Quality of Life in Bipolar Disorder: Portuguese validation of the Brief QoL.BD questionnaire

Research Article

J Psychiatry Mental Disord. 2023; 8(1): 1064.

Quality of Life in Bipolar Disorder: Portuguese validation of the Brief QoL.BD questionnaire

Julieta Azevedo1,3 Maria Roque1; Diogo Carreiras1; Paula Castilho1 and António Macedo2,4

1Univ Coimbra, Centre for Research in Neuropsychology and Cognitive Behavioural Intervention (CINEICC)

2Univ Coimbra, Institute of Psychological Medicine (IPM)

3School of Human and Behavioural Sciences, BangorUniversity, United Kingdom

4Coimbra Institute for Biomedical Imaging andTranslational Research (CIBIT), Portugal

*Corresponding author: Julieta Azevedo Faculty of Psychology and Educational Sciences, University of Coimbra, Rua do Colégio Novo, Apartado 61533001-802 Coimbra, Portugal.

Received: December 26, 2022; Accepted: February 02, 2023; Published: February 09, 2023

Abstract

The Brief Quality of Life in Bipolar Disorder (Brief QoL.BD) questionnaire is a short version of the first disorder-specific scale developed for Bipolar Disorder (BD), the QoL.BD, designed for use in research settings. This questionnaire has been considered a useful measure to assess psychological interventions’ improvements. This study aimed to validate and assess the psychometric properties of the Brief QoL.BD.

Method: The Brief QoL.BD questionnaire was translated to Portuguese according to international guidelines, and participants with BD, psychiatrists and psychologists were consulted regarding its intelligibility. It was then administered to 110 people with BD (M = 43.81 ± 11.72, 66.4% ♀; 33.6% ♂), in addition to other self-report questionnaires, to assess satisfaction with life, anxiety, depression, external shame and positive and negative affect.

Results: The Brief QoL.BD revealed good internal consistency (a = .84). Positive correlations with satisfaction with life and positive affect supported the scale’s convergent validity. Significant negative correlations supported the divergent validity with negative affect, depression, anxiety and external shame. Confirmatory factor analysis validated the original one-factor structure showing a good fit.

Conclusion: The Portuguese translation of Brief QoL.BD questionnaire (European) proved to be a valid, and reliable quality of life measure to be used with people with bipolar disorder. BD type II displayed significantly lower levels of QoL than type I. Brief QoL.BD is short and easy to apply, being recommended for research purposes, specifically tracking psychological intervention’s impact.

Keywords: QoL.BD; Confirmatory Factor Analysis; Psychometric Study; Bipolar Disorder; Quality of Life

Introduction

Quality of life (QoL) is a difficult concept to define, incorporating different views according to the field of expertise in which it is used. A recent systematic review on quality of life concluded that this construct’s methodological and conceptual clarity in health and medicine had highlighted the need for appropriate validated measures [1]. Service Policies in Mental Health are undergoing profound changes, from an emphasis on reducing symptoms and clinical indexes to an approach that looks beyond the classical indicators to a focus on recovery, well-being and quality of life [1,2].

People with Bipolar Disorder (BD) often experience a severe impact of this condition on their global functioning and specifically in the ability to complete education, access financial independence or have healthy relationships [3]. They also present lower QoL when compared to the general population and other mental health problems, and even though QoL in BD patients can be satisfactory between episodes [4,5], it is still reported as impaired in remitted euthymic patients and during euthymic periods [6-8].

As QoL is a highly subjective concept, assessing it accurately and reliably can be challenging [9], especially in BD. In addition, this disorder is associated with a lack of insight concomitant with high mood and manic states, and thus the validity of self-reported measures has been frequently questioned [10,11]. There have been, however, consistent reports that QoL measures in BD can be reliable, showing consistent scores during mania, depression and remission phases, with overall QoL rates similar to euthymic patients and healthy controls [6]. Also, even though mania and hypomania symptomatology have a less negative impact on perceived QoL when compared to depressive symptoms, this impact is still higher when compared to the general population [12].

Given the importance of this construct, QoL is frequently used as an outcome treatment goal in psychological intervention, namely in people with BD [13,14]. QoL has been shown to be a mediator between treatment adherence and therapeutic alliance [15] and effectiveness of treatment interventions [16] in people with BD. It has been increasingly included in clinical trials and observational studies [17] and specifically in scientific BD literature [18].

The Quality of Life in Bipolar Disorder (QoL.BD) is the first and only (as far as we know) disorder-specific instrument to assess QoL in BD [19]. It has rapidly become an internationally spread and well-known tool [20], which proved to be feasible, reliable and valid, with excellent internal reliability and psychometric properties, being sensitive to clinical changes in BD [19]. Its development was initiated in 2004 by the Collaborative Research Team to study psychosocial issues in BD [19] composed of 56 items and 12 main factors: physical, sleep, cognition, mood, leisure, social, finances, household, spirituality, self-esteem, identity, independence, plus two optional ones, work and education [19]. Each item is rated on a 5-point Likert-type scale (1 = strongly disagree; 5 = strongly agree), with higher scores indicating a better perceived quality of life. A 10-year review of the QoL.BD worldwide revealed that it was adapted into 14 languages, and there is now a vast body of evidence regarding its relationship with various psychological and clinical variables [21].

The Brief QoL.BD is a reliable short version that includes the 12 main domains of the original scale, each one reduced to one item (rated on the same 5-point Likert scale), minus the work and education domains, based on high loadings on the exploratory factor analysis [19]. This version also showed moderate-to-large correlations with each of the subscales of the original version and convergent validity with quality of life, subjective well-being and satisfaction with life [19].

Currently, as far as we know, there is no specific measure to assess QoL in BD for the Portuguese population. Thus, we aimed to translate and validate the Brief QoL.BD questionnaire to this population (PT-EU) and to explore the association between QoL and other related variables (e.g., anxiety, depression, negative and positive affect, and external shame). Additionally, we aimed to analyse the differences between BD type I and II in the variables in study.

Methods

Procedures

This study is part of a broader project, which was approved by the Faculty of Psychology and Educational Sciences of the University of Coimbra Ethics Committee and received further approval from the hospitals and organisations concerned. Participants gave written informed consent and data confidentiality, and anonymity was assured, as well as clear instructions about General Data Protection Regulation (GDPR).

A sociodemographic questionnaire and a battery of self-report questionnaires was presented to participants either online (using LimeSurvey platform) or in paper format (ratio 70:30). Recruitment occurred between December 2019 and January 2021. In addition, participants had to have already a well-established diagnosis by a psychiatrist or, in turn, be assessed with a clinical interview by the responsible researcher with a semi-structure interview to confirm the diagnosis and have no other identified comorbidities (73 patients were assessed with the Clinical Interview for Bipolar Disorder).

Translation of the Brief QoL.BD

Permission to translate and validate the Brief QoL.BD from English to European Portuguese was requested from the scale’s original authors [19] and obtained via e-mail and was sent to be added to the CREST. BD research team website (www.crestbd.ca). The translation was conducted independently by two native Portuguese clinical psychologists and one psychiatrist, proficient in English, resulting in three translations. A consensus version was achieved by the research team. Later, it was back-translated by a different member of the research team (a psychiatrist), and this was compared to the original version. Slight changes were undertaken. Finally, the last version was shown to psychiatrists (n=5) and psychologists (n=3) experienced in dealing with people with BD and also to patients with this disorder who provided feedback about clarity. The questionnaire was described as clear and easy to understand.

Statistical analysis

Statistical analyses were done using the SPSS software version 22 (Statistical Package for the Social Sciences: IBM Corp.). To evaluate reliability and construct validity, missing data were handled using mean-score imputation (missings < 1%).

For each Brief QoL.BD domain, Cronbach’s alpha coefficients were calculated as the measure of internal reliability with a minimal reference value of 0.70 [22]. The construct validity was evaluated via Pearson’s correlations, and different sample sizes were used as we tried to maximise data collected even though some participants did not fill the entire battery of tests. Differences in the clinical sample were tested using independent samples t-test for continuous variables and chi-square for categorical variables.

Confirmatory Factorial Analysis (CFA) was performed using AMOS 24.0 software (Analysis of Moment Structures). To assess overall model fit, several goodness of fit measures and recommended cut-points were used [23,24]. Modification indices were applied to improve the model (i.e., error correlation). Normality, homogeneity and independence of the residue were validated through Skewness and Kurtosis values (|Sk| < 3 e |Ku| < 10, Kline, 2005), analysis of the normal probability graphic and Durbin-Watson statistic, respectively. Multicollinearity between variables was verified (VIF < 5). Outliers were found through the analysis of results graphs (box diagrams) and kept to ensure ecological validity.

Measures

Participants were assessed by a clinical semi-structured diagnostic interview and answered a battery of self-report questionnaires, with additional questions to describe sociodemographic variables.

The Clinical Interview for Bipolar Disorders [25] was administered to assess the diagnosis of BD and Related Disorders in adults based on the DSM-5 criteria.

Self-report questionnaires

The Satisfaction With Life Scale - SWLS [26,27] measures subjective well-being through five items, measured on a 7-point Likert-type scale. The original scale showed an a = .87 and the Portuguese version an a = .89. In this study, the SWLS had good reliability (a = .89, n = 41).

The Positive and Negative Affect Scale - PANAS [28] is a self-report questionnaire divided into two subscales: PANAS-PA and PANAS-NA (positive and negative affect, respectively). The reliability of the Portuguese version (aPA= .86 and aNA= .89) was identical to the original version (aPA = .88 and aNA = .87). In the current study, the PANAS showed an acceptable to excellent reliability (aPA = .91 and aNA = .74).

The Hospital Anxiety and Depression Scale - HADS [29,30], assesses emotional changes in a hospital setting, with two subscales: HADS-ANX and HADS-DEP (anxiety and depression, respectively). The Portuguese version achieved values of aA= .76 and aD= .80. HADS achieved a good reliability in this study (aA = .85 and aD = .85).

The Other As Shamer Scale 2 - OAS2 – Short version of the OAS [31,32], is an abbreviated version of the OAS and measures external shame. OAS2 internal consistency in the original study was a=.82 and .94 in the current study.

Participants

Participants were 110 Portuguese adults with a mean age of 43.81 (± 11.723), of which 66.4% were female (♂ = 33.6%), diagnosed with BD of any type (I, II, non-specified) and a mean age of onset of 23.93 (± 9.954; n = 95). Participants lived mainly in urban areas (70%) and 30% in rural areas and reported a mean of 14.36 years of schooling (± 3.987). Most participants were single (35.2%), 29.7% married, 19.7% divorced, 6.6% with a civil union, 1.1% widowed, and 7.7% did not fill in that information. Further descriptive statistics of clinical and sociodemographic features can be found in Table 1.