- Top of page
- PATIENTS AND METHODS
The selection of appropriate measures to assess study outcomes is an important part of planning any clinical research project and is often based on data from prior psychometric evaluations, ideally in a similar research setting. The assessment of medical intervention outcomes has moved from the traditional focus of measuring impairments to more comprehensive assessment that incorporates individuals' own perspectives and beliefs regarding their health status (1–3). Assessment of quality of life is now a major component of clinical research and has been conducted across a wide range of settings. To comprehensively assess outcome after joint replacement surgery, a combination of measures is required to produce a holistic picture of well-being. According to the World Health Organization, health is defined as “a state of complete physical, mental, and social well-being; not merely the absence of disease and infirmity” (available at www.who.int/about/definition/en). Using this definition, it is evident that although disease-specific measures such as the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) can provide valuable information about physical status after surgery, additional measures are also necessary to obtain more complete information about physical, mental, and social well-being.
One measure that could provide a broader perspective in this setting is the World Health Organization Quality of Life short version instrument (WHOQOL-BREF), a generic quality of life measure that forms part of a range of instruments developed by the World Health Organization Quality of Life (WHOQOL) Group through international collaboration (4, 5). These instruments were designed to provide a cross-cultural assessment of quality of life (4) by taking into account an individual's beliefs and situation in life. Although there is some overlap between information obtained from the WHOQOL-BREF and other generic measures such as the Medical Outcomes Study 36-Item Short Form (SF-36), a unique feature of the WHOQOL-BREF is its focus on the degree of satisfaction with aspects of an individual's life. In addition to covering areas relevant to joint replacement surgery such as pain, physical function, and capacity for work, the WHOQOL-BREF also provides information not available from measures such as the SF-36. This includes dependence on medical services and environmental aspects such as the home environment, transportation, and access to health services. While the cross-cultural design of the WHOQOL-BREF is important in multicultural societies such as Australia, other features also enhance its potential applicability. The shortened format (26 questions) may be of value in clinical research where participant burden is a concern, and as with all generic quality of life measures, the WHOQOL-BREF allows for comparison across diseases, interventions, and research settings. Its availability in a range of languages (6) and readily accessible normative data (6–8) could also facilitate future international comparisons of joint replacement outcomes.
The psychometric properties of the WHOQOL-BREF have been examined in several population-based studies (6–8) and a variety of patient groups, including individuals with rheumatoid arthritis (9), spinal cord injury (10), human immunodeficiency virus (HIV) (11), and those who have undergone liver transplantation (12). However, there are no published reports of its use in patients undergoing joint replacement surgery. Given its potential as an adjunct to disease-specific tools, an evaluation of the WHOQOL-BREF's measurement properties in this setting is warranted. The purpose of this study was to assess the psychometric properties of the WHOQOL-BREF in patients undergoing total hip replacement (THR) or total knee replacement (TKR) surgery, and to determine its ability to detect change following surgery.
- Top of page
- PATIENTS AND METHODS
Quality of life measures such as the WHOQOL-BREF can evaluate the impact of interventions such as THR and TKR through assessment of physical, mental, and social well-being. Such measures can capture the intended and unintended effects of treatment. By focusing on an individual's perceived quality of life across a range of domains, the WHOQOL-BREF is an appropriate measure to assess these effects. Although disease-specific measures have an important role, an advantage of generic tools such as the WHOQOL-BREF is that they allow comparison between patient groups and health care interventions. Our results indicate that several psychometric properties of the WHOQOL-BREF make it a valuable adjunct for assessing quality of life in patients undergoing joint replacement surgery.
In this study, all WHOQOL-BREF domains showed good internal consistency, with values similar to or higher than those reported for persons with rheumatoid arthritis (9), spinal cord injury (10), and HIV (11). Internal consistency was also comparable with that reported by the WHOQOL Group in a large international field trial (6). The low incidence of floor effects for all WHOQOL-BREF domains at baseline was particularly encouraging, considering the known poor health of individuals awaiting joint replacement surgery (13, 36). Baseline floor effects for the WHOQOL-BREF domains were similar to or lower than those found for the WOMAC subscales. Because participants can find a comprehensive range of response options, particularly at the lower end of the WHOQOL-BREF scale, researchers and health professionals are able to obtain more precise information across the disease continuum. Minimal ceiling effects were observed for the WHOQOL-BREF postoperatively except for the social relationships domain, suggesting that despite the overwhelmingly positive effect of surgery, participants could satisfactorily rate themselves within the higher end of the scale. Our finding of higher ceiling effects for the WHOQOL-BREF social relationships domain at both time points may relate to the small number of items contributing to this score, although this was not found previously (10).
The WHOQOL-BREF demonstrated expected correlations with other questionnaires at baseline. Moderate associations were seen between the physical domain and the WOMAC pain subscale, WOMAC physical function subscale, AQOL instrument, K10, and MHAQ. The magnitudes of the correlations (r = 0.54–0.67) indicate that, although there are substantial similarities between these measures, this domain also captures different information. In this study, correlation between the physical domain and the MHAQ was similar to that reported between the physical domain and the original HAQ in persons with rheumatoid arthritis (r = −0.62 versus −0.65) (9). The relationship between this domain and the AQOL measure was also similar to that reported in an Australian population study (r = 0.67 versus 0.69) (7). The present study also demonstrates that, although there was moderate correlation between the psychological domain and the K10 scale (r = −0.71), the WHOQOL-BREF provides supplementary information. The psychological domain covers aspects not included in any of the other instruments, such as satisfaction with oneself and enjoyment of life. A similar degree of correlation was previously found between this domain and the psychological component of the Quality of Life Profile (r = 0.69) (9) and the SF-36 Mental Component Score (MCS) (r = 0.70) (7). As expected, the social relationships domain was not strongly associated with the other scales. It was not possible to provide an ideal comparative measure for this domain because it covers unique concepts such as satisfaction with personal and sexual relationships and social support. Previous clinical validation studies have also shown only small to moderate correlations between the social relationships domain and other measures (9, 11). The present study also demonstrated that, despite measuring different concepts, there was moderate correlation between the WHOQOL-BREF environment domain and the K10 scale (r = −0.63). Similar correlations were seen between the K10 and the other domains, indicating that psychological distress is associated with general dissatisfaction in many aspects of a person's life. Moderate correlations between the SF-36 MCS and each WHOQOL-BREF domain were also found in an Australian validation study (7), and similar relationships were seen between a self-evaluated happiness scale and the WHOQOL-BREF domains in individuals with HIV (11).
Although the disease-specific WOMAC showed the greatest responsiveness, the WHOQOL-BREF detected significant improvements in quality of life for the physical, psychological, and environment domains. A significant difference in social relationships was not observed; this may be due to a lack of change in relationships in the early postoperative period or to poor responsiveness of this domain. Previous studies have shown that social relationships is the least responsive domain in patients undergoing liver transplantation (12) and in patients with rheumatoid arthritis (9), although the latter study reported a larger effect size (0.31). The physical domain, which had the largest effect size of the WHOQOL-BREF domains, had an effect size greater than that calculated for the AQOL instrument and comparable with that of the MHAQ. This domain was also found to be highly responsive in persons receiving inpatient treatment for rheumatoid arthritis (9). Because the physical domain focuses on pain, physical function, and activities of daily living, it was anticipated to be efficient in detecting improvement after joint replacement. Although this domain had a responsiveness similar to the MHAQ, a major shortcoming of the MHAQ in this setting was the large number of ceiling effects at 3 months (14%). Similar findings were reported in individuals with rheumatoid arthritis (37). In this study, the efficiency of the psychological domain was similar to that of the K10 instrument. The smaller effect sizes for both measures are consistent with previous findings that physical health improves to a greater degree than psychological health after joint replacement (38, 39). A small effect size was found for the environment domain, which covers areas such as financial situation, access to health services, and transportation. These factors may only improve to a small degree after surgery, secondary to improvements in physical function and return to regular activities.
A limitation of this research is that a considerable proportion of participants could not be followed up after surgery, predominantly due to protracted waiting times (almost 3 years for some patients). Analysis of baseline and demographic data indicated that, apart from a greater proportion undergoing THR, participants who provided postoperative data were similar to those who provided baseline data only. Because preliminary analyses (data not shown) demonstrated that the WHOQOL-BREF effect sizes for the overall sample were similar to those calculated separately for THR or TKR patients, this demographic difference is unlikely to have impacted significantly on the findings. Another limitation is the focus on the early postoperative period; additional improvements may continue after this time. A 3-month assessment was used because the greatest changes in health status were expected to have occurred by this time (39, 40); our data show that WHOQOL-BREF scores were comparable with population norms for 3 of the 4 domains by this point.
This study is the first to examine the psychometric properties of the WHOQOL-BREF in a mixed sample undergoing lower-limb joint replacement. Overall, the WHOQOL-BREF has good psychometric properties, compared with disease-specific and other generic measures, in individuals with severe joint disease. While the disease-specific measures provide detailed information about pain and physical function, the WHOQOL-BREF provides complementary information and presents clinicians and researchers with an additional tool for comprehensively assessing outcomes from joint replacement surgery.