilot study was conducted with the pre-final version of OBQ11-P administrated to 30 healthy subjects (18 men and 12 women, mean age 35 years old) via internet. The subjects were asked to complete the questionnaire and indicate their comments on the content, clarity, and meaning of the questionnaire items. Subjects commented the questionnaire was clear and understandable. Consequently, the final OBQ11-P was approved by the expert committee for psychometric testing (Figure 1).

Figure 1.

The steps for translation and cultural adaptation of the Occupational Balance Questionnaire-11 to Persian language.

Occupational Balance Questionnaire-11

The OBQ11 is a self-report assessment tool that measures OB in different areas. This questionnaire is the modified version of OBQ which was developed by Wagman and Håkansson (2014). They found that the OBQ has good internal consistency (Cronach’s α = 0.936) and sufficient test–retest reliability (Spearman’s ρ = 0.926). Håkansson et al. retested the validity of the OBQ in 2020 and revised the original OBQ named as OBQ11. Each item of the OBQ11 is scored on a four-point scale from “strongly disagree” (0 points) to “strongly agree” (3 points) with a total score between 0 and 33. A higher score indicates a better OB (Håkansson et al., 2020).

Stage II: Reliability and validity study

Participants

Individuals eligible for the study were invited to participate. Written informed consent was obtained from all participants. Inclusion criteria were: age 20 years and older, being a native Persian speaker, being healthy (no acute or chronic physical or mental disorder that disrupted a person’s life) and able to read and write Persian language and able to use the internet. The sample size of a minimum 100 subjects was determined following the guideline (Terwee et al., 2007).

Procedure

Eligible healthy subjects completed two questionnaires comprising a demographic questionnaire and a Persian version of the occupational balance questionnaire (OBQ11-P). For this study, a web-based investigation was conducted due to the pandemic of COVID-19. The questionnaires were sent to the subjects via the internet (by sharing the questionnaire with the subjects through the link of Google form). The OBQ11-P was administrated 7 days later to examine test–retest reliability. A one-week interval was determined to reduce the bias associated with a shorter time. For discriminant validity, differences on OBQ11-P total scores were analyzed between groups of elderly 60 ⩽ years versus non-elderly <59 years, employed versus unemployed and single versus married.

Data analysis

Statistical analysis was carried out using the SPSS statistical software version 22 (SPSS Inc., Chicago, Illinois, USA). For discriminant validity, differences on OBQ11-P total scores were analyzed between groups of elderly 60 ⩽ years versus non-elderly <59 years, employed versus unemployed and single versus married using independent t-test. Construct validity was analyzed with exploratory factor analysis (EFA). Factor structure of the OBQ11-P was examined with the extraction method of principal component analysis (PCA) and varimax rotation. Kaiser–Meyer–Olkin (KMO) and the Bartlett’s test of sphericity were calculated for sampling adequacy and appropriateness of correlation matrix. Eigenvalues >1, the scree plot and the variance >10% were used to determine the number of factors. The correlation coefficients of ⩾0.40 were set to include items in a factor (Costello and Osborne, 2005). Factor structure helps to define the extent to which the structure of the instrument is replicable in different populations (Meyers et al., 2016). Internal consistency reliability was measured using Cronbach’s α for the total score. The Cronbach’s α coefficient of 0.7 was defined acceptable. Item-total correlation (ITC) was used to analyze the internal structure of the OBQ11-P and defined at ⩾0.2 as acceptable (Nakhostin Ansari et al., 2022). For test–retest reliability, ICC_agreement (two-way random effects model, single measure) was used and interpreted as follows: >0.75 excellent, 0.75–0.60 good, 0.59–0.40 fair (Fleiss et al., 2013). Floor and ceiling effects were defined as >15% of patients with minimum and maximal total scores, respectively (Terwee et al., 2007). The absolute measures of reliability of SEM (σ√(1 − ICC) (Bruton et al., 2000) and the SDC (1.96 × √2 × SEM) were calculated (Terwee et al., 2007). An α of <0.05 was considered statistically significant.

Results

Subjects

In total, 157 individuals (82 female, 75 male) with a mean age of 42 ± 18 years (range 20–70) participated in this study. The characteristics of subjects are illustrated in Table 1.

Table 1.

Demographic characteristics of the participants (N = 157).

Variables	N (%)	Percent
Sex
Male	75 (48)	48
Female	82 (52)	52
Marital status
Single	77 (49)	49
Married	80 (51)	51
Education
Primary school	10 (6)	6
High school	10 (6)	6
Diploma	28 (18)	18
Master diploma	17 (11)	11
Bachelor	57 (36)	36
Master bachelor	23 (15)	15
PhD and higher	12 (8)	8
Job
Employed	85 (53)	53
Unemployed	26 (17)	17
Student	23 (15)	15
Retired	23 (15)	15

Translation and cultural adaptation

We encountered no problems during the translation process of OBQ11 into Persian language. Participants reported the questions and response items were easy to understand. Participants responded to all items of OBQ11-P. There were no missing data.

Floor or ceiling effects

Floor or ceiling effect was not found for OBQ11-P total score in test (17.71 ± 4.83, range = 5–30) and retest (17.63 ± 4.91, range = 4–28). No individuals were scored the lowest “0” or highest “33” possible total score on OBQ11-P.

Factor analysis

The KMO coefficient of sampling adequacy was 0.838 and Bartlett’s test of sphericity was significant (χ² = 461.79, df = 55, p < 0.0001) confirming that the sample size and the correlation matrix were suitable for factor analysis. A PCA with varimax rotation loaded three latent factors with eigenvalues greater than 1, which jointly accounted for 57.62% of the total variance (Table 2). However, only the first two factors had the criteria of eigenvalues >1 and variance >10% which explained 9.62% of the total variance (Table 2, Figure 2).

Table 2.

The factor structure of the OBQ11-P.

OBQ11-P items	Factors
	1	2	3
	(OB and time)	(OB and rest)	(OB and amount)
1. Having sufficient things to do during a regular week			0.853
2. Balance between doing things for others/for oneself	0.533
3. Time for doing things wanted	0.775
4. Balance between work, home, family, leisure, rest and sleep	0.687
5. Have sufficient time for doing obligatory occupations	0.655
6. Balance between physical, social, mental and restful occupations		0.530
7. Satisfaction with how time is spent in everyday life	0.507
8. Satisfaction with the number of activities during a regular week	0.505
9. Balance between obligatory/voluntary occupations		0.569
10. Balance between energy-giving/energy-taking activities		0.818
11. Satisfaction with time spent in rest, recovery and sleep		0.816

OB: occupational balance; OBQ11-P: Persian Occupational Balance Questionnaire-11.

Figure 2.

Scree plot of Persian Occupation Balance Questionnaire-11.

Discriminant validity

Independent t-test showed that there were no significant differences between the groups on OBQ11-P total scores (Table 3).

Table 3.

Results of independent t-test for discriminant validity of OBQ11-P.

OBQ11-P total score
	Mean ± SD	Min–max	p Value
Age
Elderly (n = 50)	18.36 ± 5.38	9–30	0.28
Non-elderly (n = 107)	17.41 ± 4.55	5–29
Job
Employed (n = 85)	18.03 ± 4.43	6–29	0.06
Unemployed (n = 26)	16.15 ± 4.85	5–24
Marital status
Single (n = 77)	17 ± 5.37	5–29	0.07
Married (n = 80)	18.4 ± 4.16	10–30

OBQ11-P: Persian Occupational Balance Questionnaire-11; SD: standard deviation.

Internal consistency reliability

Internal consistency was high for test (Cronbach’s α = 0.813, Cronbach’s α, if an item deleted 0.78–0.83) and retest (Cronbach’s α = 0.853, and Cronbach’s α if an item deleted 0.82–0.86). Item 1 had the lowest value of ITC for both test (0.10) and retest (0.12) (Table 4).

Table 4.

Cronbach’s α if an item deleted and corrected ITC for OBQ11-P.

Items	Scale mean if item deleted		Scale variance if item deleted		Corrected ITC		Squared multiple correlation		Cronbach’s α if item deleted
	Test	Retest	Test	Retest	Test	Retest	Test	Retest	Test	Retest
1	15.43	15.46	22.08	23.08	0.10	0.12	0.11	0.19	0.83	0.86
2	16.10	16.05	19.95	21.79	0.42	0.37	0.23	0.24	0.80	0.85
3	15.92	15.90	19.32	19.76	0.50	0.62	0.33	0.49	0.79	0.83
4	16.06	16.11	18.68	19.34	0.54	0.63	0.38	0.43	0.79	0.83
5	15.99	16.00	19.92	20.35	0.41	0.49	0.30	0.30	0.80	0.84
6	16.37	16.28	18.95	19.72	0.60	0.68	0.43	0.56	0.78	0.82
7	16.20	16.24	19.24	19.75	0.58	0.67	0.40	0.53	0.78	0.83
8	16.22	15.83	19.27	18.72	0.60	0.62	0.46	0.43	0.78	0.83
9	16.31	16.20	19.48	19.47	0.59	0.69	0.46	0.58	0.78	0.82
10	16.38	16.29	20.16	20.86	0.45	0.54	0.31	0.39	0.80	0.84
11	16.15	16.03	19.48	20.20	0.45	0.47	0.31	0.30	0.80	0.84

ITC: item-total correlation; OBQ11-P: Persian Occupational Balance Questionnaire-11.

Test–retest reliability

Eighty subjects (mean age 34.39 ± 13.2 years) participated in the test–retest reliability study. The ICC_agreement for test–retest reliability of the OBQ11-P total score was excellent (0.88, 95% confidence interval: 0.819–0.921, p < 0.0001).

SEM and SDC

The absolute reliability values for OBQ11-P were SEM = 2.02, SDC = 5.60.

Discussion

We followed a standard method to translate and culturally adapt the self-reported OBQ11 into Persian language. The OBQ11-P will allow Iranian clinicians and occupational therapists to assess the related problems and to adopt an appropriate strategy to help clients to restore it. As far as we know, this is the first tool in Persian language for measuring various aspects of the OB. Given that occupational therapists are involved in enabling clients to engage successfully in occupations, this study provided an important tool to assess the clients with the OB problems in clinical practice. The results of the present study, consistent with translated versions of Turkish (Günal et al., 2020), Arabic (Dhas et al., 2022) and Norwegian (Uhrmann et al., 2019), demonstrated that the Persian version of the OBQ11 was equivalent to the original version (Håkansson et al., 2020), easily understood by subjects, and reliable and valid.

Translation and cultural adaptation

There were no problems observed in the process of translation and adaption. The expert panel made no significant changes in the translations, and subjects commented the questionnaire items were clear and easy to understand as reflected in the responses of participants to all items of the questionnaire. This finding, consistent with those reported for the original version (Håkansson et al., 2020) and adapted Turkish (Günal et al., 2020), Arabic (Dhas et al., 2022) and Norwegian (Uhrmann et al., 2019) versions of OBQ11 indicates the feasibility and acceptability of OBQ11-P.

Floor or ceiling effects

In this study, a floor or ceiling effect was not found for the OBQ11-P. This finding is in line with the original and adapted versions of OBQ11 (Dhas et al., 2022; Günal et al., 2020; Yu et al., 2018). The absence of floor or ceiling effect indicates the responsiveness and content validity of the OBQ11-P. This indicates that the OBQ11-P can detect changes in improvements or deteriorations in OB of individuals over time. The absence of floor or ceiling effect along with responding to all questions supports the reliability, face validity and content validity of the OBQ11-P. We did not evaluate the responsiveness using the effect size approach in a clinical situation. Therefore, a further study in a clinical setting is also required to determine the responsiveness of OBQ11-P.

Internal consistency reliability

The internal consistency reliability of OBQ11-P was excellent for both test (Cronbach’s α = 0.813) and retest (Cronbach’s α = 0.853). When each item was omitted, Cronbach’s α did not change significantly, range 0.82–0.86, and all α values are around the overall α (0.813) confirming that all items of OBQ11-P are related and homogenous. This indicates that none of the OBQ11-P items would affect the internal consistency reliability if they were omitted. The internal consistency reliability of OBQ11-P is in agreement with those of the Turkish (Cronbach’s α = 0.785) (Günal et al., 2020) and the Arabic versions (Cronbach’s α = 0.864) (Dhas et al., 2022). The Norwegian version of OBQ11 is not evaluated for internal consistency reliability (Uhrmann et al., 2019). However, the ITC value for item 1 was poor that indicates item 1 represents construct other than the overall OBQ11-P and thus it may be redundant in OBQ11-P. Further investigations to confirm this finding are required to clarify whether or not item 1 is redundant. Therefore, while the Cronbach’s α values on OBQ11-P indicates that the item scores were related confirming the homogeneity of the OBQ11-P, poor ITC found for item 1 questions the internal structure of the OBQ11-P as an overall one-factor questionnaire. The ITC was not reported for the original version and adapted versions of the OBQ11 in other languages (Håkansson et al., 2020; Yu et al., 2018).

Test–retest reliability

The test–retest reliability was excellent for OBQ11-P in agreement with those found in English (ICC = 0.74) (Yu et al., 2018), Turkish (ICC = 0.922) (Günal et al., 2020) and Swedish (ICC = 0.926) (Håkansson et al., 2020). The difference between repeated measurements was small with no significant differences between the test–retest total scores. Excellent test–retest reliability found in the present study (ICC = 0.880) indicates that the OBQ11-P scores were stable from test to retest supporting the reliability of the OBQ11-P for repeated assessments.

SEM and SDC

In this study, both the SEM and SDC were calculated as the measures of the absolute reliability. The SEM is the standard deviation of questionnaire errors. The SDC is based on SEM and estimates the real smallest change in score measured for a patient that is greater than the measurement error (Terwee et al., 2007). The SDC estimated for OBQ11-P was 5.60; this indicates that a score on the OBQ11-P must be greater than 5.60 to be interpreted real and clinically important. Otherwise, any changes less than 5.60 after an intervention would be considered due to measurement error. The SEM and SDC are not reported for the original as well as the adapted versions of OBQ11 (Dhas et al., 2022; Günal et al., 2020; Håkansson et al., 2020; Uhrmann et al., 2019).

Discriminant validity

In this study, the analyses did not demonstrate significant differences on OBQ11-P total scores between healthy groups particularly elderly versus non-elderly, employed versus unemployed and single versus married. This finding, in line with the original OBQ11 (Håkansson et al., 2020), indicated that the scores on OBQ11-P were similar among the healthy groups. The original OBQ11 evaluated for differences on age and gender reported invariance of scores across age and gender groups (Håkansson et al., 2020). In this study, subjects were healthy people and thus the similar scores on OBQ11-P across groups may indicate that the OB is a similarly perceived construct among healthy groups. In other words, the OBQ11-P is not affected by age, employment state and marital status in healthy individuals. Nevertheless, previous studies have shown that OB differs between genders of men and women (Håkansson and Ahlborg, 2010) and between people with and without children (Wagman and Håkansson, 2014). In this study, healthy groups were included; the OB may differ between people with pathologies or disabilities when compared with people without them. Further investigations are imperative regarding the discriminant validity of the OBQ11-P. The adapted Turkish, Arabic and Norwegian versions of the OBQ11 were not evaluated for discriminant validity (Dhas et al., 2022; Günal et al., 2020; Uhrmann et al., 2019).

Construct validity

We performed EFA to further explore the internal structure of the OBQ11-P. The structure of the OBQ11-P as a priori hypothesized was not supported by a single factor solution (Håkansson et al., 2020) but suggesting multidimensionality. The EFA produced a three-factor solution for OBQ11-P with the third one as a single item factor (item 1: having sufficient things to do during a regular week). The third factor, however, had explained less than 10% of the total variance and along with scree plot was not supported as an independent factor. Nevertheless, while the ITC showed all items of OBQ11-P well correlated with the total score, item 1 had lowest correlation (ITC = 0.1) indicating that the item 1 might be considered as a misfit with the total scale. The EFA of Turkish version revealed one factor; however, item 1 had the lowest factor loading, 0.243 (Günal et al., 2020). Further, item 1 in the Arabic version had the lowest factor loading (Dhas et al., 2022). The OBQ11 is an instrument for measuring OB based on people’s amount of occupations (Håkansson et al., 2020). Considering the amount of occupations mentioned in item 1 as an independent factor in this study may indicate that item 1 may be considered as redundant for evaluating OB. Studies are required to confirm the contribution of item 1 in the measurement of OB. Dhas et al. (2022) also stated the reason is that item 1 does not appear to contribute more in terms of measuring OB.

Study limitations

The current study has some limitations. First, this study was conducted during the COVID-19 pandemic and consequently data collection was performed via internet. The data collection was performed when people in Iran had not returned to a normal condition when restrictions and quarantine were all in place. It is possible that it might have affected subjects. Nevertheless, the current study is important as we conducted this study in accordance with standard methodology for translations and adaptation of health questionnaires. Second, the people with chronic health conditions and disabilities were not included. In this study, there were no differences among the healthy groups in line with previous reports (Yu et al., 2018). The OB, however, may differ among people with disabilities and between healthy subjects and ill subjects. Third, factor analysis was used to evaluate the construct validity of OBQ11-P; this could affect the construct validity. The relationships between the OBQ11-P scores and questionnaires reflecting OB construct must be examined for convergent construct validity.

Conclusion

This study described the process for development and metric assessment of the Persian version of OBQ11. The OBQ11-P was found easily understandable by healthy Iranian participants and showed satisfactory reliability and validity. The Persian-speaking occupational therapists can use it in clinical practice and research for measuring the OB of healthy individuals. Future studies on reliability and validity of the OBQ11-P in people with health conditions and disability are suggested.

Key findings

• OB is an essential factor in enhancing health and well-being.

• OBQ11-P is used by clinicians and researchers to evaluate OB in a short period in healthy Persian-speaking participants.

What the study has added

OB is a key concept in occupational science. The OBQ11-P covers different dimensions of OB and has validity and reliability in healthy Persian-speaking participants.