Validity and reliability of the WHOQOL-BREF in a pregnant population
Health and Quality of Life Outcomes volume 21, Article number: 96 (2023)
Physical end emotional changes during pregnancy may not only affect pregnant womens’ quality of life, but also how instruments assessing quality of life perform in such populations. To date, there is insufficient evidence on psychometric properties for both generic and condition-specific instruments measuring quality of life during pregnancy. The aim of this study was thus to examine the structural validity, internal consistency, and construct validity of the WHOQOL-BREF in a sample of pregnant women.
A convenience sample of 1015 pregnant women in Sweden completed the WHOQOL-BREF online. We examined the psychometric properties of the instrument using principal component analysis (PCA), confirmatory factor analysis (CFA), Cronbach’s alpha, item-domain correlations, correlations with overall QOL and general health, and multiple linear regression with items on overall QOL and general health as outcomes.
Principal Component Analysis in a random subsample (n = 502) supported a four-factor model, encompassing the domains physical, psychological, social and environmental quality of life, but with four of the items originally in the environmental domain relocated to the other domains. The proposed domain structure showed good fit in confirmatory factor analysis in the other random subsample (n = 513). The physical and psychological domains showed good internal consistency (Cronbach’s alpha = 0.885 and 0.826 respectively), while the social and environmental domains were weaker in this regard. All domains showed significant positive correlations with items on overall QOL and general health. The physical and psychological domains were the most evident predictors in the regression models.
We find the Swedish version of the WHOQOL-BREF to have good psychometric properties to be used in samples of pregnant women, and propose an alternative domain structure that might be even more useful for assessing quality of life during pregnancy. The physical and psychological domains showed good internal consistency and construct validity.
Pregnancy is generally perceived as a period of transition, affecting women both physically and emotionally. Researchers have shown a dramatically increasing interest in women’s quality of life (QOL) during this period in life .
QOL has been conceptualized and defined in various ways. While health related QOL focuses on the individual’s perception of how their functioning and well-being is affected by their health status or a specific condition, generic QOL broadens the picture to encompass the individual’s satisfaction with life in general, not solely in relation to disease-related limitations on functioning . The World Health Organization (WHO) defines QOL in this broad sense, as “individuals’ perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” . They highlight that QOL is a subjective experience of a multidimensional construct that pertains both positive and negative aspects of life. While there is no common understanding of the definite dimensions of QOL, most conceptualizations include physical, psychological and social dimensions .
The physical and emotional changes accompanying pregnancy may have an impact on womens’ QOL. Several studies have shown a generally lower level of QOL among pregnant women, when compared to non-pregnant women in the same age . Above all, there seem to be an overall trend of decreasing physical QOL over the course of pregnancy, while psychological QOL often is shown to be stable or even increase . Interestingly, Bai et al.  identified four different trajectories of QOL change throughout pregnancy: healthy (63%), consistently low (11%), small increase (13%) and large decrease (13%), suggesting individual variations from the group pattern.
With the different definitions of health related and generic QOL, instruments aimed to measure QOL also have their differences. Health related QOL can be measured either in a general way, such as in the SF-36 , but also using condition specific instruments. By including domains that are of central importance for a specific condition, these instruments are believed to be better equipped to capture small but important differences or changes in QOL in this particular group . The downside of this specificity is that their use is limited to a particular patient group and to the individual’s perception of how they are affected by the specific health condition. In contrast to the disease-based paradigm, generic QOL instruments are designed to capture QOL in its broad sense, as an individual’s satisfaction with life regardless of potential health conditions . These instruments comprise the individual’s subjective experience regarding multidimensional aspects of their life (e.g. self-esteem, body image, pain and discomfort, social relationships, and physical safety and security), in line with the pivotal definition of QOL by the WHO . This personal view enables universal QOL comparisons between healthy and clinical groups, and repeated assessments of an individual’s QOL over different life stages. Hence, to compare QOL between pregnant and non-pregnant women, or study longitudinal changes in womens’ QOL over longer periods of time, we would need to use a generic measure of QOL. However, conditions associated with pregnancy might affect the validity of generic instruments during this specific period in life. Hence, generic instruments need to be validated in pregnant populations.
In a recent systematic scoping review, Brekke et al.  identified the use of twelve generic and seven specific QOL instruments in pregnant populations. One of the generic instruments that was most used was the World Health Organization’s Quality of Life-BREF (WHOQOL-BREF). It was developed by the WHOQOL Group as an abbreviated version of the WHOQOL-100 , reflecting physical, psychological, social, and environmental dimensions of QOL.
In their review, Brekke et al.  identified six studies that included some evaluation of the WHOQOL-BREF in a pregnant population [11,12,13,14,15,16]. Most simply reported Cronbach’s alpha of either the full scale (alpha = 0.84–0.92) [12,13,14,15] or specific domains (Physical: 0.64–0.86; Psychological: 0.58–0.78; Social: 0.44–0.77; and Environmental: 0.76–0.80) [11, 14,15,16]. The correlations between the physical, psychological and social domains were presented in a paper by Brandão et al.  (r = 0.461–0.527), while Vachkova et al.  and Mortazavi et al.  presented some basic descriptive statistics for the specific domains. Beyond that, psychometric evaluations of the WHOQOL-BREF were lacking. Similar or even sparser results were found for the other instruments identified. While these basic statistics might give a clue of scale reliability, proper psychometric evaluation requires a much more thorough approach . As their overall conclusion, Brekke et al.  thus found the evidence on all psychometric properties insufficient, and strongly encouraged primary studies evaluating the psychometric properties for instruments that measure QOL during pregnancy.
In line with their suggestion, the aim of this study was to examine the psychometric properties of the WHOQOL-BREF, by evaluating its structural validity, internal consistency, and construct validity in a sample of pregnant women. If general measures are to be used to assess the QOL of pregnant women, it is imperative to understand how these instruments perform in such populations. Availability of a good, validated, generic QOL instrument would enable cross-sectional and longitudinal research, be useful as a validated patient-reported outcome measure in clinical trials, and facilitate clinical identification of women in need of support during pregnancy.
This cross-sectional study was part of a project focusing on quality of life during pregnancy. Ethical approval was obtained from the regional ethical board in Umeå, Sweden (2018-451-31).
Participants and procedure
Participants were self-recruited via an advertisement in social media (Facebook and Instagram). Interested individuals could follow a link to the online survey software Qualtrics, where the questionnaire was available from February 11 to March 11, 2019. Before starting the questionnaire, potential participants gave their digital informed consent to participate. The first items of the questionnaire confirmed that participants fulfilled the inclusion criteria, i.e. being at least 18 years of age and currently pregnant. No identifiable data was collected, and no exclusion criteria were adopted.
A total of 1016 women completed the questionnaire. One woman was excluded due to non-eligibility (low age), leaving a final sample of 1015 pregnant women.
The participating women had a mean age of 29.8 years (SD = 3.93). Most were in the second or third trimester of pregnancy. The vast majority were born in Sweden and lived together with a partner. Please see Table 1 for more participant details.
Participants responded to the Swedish version of the WHOQOL-BREF . The instrument has 26 items where responses are given on a five-point Likert scale (range 1–5). Items 3–26 are used to derive four domain scores: physical health, psychological, social relationships, and environment . Details on how the items are distributed between the domains can be found in Table 2. Items 1 and 2 are used as general measures of an individual’s perception of overall QOL (Q1) and general health (Q2). As suggested by Skevington et al.  the two general questions can also be added to a combined measure of overall quality of life and general health (Q1 + Q2). Although the WHOQOL-BREF has only been sparingly validated in pregnant populations, the instrument has been worldwide field-tested and its psychometric properties have demonstrated to be good to excellent, indicating it to be a valid instrument to be used across cultures and with a variety of population groups, in large epidemiological surveys, clinical settings and clinical trials .
All cases had complete data. The dataset was split in two random samples (n = 502 and n = 513), using the function “random sample of cases” in SPSS. Participant characteristics were computed for the complete sample as well as for the two subsamples, and the samples were compared using a one-way ANOVA (regarding the four original domains of the WHOQOL-BREF, items 1 and 2 separately, age, and gestational week) and Pearson’s Chi-Square tests (regarding parity, trimester, civil status, education, occupational status, country of birth, and site of living). No significant differences were identified between the two subsamples.
To assess the structural validity, we first explored the factor structure of items 3–26 using Principal Component Analysis (PCA) with Varimax rotation in one of the random subsamples (the PCA-sample, n = 502). The Kaiser-Meier Olkin (KMO) measure and Bartlett’s test of sphericity were inspected to verify the adequacy of the sampling and the correlation structure. The resulting model was examined with regard to its initial eigenvalues, the sum of squared loadings and variance explained by each factor after rotation, and the communalities and cross loadings of each item. As suggested by Costello and Osborne , communalities lower than 0.40, loadings lower than 0.32, and cross-loadings of 0.32 or higher were considered as problematic.
The PCA was followed by a confirmatory factor analysis (CFA) of the identified factor structure in the other random sample (CFA sample, n = 513), using diagonal weighted least squares (DWLS) as estimator . The goodness of fit was evaluated using a combination of methods; the normed Chi Square (χ2/df) where values lower than 2, or in more generous recommendations lower than 5, have been suggested acceptable , the Comparative Fit Index (CFI) where values ≥ 0.95 suggest good fit, the Root Mean Square Error of Approximation (RMSEA) where values ≤ 0.06 indicate a good fit and vales ≤ 0.08 could suggest an acceptable fit especially if the upper limit of the 90% confidence interval falls below this threshold, and finally the Standardized Root Mean Square Residual (SRMR) where 0.08 indicates a good fit [24,25,26]. We then inspected the modification indices to explore if the model fit could be improved by allowing item residuals with high covariance to correlate. Pairs of correlating residuals were added to the model one by one, beginning with the ones with highest covariance, until new additions only made minor changes in the fit indices. Residuals were only allowed to correlate if the items belonged to the same domain and correlations appeared theoretically meaningful. For comparative purposes, we also tested the fit of the original four domains, both in a basic model and in a model with correlated residuals, applying the same procedure as described above.
Internal consistency of both the original and proposed domains was assessed using Cronbach’s alpha, alpha if item was deleted, and item-domain correlations. A Cronbach’s alpha of 0.70 or higher and item-domain correlations of 0.30 or higher were considered adequate .
As a preliminary test of construct validity, we calculated Person’s correlations (one-tailed) between both original and proposed domains and overall QOL and general health (Q1, Q2 and Q1 + Q2). We also conducted multiple linear regression analyses (enter method) with overall QOL and general health as outcomes, using the complete sample (n = 1015). Separate analyses were conducted using the original and proposed domains as predictors. Domain scores were calculated by the mean of the included items multiplied by 4 . All domains were expected to show a significant positive correlation with the general items (Q1, Q2, Q1 + 2), and especially strong correlations were expected between the physical domain and the general health item (Q2), and between the psychological domain and overall QOL item (Q1). We also expected to see a unique and positive contribution of each domain in the multiple regression models. Correlations were interpreted as suggested by Dancey and Reidy: r = 0.10–0.39 indicating a weak correlation, r = 0.40–0.69 indicating a moderate correlation, and r ≥ 0.70 indicating a strong correlation .
All statistical analyses, except the CFA, were conducted in SPSS, version 27. The CFA was conducted using JASP, version 0.16.1.
In the PCA subsample, The KMO measure verified the sampling adequacy (KMO = 0.917) and a significant result of the Bartlett’s test of sphericity (p < .001) indicated adequacy of the correlation structure. The ratio of participants per item was 1:21, which is in line with the thresholds of sample size recommendations . Inspection of the eigenvalues indicated a possible fifth factor with an initial eigenvalue just above one (1.040). However, we found this factor neither theoretically nor statistically convincing, as one of its two items cross-loaded with the psychological domain and both items saturated with their original domain (psychological QOL) in a four-factor solution.
After rotation, the four-factor solution explained 54.71% of the total variance. The factor structure, including the sum of squared loadings and variance explained by each factor, is shown in Table 2. The identified factors showed an overall resemblance with the original domains of the WHOQOL-BREF , leading us to use the same domain names, that is, physical, psychological, social, and environmental QOL. However, factor loadings suggested that four of the items that originally were part of the environmental domain related more strongly to other factors in this sample (item 14 with the physical domain, items 8 and 9 with the psychological domain, and item 23 with the social domain). All items had a factor loading > 0.32 in their respective new domain. Three items (items 4, 9 and 11) showed low communalities (< 0.40) and eight items (items 7, 8, 10, 12, 19, 20, 24, and 25) had cross loadings larger than 0.32. With the intention to validate the original scale, we decided to retain all items, and rather identify and discuss their shortcomings in the current sample.
In the confirmatory factor analysis, we tested the factor structure proposed by the PCA in another sample (n = 513). Factor loadings are displayed in Table 3 and fit indices in Table 4. Fit indices of the proposed domain model indicated a good model fit, with CFI > 0.95 and SRMR < 0.08. Although the RMSEA did not reach below 0.06, the upper limit of the 90% CI was below 0.08 indicating acceptable fit. Fit indices were further improved after consulting the modification indices, and one by one allowing residual covariances to correlate. With four correlating pairs of residuals, RMSEA was below the 0.06 threshold. All factor loadings were acceptable (> 0.32) both in the basic model and in the model with correlated residuals (see Table 3).
For comparative purposes, we also tested the fit of the original domain structure. Although showing relatively good fit indices too, the original model had a poorer fit with our data than the proposed model had (see Table 4).
Cronbach’s alpha for the complete scale was 0.90 (0.91 in the PCA sample and 0.90 in the CFA sample respectively). The alpha values of the original and the proposed four domains are presented in Table 5, alongside alpha values if items were deleted and the corrected item-total correlations. The physical and psychological domains showed high internal consistency in both solutions, the social domain was below 0.70 in both, and the environmental domain reached above 0.70 in the original whilst not in the proposed domain structure. Cronbach’s alpha if items were deleted indicated some issues with items 11 and 21 in both domain structures, and also item 23 in the proposed structure. Item-domain correlations supported a general pattern of higher internal consistency in the physical and psychological domains and lower correlations in the social and environmental domains. All item-domain correlations were above 0.30. Five correlations in the original domains solution and four in the proposed domain solution were weak (< 0.40).
Pearson’s correlations, displayed in Table 6, showed that all original and proposed domains correlated positively with each other and with the items on overall QOL and general health (p < .001). The physical and psychological domains showed moderate correlations with the overall items (r = 0.55–0.68), the social domain displayed weak to moderate correlations (r = 0.38–0.47), and the environmental domain weak correlations in the proposed structure (r = 0.29–0.33) whilst moderate when using the original domain structure (r = 0.43–0.49).
When entering the domains simultaneously into multiple linear regression models (see Table 7), the physical and above all the psychological domains were strong independent predictors of overall QOL and general health. The social domain contributed significantly and positively to overall QOL (Q1) and the composite outcome (Q1 + Q2), whilst not to general health (Q2). The environmental domain showed different patterns when using the original and proposed items; in the first case showing small and only occasionally significant positive associations with all three outcomes, in the second case again showing small and marginally significant, but negative associations with all three outcomes.
In this study, we sought to evaluate the structural validity, internal consistency, and construct validity of the WHOQOL-BREF in a sample of pregnant women. At a general level, all analyses supported a four-factor model, encompassing the domains physical, psychological, social, and environmental QOL. However, the PCA suggested an alternative distribution of items between these domains, with four of the items originating in the environmental domain  relocated to the other domains. Although the CFA revealed a relatively good fit of the original domain structure, the proposed model structure was superior in all fit indices, indicating the proposed distribution of items to be more adequate in this sample of pregnant women.
In further comparisons between the original and proposed domain structures, the general pattern of findings suggested that the physical, psychological and social domains worked slightly better in the proposed structure, whilst the environmental domain performed better in its original eight-item version. The physical and psychological domains showed good internal consistency in both models, with a Cronbach’s alpha that was higher than in other studies of pregnant populations [11, 14,15,16]. However, just like the majority of centers in the original validation by Skevington et al. , the alpha values of domains with fewer items were lower. In our case, the social and the proposed four-item environmental domains only reached marginal alpha values. The social domain has systematically presented marginal levels of internal consistency [19, 29,30,31], also in pregnant samples [11, 14, 16]. Despite the low alpha values, each item had strong factor loadings and contributed to the reliability (as inspected by alpha if item deleted). However, the alpha statistics of the proposed four-item environmental domain was lower than typically shown in pregnant samples [11, 14,15,16]. Although the lower internal consistency of these domains might be partly explained by the low number of items in these domains , the relatively low item-domain correlations further questions the consistency of these domains.
The relocation of item 14 was found both relevant and statistically successful. Physical limitations during pregnancy might well restrict the individual’s opportunities to take part in leisure activities. Interestingly, Taylor et al.  also found item 14 to load on the physical domain when validating the WHOQOL-BREF in people with rheumatoid arthritis, which further supports this hypothesis. In our study, the item performed well in its new domain, with high loadings in the CFA and higher item-domain correlations when compared to its original placing in the environmental domain.
Item 8, asking about safety in daily life, was found to be strongly associated to both the environmental and psychological domains. Although cross-loading on the environmental domain, the PCA showed a clear belonging in the psychological domain which was supported by higher item-domain correlations in the proposed domain structure. As the Swedish word for “safe” used in the translation have strong connotations of psychological safety, this double loading seem reasonable. Neither is it unique to this study. In the original cross-cultural validation of the scale, this item correlated more strongly with the psychological domain in seven of the 24 countries represented , and high loadings on this domain have been found in other validation studies [29, 34].
Items 9 (physical environment) and 23 (conditions of living place) could best be described as generally problematic in this sample, with low communalities and factor loadings regardless of domain. Item 9 has no apparent connection to psychological QOL, and the loading on that domain was surprising to us. One hypothesis might be that its place in the questionnaire, following three psychological questions and item 8 on safety in daily life, might have primed the participants to follow a similar response pattern. Another could be that an unhealthy environment leads to worry, associating the item with other issues of a psychological nature. With regard to item 23, our guess would be that it its focus on the domestic parts of life might explain its association to the social domain. With indications that Cronbach’s alpha would decrease rather than increase if deleted, the item appeared to make a significant contribution to its new domain.
As hypothesized, all domains showed significant positive correlations with the items on overall QOL and general health. As expected and in line with previous research [19, 30], the physical domain was the strongest predictor of general health (Q2) and the psychological domain was the strongest predictor or overall QOL (Q1) and also the compound variable of Q1 + Q2. While the contribution of the social domain was smaller, this domain made a positive and significant prediction of the outcomes including overall QOL (Q1 and Q1 + Q2), but could not predict general health (Q2). The environmental domain was the weakest predictor of all outcomes, regardless of model. Its non-significant or week associations with the general items suggest that specific environmental issues (money, access to information, health services, and transports) play a minor role in determining the overall QOL and general health of pregnant women in a high-income context such as the Swedish, while other environmental factors may assume a psychological and social dimension in pregnant women. As suggested by Xia et al. , the environmental domain might be context rather than health related.
This study was conducted in a convenience sample of self-recruited pregnant women, which might pose a risk to the representativeness of the sample and the generalizability of results. With the sample being rather large, and representing women of different parities, trimesters, educational levels, countries of birth, and sites of living, we believe that the findings still might be generalizable to the general population of pregnant women living in Sweden. It is also worth mentioning that we collected our data before the outbreak of the COVID-19-pandemic and the additional challenges faced by pregnant women under these difficult circumstances.
The study might have benefited from having a comparative sample of non-pregnant women of the same age. Unfortunately, no such data was available to us.
In this study, the Swedish version of the WHOQOL-BREF was found to have good psychometric properties to be used in samples of pregnant women. While the original domain structure showed acceptable fit, we propose an alternative domain structure with even better fit that might be more useful for assessing QOL in pregnant samples. The physical and psychological domains showed good internal consistency and construct validity, while some uncertainty remains regarding the social and environmental domains. Future studies of QOL in pregnant populations are needed to evaluate the usefulness of this or other alternative domain structures.
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Morin M, Vayssiere C, Claris O, Irague F, Mallah S, Molinier L, et al. Evaluation of the quality of life of pregnant women from 2005 to 2015. Eur J Obstet Gynecol Reprod Biol. 2017;214:115–30.
Laudet AB. The case for considering quality of life in Addiction Research and Clinical Practice. Addict Sci Clin Pract. 2011;6:44–55.
The WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): position paper from the World Health Organization. Soc Sci Med. 1995;41:1403–9.
Fayers PM, Machin D. Quality of life: the assessment, analysis and interpretation of patient-reported outcomes. Second ed. Chichester: John Wiley & Sons Ltd; 2007.
Lagadec N, Steinecker M, Kapassi A, Magnier AM, Chastang J, Robert S, et al. Factors influencing the quality of life of pregnant women: a systematic review. BMC Pregnancy Childbirth. 2018;18:455.
Bai G, Raat H, Jaddoe VWV, Mautner E, Korfage IJ. Trajectories and predictors of women’s health-related quality of life during pregnancy: a large longitudinal cohort study. PLoS ONE. 2018;13:e0194999.
Ware JE, Gandek B, Group IP. The SF-36 Health Survey: development and use in Mental Health Research and the IQOLA Project. Int J Ment Health. 1994;23:49–73.
Mogos MF, August EM, Salinas-Miranda AA, Sultan DH, Salihu HM. A systematic review of quality of life measures in pregnant and Postpartum mothers. Appl Res Qual Life. 2013;8:219–50.
Brekke M, Berg RC, Amro A, Glavin K, Haugland T. Quality of Life instruments and their psychometric properties for use in parents during pregnancy and the postpartum period: a systematic scoping review. Health Qual Life Outcomes. 2022;20:107.
The WHOQOL Group. Development of the World Health Organization WHOQOL-BREF Quality of Life Assessment. Psychol Med. 1998;28:551–8.
Brandão T, Brites R, Hipólito J, Pires M, Nunes O. Dyadic coping, marital adjustment and quality of life in couples during pregnancy: an actor–partner approach. J Reprod Infant Psychol. 2020;38:49–59.
Chen Y-H, Huang J-P, Au H-K, Chen Y-H. High risk of depression, anxiety, and poor quality of life among experienced fathers, but not mothers: a prospective longitudinal study. J Affect Disord. 2019;242:39–47.
Daglar G, Bilgic D, Ozkan S. Determinants of quality of life among pregnant women in the city centre of the Central Anatolia region of Turkey. Niger J Clin Pract. 2020;23:416–6.
Khwepeya M, Huang H-C, Lee GT, Shu-Yu K. Link to external site this link will open in a new window. Validation of the Wijma delivery expectancy/experience questionnaire for pregnant women in Malawi: a descriptive, cross-sectional study. BMC Pregnancy Childbirth. 2020;20:1–11.
Mortazavi F, Mousavi SA, Chaman R, Khosravi A. Maternal quality of Life during the transition to Motherhood. Iran Red Crescent Med J. 2014;16:e8443.
Vachkova E, Jezek S, Mares J, Moravcova M. The evaluation of the psychometric properties of a specific quality of life questionnaire for physiological pregnancy. Health Qual Life Outcomes. 2013;11.
Boateng GO, Neilands TB, Frongillo EA, Melgar-Quiñonez HR, Young SL. Best Practices for developing and validating Scales for Health, Social, and behavioral research: a primer. Front Public Health. 2018;6:149.
World Health Organization. WHOQOL-BREF: Introduction, Administration, Scoring and Generic Version of the Assessment. Field trial version [Internet]. Geneva: World Health Organization. 1996. Available from: https://www.who.int/publications/i/item/WHOQOL-BREF
Skevington SM, Lotfy M, O’Connell KA. The World Health Organization’s WHOQOL-BREF quality of life assessment: psychometric properties and results of the international field trial. A report from the WHOQOL Group. Qual Life Res. 2004;13:299–310.
Gagnier JJ, Lai J, Mokkink LB, Terwee CB. COSMIN reporting guideline for studies on measurement properties of patient-reported outcome measures. Qual Life Res. 2021;30:2197–218.
Mokkink LB, Prinsen CA, Patrick DL, Alonso J, Bouter LM, de Vet HC et al. COSMIN Study Design checklist for Patient-reported outcome measurement instruments [Internet]. Amsterdam; 2019 p. 1–32. Available from: https://cosmin.nl/wp-content/uploads/COSMIN-study-designing-checklist_final.pdf
Costello AB, Osborne J. Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Pract Assess Res Eval. 2005;10:1–9.
DiStefano C, Morgan GB. A comparison of diagonal weighted least squares robust estimation techniques for Ordinal Data. Struct Equ Model Multidiscip J. 2014;21:425–38.
Hooper D, Couglan J, Mullen MR. Structural equation modelling: guidelines for determining model fit. 2008;6:53–60.
Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Model Multidiscip J. 1999;6:1–55.
Brown TA. Confirmatory factor analysis for applied research. New York: The Guilford Press; 2006.
Cristobal E, Flavián C, Guinalíu M. Perceived e-service quality (PeSQ): measurement validation and effects on consumer satisfaction and web site loyalty. Manag Serv Qual Int J. 2007;17:317–40.
Dancey CP, Reidy J. Statistics Wihtout Maths for psychology. Sixth edition. Harlow: Pearson Education Limited; 2014.
Kalfoss MH, Reidunsdatter RJ, Klöckner CA, Nilsen M. Validation of the WHOQOL-Bref: psychometric properties and normative data for the norwegian general population. Health Qual Life Outcomes. 2021;19:13.
Oliveira SEH, Carvalho H, Esteves F. Toward an understanding of the quality of life construct: validity and reliability of the WHOQOL-Bref in a psychiatric sample. Psychiatry Res. 2016;244:37–44.
Xia P, Li N, Hau K-T, Liu C, Lu Y. Quality of life of chinese urban community residents: a psychometric study of the mainland chinese version of the WHOQOL-BREF. BMC Med Res Methodol. 2012;12:37.
Cortina JM. What is coefficient alpha? An examination of theory and applications. J Appl Psychol. 1993;78:98–104.
Taylor WJ, Myers J, Simpson RT, McPherson KM, Weatherall M. Quality of life of people with rheumatoid arthritis as measured by the World Health Organization Quality of Life Instrument, Short Form (WHOQOL-BREF): score distributions and psychometric properties. Arthritis Care Res. 2004;51:350–7.
Najafi M, Sheikhvatan M, Montazeri A, Sheikhfatollahi M. Factor structure of the World Health Organization’s quality of life Questionnaire–BREF in patients with coronary artery disease. Int J Prev Med. 2013;4.
The authors declare that no external funds, grants, or other support were received during the preparation of this manuscript.
Open access funding provided by Mid Sweden University.
Ethics approval and consent to participate
This study was performed in line with the principles of the Declaration of Helsinki and informed consent was obtained from all individual participants included in the study. Ethical approval was granted by the regional ethical board in Umeå, Sweden (2018-451-31).
Consent for publication
This work does not include any individual person’s data in any form.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Rondung, E., Oliveira, S. & Esteves, F. Validity and reliability of the WHOQOL-BREF in a pregnant population. Health Qual Life Outcomes 21, 96 (2023). https://doi.org/10.1186/s12955-023-02166-2