The LupusQoL and Associations with Demographics and Clinical Measurements in Patients with Systemic Lupus Erythematosus

MATERIALS AND METHODS

The study was granted Multi-centre Research Ethics Committee approval and participants from the collaborating centers gave written informed consent. The collaborating rheumatology units were UK centers with an interest in SLE as part of the British Isles Lupus Assessment Group (BILAG): Bangor, Birmingham, Blackburn, University College London, Manchester, Newcastle, and Sheffield. All patients fulfilled 4 or more of the revised American College of Rheumatology (ACR) criteria for SLE^8,9.

The study design was cross-sectional and patients were approachedduring their routine rheumatology outpatient attendance. Those who consented to take part in the study either completed the LupusQoL at the outpatient clinic or took it home to complete and were instructed to return it by post within a week. Demographic (age, sex, ethnic group, marital status) and clinical information (duration of SLE, disease activity, damage) were recorded at the time of consent. Ethnicity was defined in accord with the UK Office of National Statistics¹⁰ and the main groups included White (British, Irish), Asian (Indian, Pakistani, Bangladeshi), Black (Black Caribbean, Black African), Mixed (White and Black Caribbean, White and Black African, White and Asian), Chinese, or others. The marital status data collected were as follows: married/living with partners, single, widowed, separated/divorced, and for statistical analysis, the patients were divided into 2 groups. Group 1 included those assumed to be not living with spouses or partners (single, widowed, separated/divorced) and Group 2 included those married or living with partners.

Disease activity was measured by the BILAG index^11,12 and damage by the Systemic Lupus International Collaborating Clinics ACR-Damage Index (SLICC ACR-DI)¹³, both of which were assessed on the date of consent. The BILAG index consists of 86 items (symptoms, signs, and laboratory results) that were scored as absent or present, and if present, whether that item was new, worse, the same, or better over the previous 4 weeks. These items are 8 organ systems (general, mucocutaneous, neurological, musculoskeletal, cardiovascular and respiratory, vasculitis, renal, and hematological), which are then each scored on a 5-point ordinal scale (A to E) based on the principles of the “physician’s intention to treat.” Although it was not the original intention of the BILAG to produce an overall score, for some types of analysis it became necessary to do so, and after some discussion a scoring system was developed: A = 9, B = 3, C = 1, D and E = 0¹⁴. The SLICC ACR-DI assesses damage over 12 organs or systems: ocular (range 0–2), neuropsychiatric (0–6), renal (0–3), pulmonary (0–5), cardiovascular (0–6), peripheral vascular (0–5), gastrointestinal (0–6), musculoskeletal (0–7), skin (0–2), gonadal (0–1), endocrine damage (0–1), and malignancy (0–2). The maximum possible total score is 48.

HRQOL using the LupusQoL was also described and the associations between the 8 domains of the LupusQoL and demographic and other clinical measures in these patients were explored.

RESULTS

DISCUSSION

Our study found that HRQOL in patients with SLE as measured by the LupusQoL was impaired in all domains, especially those of fatigue, physical health, and burden to others. The domain most impaired was fatigue, consistent with other studies^15,16,17,18. Fibromyalgia may have contributed to fatigue but we did not assess our patients for fibromyalgia, which may be a limitation of our study. However, a study has shown a low prevalence (10%) of fibromyalgia in the UK SLE population¹⁹, so it is unlikely to have contributed significantly to the high level of fatigue in our patients. Physical health has also been documented to be poor in patients with SLE in studies employing generic measures such as the SF-20^20,21 and the SF-36^{22,23,24,25,26}. The patients also perceive that having SLE places a significant burden on others close to them in both physical and psychological terms. Burden to others is a domain that has not been assessed in other HRQOL measures, and if our finding is confirmed, it may warrant specific intervention strategies.

Body image was the least impaired domain in this group of patients and this was at odds with the emphasis placed on it by patients during the semistructured interviews conducted for the item generation of the LupusQoL measure⁶. The literature is sparse regarding body image in patients with SLE and it has been reported to be worse in SLE than in normal controls or people with rheumatoid arthritis²⁷. When patients were subclassified into those with active scores (A or B) or inactive scores (C, D, or E) in the mucocutaneous system of the BILAG index, the median body image scores were significantly lower (poorer) in the active disease group (p < 0.05, data not shown). The number of patients with skin damage on the SLICC ACR-DI was small (n = 11) and thus statistical analysis was not undertaken. If the patients were subgrouped into ever having any skin involvement (whether currently active and with or without with skin damage), the body image median scores were not statistically different between the patients with and those without cutaneous involvement (data not shown). This emphasizes the difficulty in defining the concept of body image²⁸ and underscores that “body image” is a complex term that encompasses more than SLE-related skin rashes.

In our study, older age was associated with worse physical health and intimate relationships even though the associations were weak and unlikely to be clinically important. In terms of physical health, the findings were consistent with some studies^{22,24,25,29,30,31,32,33,34,35} but not others, where there was no association reported between age and poorer physical health^21,36,37,38. It would seem logical that older patients may have poorer physical health, because patients with longer disease duration have greater damage³⁹, including more degenerative joint disease, which damages physical health. Longer disease duration was found to be associated with better emotional health in the LupusQoL, but again this was unlikely to be clinically relevant. The literature findings were varied: some investigators reported no significant association between the two^21,25,32,37; others found that longer disease duration associated to better physical health⁴⁰, HRQOL (SF-36)³⁸, mental health^22,35, or role emotional³⁰; and one reported the association of disease duration with worse physical health²⁹. These discrepancies may be due to many reasons including demographic and disease-related differences between different cohorts.

The effect of gender on HRQOL has been examined in only 3 studies^35,38,40, and like our study, those found no differences between the sexes in terms of HRQOL. For marital status, patients were divided into 2 groups and HRQOL was similar in both. Two studies^36,40 reported no differences in HRQOL involving marital status; one reported that marriage improved social functioning³⁸ and another that those who lived alone had better HRQOL than those who did not live alone³⁵. However, the way we subdivided our groups and the assumptions we made about patients who were “living alone” could be erroneous and are potential limitations of our study. Further studies are needed with groups carefully defined from the onset.

Comparing the HRQOL scores in the ethnic groups, there were some differences in 2 of the LupusQoL domains. Black Caribbean patients were found to have better intimate relationships compared to White and Asian patients, but this was because of their younger age rather than their ethnicity. There were also statistically significant differences between Whites and Black Caribbeans in terms of age and marital status, Whites being older and more likely to be married (data not shown). Perhaps older age and being married reduce the need for sexual intimacy as a means of maintaining the partnership; this is supported by another study⁴¹ and may explain some of the ethnic differences in the intimate relationship domain. Although not statistically significant, the different levels of HRQOL between Black Caribbeans and Asians in terms of the “burden to others” domain may be clinically important. This HRQOL domain has not been explored in SLE populations. A UK study⁴² in patients with endstage renal disease reported that Asian patients perceive kidney disease as a social burden even after a successful kidney transplant. Ethnic differences in HRQOL have been examined by only a few authors, possibly because most study populations consisted of predominantly 1 ethnic group, usually Whites, in studies conducted in the West (Europe and the United States), and Chinese patients in studies conducted in Singapore. The relationship between ethnicity and HRQOL deserves further study and may be modulated by clinical status, socioeconomic factors such as poverty and access to medical services^35,43, and patient factors such as their beliefs in medicines and disease⁴⁴. In our study, socioeconomic status of the patients was not collected and therefore could not be analyzed, but in the UK, there is less association between ethnic minorities and poor socioeconomic status than in the United States¹.

In our study, 7% of patients had high disease activity as assessed by the BILAG index (A scores), 7% had no disease activity (D or E scores only), and the majority of patients had moderate or mild disease activity (scores no higher than B or C), and this is similar to other studies⁴⁵. The BILAG index assessments were carried out at a clinic and most patients completed the LupusQoL at that time. The remainder completed the questionnaire at home and returned it within 1 week, allowing appropriate correlation of their LupusQoL responses to the BILAG index assessments. The small number of patients completing the LupusQoL at home worked against meaningful comparison with those who completed the questionnaire at a clinic. Sixty percent of patients had no damage as assessed by the SLICC ACR-DI, and this is compatible with an outpatient group of patients with SLE. Recent UK studies also show low levels of damage, a condition probably related to a number of factors including diagnosis of milder cases, more aggressive treatment of SLE, and disease duration. In the study by Chambers, et al³⁹, 67% of patients had no damage at 5 years, dropping to 49% at 10 years. Most of our patients had disease duration < 10 years, with a median of 8 years (IQR 4 to 16 yrs). So the figure of 60% of patients without damage is plausible. In addition, many of the data in the literature are based on data from university teaching hospitals, which are tertiary referral centers treating patients with more severe disease, while our group of patients included those followed up at district general hospitals. Studies of patients with SLE in the United States may also show a higher level of damage in these patients compared to patients with SLE in the UK, where there is universal access to healthcare and education.

In our study, we found that active disease in some systems of the BILAG index and damage in some organs of the SLICC ACR-DI were associated with impairment in some domains of the LupusQoL. It is perhaps not surprising that the presence of general constitutional symptoms was related to impaired domains of physical health and fatigue, especially as there is an item measuring fatigue in that system. Activity in the musculoskeletal system may alter the patients’ perceptions of the physical health, body image, pain, and fatigue domains. Overall damage was, not surprisingly, associated with poorer physical health, pain, and planning. More surprisingly, damage in the ocular system of the SLICC ACR-DI was significantly correlated to the physical health, pain, planning, and intimate relationship domains. Ocular damage in SLE is most commonly due to cataracts and is likely to be a marker of steroid therapy prescribed for active SLE. Nevertheless, all these correlations, even if statistically significant, were weak and highlight the lack of association between disease activity, damage, and HRQOL. This lack of concordance between the 3 assessments of SLE (disease activity, damage, and HRQOL) underlines that each component measures different aspects of the effect of SLE on the patient and that to fully assess a patient with SLE, we need to measure not only disease activity and damage but also HRQOL. Thus the newly developed and validated disease-specific HRQOL measure, the LupusQoL, provides additional information to the clinical indices of activity and damage.