STRENGTHS AND LIMITATIONS OF THIS STUDY

Participants were randomly selected from a region similar to the wider Australian population, allowing data to be extrapolated to national levels.

Data presented may not be cross-culturally representative.

There may be a potential healthy participant bias.

Introduction

Quality of life (QoL) is an interdisciplinary concept that has gained prominence over the years as an important outcome measure in healthcare. The WHO defines QoL as an individual’s ‘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’.1 Thus, QoL can be understood as a subjective evaluation encompassing multiple aspects of people’s lives.

Measures of QoL are increasingly being utilised in both clinical practice and research. In the clinical setting, QoL is the primary focus in the care of people with chronic and disabling conditions for which there is no cure. Similarly, interventions aimed at these populations frequently use QoL measures as an indicator of success.2 In other populations, QoL measures are also commonly used to supplement objective measures of disease and health status.3 The inclusion of QoL measures in addition to traditional physiological markers of health is reflective of the paradigm shift from a mechanistic model (concerned only with eradication of disease) towards a more holistic and patient-orientated approach to healthcare.

Although health status is related to QoL, it is not an equivalent concept.4 While healthy groups generally report better QoL compared with those with a poorer health status, disease does not necessarily equate to a lower QoL. This is sometimes referred to as a ‘disability paradox’ whereby individuals experiencing significant physical impairment and functional limitations report their QoL as good.5 Because QoL is broader than health status, contextual factors are likely to have a moderating effect. Several sociodemographic characteristics have been reported to influence QoL, including age,6–8 sex,7 9 various indices of socioeconomic status (SES)7 10–15 and marital status.7 9 16 To date, QoL research has predominantly focused on disease and treatment with few studies examining factors that promote QoL.17

Several tools have been developed to assess QoL. These can be categorised as condition-specific or generic. While condition-specific instruments are sensitive to the specific aspects of QoL related to certain diseases they are not widely applicable. On the other hand, generic QoL measures can be used in both healthy and ill populations and allow for comparison between and across these groups. A systematic review published in 2019 found the most commonly used generic instruments were the Short Form-36 (SF-36), SF-12, EQ-5D and the WHO Quality of Life brief version-BREF (WHOQOL-BREF).18 While most generic QoL tools capture aspects of QoL relating to physical and mental well-being, the WHOQOL-BREF is unique in its assessment of broader aspects of QoL, such as social relationships and environment.

The WHOQOL-BREF has been shown to have good psychometric properties and provides a valid and reliable alternative to the WHOQOL-100 which may be too lengthy for practical use.19 While the WHOQOL-BREF has been validated for use in numerous patient populations, reference ranges for the general population are limited. Reference ranges for the WHOQOL-BREF have been published for general populations in Brazil, Denmark, Indonesia, Norway and Portugal.20–24 However, due to the contextual nature of QoL, these results are not applicable to other countries. To date, only one study has published population norms using Australian data.25 Hawthorne et al reported preliminary estimates based on a combined sample of 931 adults. The pooling of two separate samples to obtain a sufficient sample size is recognised by the authors as a key limitation. Although both samples were randomly recruited using similar techniques, considerable differences in the age profile of each sample may have biased the findings. Furthermore, the number of participants aged 20–59 years was small. Hence, further research within the Australian setting is warranted.

In the absence of population-specific normative data, the usefulness of WHOQOL assessments in Australia is currently limited. Thus, the primary aim of the current study was to derive population norms for the WHOQOL-BREF based on a representative sample of Australian adults. A secondary aim was to explore sociodemographic factors potentially related to QoL.

MethodsStudy sample

This cross-sectional study used data from the Geelong Osteoporosis Study (GOS), an ongoing population-based study situated in south-eastern Australia.26 The GOS was established in 1993 to investigate the epidemiology of osteoporosis however has since expanded to investigate determinants of healthy ageing more broadly. Age-stratified samples of adults aged 20 years and older were drawn at random from the Commonwealth electoral rolls for the Barwon Statistical Division, an area comprising regional and urban communities. As voting is compulsory in Australia for adults aged ≥18 years, the electoral roll provides a comprehensive listing of residents within the region. Persons who were resident for less than 6 months and those unable to provide written informed consent were excluded from the study. The initial sample comprised 1540 men (participation 67%) and 1494 women (participation 77%) and is similar in age distribution, country of birth, education, marital status, employment and income profiles to the wider Australian population, allowing data to be extrapolated to national levels.

Baseline data for women were collected from years 1993–1997 and for men from years 2001–2006. Participants were recalled for assessment every 2–5 years. The WHOQOL-BREF instrument was introduced at the 15-year follow-up assessment for women and the 5-year follow-up assessment for men. For these follow-up assessment phases, 81% of eligible men participated and 75% of women. Data for men were collected from 2006 to 2011 and data for women from 2011 to 2014. From a potential pool of 978 men and 849 women, 68 participants did not provide QoL data or had more than 20% missing data and were therefore excluded from the analysis. Thus, the final sample for the current study included 1759 adults aged 24–94 years.

MeasuresDemographics

Area-based SES was ascertained by cross-referencing residential addresses with the Australian Bureau of Statistics 2006 census data for men, and 2011 census data for women. We categorised participants into three groups (low SES, mid SES and upper SES) based on the Index of Relative Socioeconomic Disadvantage and Advantage (IRSAD).27 IRSAD quintiles 1 and 2 were pooled to create the low SES group and quintiles 4 and 5 were pooled to create the upper SES group. Quintile 3 was categorised as mid SES. Participants completed self-report questionnaires on sociodemographic variables. Highest educational attainment was grouped into four categories: never completed secondary school, completed secondary school, completion of a Technical and Further Education (TAFE) or Trade qualification and completion of a university degree. Marital status was grouped into three categories: single and never married, in a relationship or married and separated or widowed. Employment status was grouped into three categories: not full-time employed, including those working casually, part-time or unemployed; full-time employed; and those who were retired or on home duties. For men, we classified full-time employment as those who reported working ≥38 hours per week. For women, full-time employment was self-reported.

WHOQOL-BREF

We used the Australian version of the WHOQOL-BREF; a 26-item abbreviated version of the WHOQOL-100 assessment. The WHOQOL-BREF contains one item from each of the 24 facets contained in the WHOQOL-100 and two global items pertaining to overall QoL and general health. The WHOQOL-BREF assesses QoL across four domains. The physical health domain comprises seven items relating to activities of daily living, dependence on medication, energy level, mobility, pain, sleep and capacity to work. The psychological domain consists of six items pertaining to body image, negative feelings, positive feelings, self-esteem, spirituality and cognition. The social relationships domain has three items relating to personal relationships, social support and sex life. The environment domain consists of eight items relating to financial resources, safety, access to health and social services, home environment, opportunity to acquire new skills and knowledge, recreation, transport and physical environment (ie, pollution, noise, traffic and climate). Items are rated on a 5-point Likert scale with the mean score of items within each domain used to calculate the domain score. Domain scores are scaled in a positive direction (ie, higher scores indicate a higher QoL). The WHOQOL-BREF correlates highly with the WHOQOL-100 domain scores and demonstrates good discriminant validity, internal consistency and test–retest reliability.1 19 Participants were mailed the WHOQOL-BREF questionnaire; however, where this was not possible due to language or disability, a trained research assistant administered the questionnaire in-person.

The WHOQOL-BREF data were cleaned and domain scores computed in accordance with the WHOQOL manual. In the current study, Cronbach’s alpha for the physical health, psychological, social relationships and environment domains was 0.83, 0.83, 0.71 and 0.81, respectively.

Patient and public involvement

Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

Statistical analysis

To derive normative data, we calculated means, SD and 95% CI for each QoL domain. Percentile scores were also generated. To examine associations between sociodemographic characteristics and QoL domains, we dichotomised QoL as high or low using the median cut-point for each sex. To describe differences in characteristics between participants with high and low QoL, we used independent sample t-tests for continuous data and the χ2 test for categorical data. The Mann-Whitney U test was used where continuous variables deviated from the normal distribution. To determine associations between sociodemographic variables and QoL in each of the four domains, we used logistic regression (ORs with 95% CIs, adjusted for age) in which domain scores were measured as dependent variables and sociodemographic variables were included as independent variables. Backwards stepwise regression techniques were used to determine the best model, with interaction terms checked for effect modification. Multicollinearity was assessed using variance inflation factor. The statistical software Stata V.17 (StataCorp) was used for analysis of the data.

Results

The final study sample (n=1759) comprised 929 men and 830 women (53% men). The age of participants ranged from 24 years to 94 years, median age 57.0 years (IQR=44.0–70.0). Overall mean scores for WHOQOL-BREF domains were 74.52 (SD=16.22) for physical health, 72.07 (SD=15.35) for psychological, 72.87 (SD=18.78) for social relationships and 79.68 (SD=12.55) for environment (online supplemental table 1). Percentile scores for WHOQOL-BREF domains were generated by age and sex (online supplemental table 2). Characteristics for high and low QoL groups are presented in online supplemental table 3. We report differences regarding age, SES, education level, employment status and marital status. Several variables were found to be significantly associated with a high QoL in the age-adjusted models.

Physical health

Increasing age was associated with a lower likelihood for high QoL in both men (OR 0.97, 95% CI 0.96 to 0.98, p<0.001) and women (OR 0.98, 95% CI 0.97 to 0.99, p=0.005). Compared with low SES, mid (OR 1.57, 95% CI 1.04 to 2.36, p=0.032) and upper SES (OR 1.41, 95% CI 1.01 to 1.97. p=0.044) was associated with high QoL in men; whereas only upper SES was associated with high QoL in women (OR 1.58 95% CI 1.07 to 2.35, p=0.021). Compared with men who had not completed secondary school, men with a university education (OR 1.64, 95% CI 1.14 to 2.36, p=0.008) had increased likelihood of high QoL. Women who had completed secondary school (OR 1.79, 95% CI 1.14 to 2.81, p=0.011), a TAFE/Trade qualification (OR 1.62, 95% CI 1.07 to 2.46, p=0.022) or university had an increased likelihood for high QoL (OR 1.83, 95% CI 1.14 to 2.94, p=0.012) compared with women who had not completed secondary school. Compared with women not working full time, women who were employed full time had an increased likelihood for high QoL (OR 1.93, 95% CI 1.26 to 2.97, p=0.003). Women who were married or in a relationship had an increased likelihood for high QoL compared with those who were single (OR 2.46, 95% CI 1.36 to 4.44, p=0.003).

Psychological

Men with mid SES had increased likelihood of high QoL compared with men from low SES background (OR 2.09, 95% CI 1.42 to 3.06, p<0.001). Compared with those who had not completed secondary school, university education was associated with high QoL for both men (OR 1.42, 95% CI 1.00 to 2.01, p=0.049) and women (OR 2.56, 95% CI 1.64 to 4.00, p<0.001). Compared with those who were single, being married or in a relationship was associated with increased odds for high QoL in both men (OR 2.07, 95% CI 1.20 to 3.55, p=0.009) and women (OR 2.15, 95% CI 1.21 to 3.81, p=0.009). There were no significant associations with age or employment status.

Social relationships

Compared with those who were single, being married or in a relationship was associated with increased odds for high QoL in both men (OR 2.28, 95% CI 1.29 to 4.04, p=0.005) and women (OR 2.77, 95% CI 1.42 to 5.41, p=0.003). There were no significant associations between social relationships domain of QoL and age, SES, education and employment status.

Environment

Increasing age was associated with a higher likelihood for high QoL in men (OR 1.02, 95% CI 1.01 to 1.03, p<0.001). Compared with low SES, mid SES and upper SES were associated with an increased odd for high QoL in both men (OR 1.91, 95% CI 1.30 to 2.82, p=0.001; OR 1.48, 95% CI 1.07 to 2.03, p=0.017) and women (OR 1.91, 95% CI 1.29 to 2.83, p=0.001; OR 2.02, 95% CI 1.35 to 3.01, p=0.001) respectively. Compared with those who had not completed secondary school, men (OR 1.96, 95% CI 1.37 to 2.79, p<0.001) and women (OR 1.78, 95% CI 1.13 to 2.82, p=0.013) with a university education had an increased likelihood for high QoL. Compared with women who were single, women who were married or in a relationship also had an increased likelihood for high QoL (OR 2.07, 95% CI:1.13 to 3.80, p=0.019). There were no significant associations with employment status.

Discussion

This study provides scores of QoL measured by the WHOQOL-BREF based on a population-based sample from south-eastern Australia. These data can be used as reference for comparison of groups or individuals. In our sample, average QoL scores for the social relationships and environment domains were similar among men and women. However, men reported higher averages for the physical health (75.93 vs 72.94) and psychological (74.92 vs 68.88) domains compared with women. This aligns with previous research that suggests men experience better overall QoL than women.9 Furthermore, women are known to experience higher rates of mood and anxiety disorders which may explain lower ratings for psychological-related QoL.28

Point estimates for mean scores were slightly higher compared with preliminary norms previously reported by Hawthorne et al25; physical health (74.5 vs 73.5), psychological (72.1 vs 70.6), social relationships (72.9 vs 71.5) and environment (79.7 vs 74.8). This could be due to differences in the age profile of each study or change in QoL over time. In comparison to international normative data, we report similar overall QoL estimates to those obtained from Danish and Norwegian populations.21 23 Compared with low-income and middle-income countries such as Indonesia and Brazil, our sample of Australian adults generally rated their QoL higher across WHOQOL-BREF domains.20 22

We found associations between SES and the physical health, psychological and environment domains of QoL. Compared with low SES, men from mid SES were more likely to have high QoL in physical health and psychological domains. For women, upper SES was associated with high QoL in the physical health domain. These results are consistent with the well-established ‘SES-health gradient’ whereby people in lower SES groups are at greater risk of poor health. Indeed, material and social resources are vital to maintaining good physical and mental health.29 Men and women from both mid and upper SES groups were twice as likely to have high QoL in the environment domain compared with those from low SES. Previous work demonstrates SES is strongly related to exposure to air pollutants, ambient noise, residential crowding, housing quality and other negative environmental factors.30

Education level, a common indicator of SES, was also associated with the physical health, psychological and environment QoL domains. In most cases, a university education was associated with high QoL. However, in the case of women’s physical health related QoL, completion of secondary school, TAFE/Trade qualification and university education were all associated with a high QoL in this domain. Similar to benefits of higher SES, the positive impact of higher education in relation to QoL has been well documented.29

Notably, marital status was associated with all four QoL domains. For both men and women, those who were married or in a relationship were 2–3 fold more likely to have high QoL on the psychological and social relationships domains. For women, being in a relationship was also associated with high QoL in the physical health and environment domains. This is consistent with other studies that suggest those who are married or in a partnership have better physical and mental health31 and are less likely to experience psychological distress32 and loneliness.33 High QoL on the physical health and environment domains for women in a relationship may be reflective of the socioeconomic conditions typically associated with partnership, such as increased income and access to resources.

Employment status was only associated with the physical health domain for women. We found women who worked full time were more likely to have high QoL compared with those not working full time. It is likely that the ‘not working full-time’ group (which includes part-time work and unemployed) may comprise a number of people unable to work full-time due to physical health limitations. Previous work has suggested retirement is associated with better psychological well-being,34 however, we did not detect this in our sample. This may be unique of our cohort or due to the heterogeneous grouping of those on home duties with those who are retired. Further investigation that looks at the effect of employment status on QoL in more detail may be warranted. We found no associations between employment status and QoL domains for men in the adjusted models. This suggests employment status may not be a salient sociodemographic factor affecting QoL for men.

Strengths and limitations

The major strength of our study is that participants were randomly selected from the general population rather than from convenience samples or on the basis of disease. We report a relatively high response rate from this process. For the initial sample, 67% of men and 77% of women participated. For the follow-up assessment phases used in this study, 81% of eligible men participated and 75% of women. Furthermore, as participants were not excluded based on disease, data presented here are representative of the underlying population. Another strength of this study is the inclusion of data on older age groups (ie, over 80 years old). This age group has not been captured in other normative data sets.20 21 23

Our study also has limitations. Approximately 98% of participants within the GOS reported their race as white,26 therefore, the data presented here may not be cross-culturally representative. Those using these data should take caution when extrapolating to other ethnicities within the wider Australian population. There may also be a potential healthy participant bias, as all study participants had to be well enough to attend study visits. Furthermore, it should be noted that younger adults (<25 years) are not represented here, and numbers for those aged under 30 are small. Replication of this study in other regions and with younger samples could contribute to achievement of more comprehensive national values.

Conclusion

This study provides representative estimates for QoL for the general population in south-eastern Australia as assessed by WHOQOL-BREF. Our results can be useful to researchers and clinicians who can compare their findings to these data. We also report several associations with common sociodemographic factors impacting QoL in this population.

Data availability statement

No data are available. No additional data available.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

This study involves human participants and the study was approved by the Barwon Health Human Research Ethics Committee (ID 92/01 and ID 00/56). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

We would like to thank all participants of the Geelong Osteoporosis Study.