Gender Differences in Quality of Life in Patients with Systemic Lupus Erythematosus

Abstract

Systemic Lupus Erythematosus (SLE) predominantly affects women. Clinical phenotype and outcomes in SLE may vary by sex; complicated further by unique concerns depending on gender defined roles. Herein, we describe gender differences in disease specific quality of life (QOL) assessment scores using LupusPRO in a large International study.

Methods

Cross-sectional data from 1,803 SLE patients on demographics, self-identified gender status, LupusPRO and disease activity were analyzed. LupusPRO has two constructs: health-related QOL (HRQOL) and non-HRQOL. Disease activity and damage were evaluated using SELENA-SLEDAI and the SLICC/ACR-Damage Index (SDI), respectively. Non-parametric tests were used to compare QOL and disease activity by gender.

Results

122 men and 1681 women with SLE participated. Mean age was similar by gender. Damage score was greater among men. Men fared worse on the non-HRQOL social support domain than women (p=0.02). When comparing disease and QOL among men and women aged ≤45 years, men were found to have greater damage and worse social support than women. However, women fared significantly worse on lupus symptoms, cognition and procreation domains with trends for worse functioning on physical health and pain-vitality domains.

Conclusions

In the largest study of diverse SLE patients with a disease specific QOL tool, gender differences in QOL were observed on both HRQOL and non-HRQOL constructs. While men performed worse in the social support domain, women, especially those in the reproductive age group, fared worse on other domains. These observations may assist physicians in appropriately addressing QOL issues in a gender focused manner.

Introduction

Systemic Lupus Erythematosus (SLE) is an autoimmune disease that predominantly affects young women. Quality of life (QOL) and Health-related quality of life (HRQOL) among patients with SLE are poor (6–11). Among men with SLE, a higher prevalence of Caucasian ethnicity, serositis, and nephritis are noted. In addition there is early and greater damage accrual along with higher mortality (1;2). There are only a handful of studies focusing on SLE in men (1–5). QOL in men with SLE has not been previously reported, though one could expect a worse QOL among men with SLE based on greater and early disease damage accrual (1–5). Furthermore QOL is known to vary by age and gender among healthy populations (12;13) and in other diseases (14–16). Gender based life roles may also differentially impact QOL, especially as SLE preferentially affects the younger age group and women. Some of the roles in lives vary by gender, especially in context of age. Financial and physical independence, though relevant to those of either gender at all ages, may be of greater relevance among middle aged and older individuals than young. However, completion of education, embarking in the job market, dating and marriage are relatively more important milestones for younger individuals of either gender than older. Though procreation and parenting are important roles for either gender (in the reproductive age group), child bearing role may be more relevant among women of the reproductive age group. Hence, it is plausible that overall QOL and specific domains of QOL affected may differ by gender and age in SLE, especially in the younger and those in the reproductive age group.

We undertook this study to compare the QOL (overall and domains) among men and women with SLE, to better understand the gender differences in their QOL. Furthermore, we also evaluated QOL differences by gender, especially in the reproductive age group patients with SLE. Information generated may help inform development of personalized and targeted care strategies.

Methods

The study was approved by the Institutional Review Board of all participating sites. The Study of Outcomes in Lupus (SOUL) data repository includes data obtained from SLE patients enrolled from various sites within the United States of America, Canada, Asia and Europe for validation of the LupusPRO tool.

All patients met the American College of Rheumatology classification criteria for SLE (17). Demographic data (age, gender, ethnicity), and QOL data were available for all patients. A subset of patients had data available on disease activity and damage. Disease activity was measured using the Safety of Estrogens in Lupus Erythematosus National Assessment Version-SLE Disease Activity Index (SELENA-SLEDAI) (18). Latter includes a physician global assessment (PhGA) which ranges from 0–3. Disease damage was assessed using the Systemic Lupus International Collaborative Clinics/American College of Rheumatology SLICC-ACR damage Index (SDI) (19). Higher scores for the SELENA-SLEDAI and SDI represent greater disease activity and damage respectively.

QOL was measured using the LupusPRO (20). It has been validated in several languages (21–27), and shows measurement equivalence (21–28). The LupusPRO has a 5-point Likert response option: 0 = none of the time, 1 = a little of the time, 2 = some of the time, 3 = most of the time, 4 = all of the time, 5 = not applicable (recoded as 0 for scoring). It has two constructs (HRQOL and non-HRQOL) and 12 domains (8 HRQOL domains and 4 non- HRQOL domains). The HRQOL domains are lupus symptoms, cognition, lupus medications, procreation, physical health, pain-vitality, emotional health and body image. The non-HRQOL domains are desires-goals, social support, coping and satisfaction with treatment. Item scores are totaled for each domain item and the mean domain score is obtained by dividing the total score by the number of items in that domain. The mean raw domain score is transformed to scores ranging from 0 (worst QOL) to 100 (best QOL) by dividing by 4 (the number of Likert responses {5 responses} minus 1) and then multiplying by 100, as shown: (Mean raw domain score/4) × 100 = Transformed score for the domain. Transformed domain scores are obtainable when at least 50% of the items are answered. Total HRQOL and non-HRQOL scores are obtained by averaging the transformed domain scores within each construct.

Statistical analyses

All analyses were performed using IBM SPSS software, edition 19. Descriptive summary statistics were obtained for the two study groups (men and women). Chi-square analysis and Mann-Whitney U test were used to compare discrete and continuous variables, respectively among the two gender groups, as most data was not normative in distribution. When data was distributed normatively, the Student’s t test was used.

For sub-analysis, patients ≤ 45 years were selected and matched for age. Similar comparative analyses were undertaken as for the whole group.

For all tests, a two-sided p value ≤ 0.05 was considered significant.

Results

1803 patients with SLE, with mean (SD) age 40.8 (13.1) years participated (Table 1). Of these 122 (6.7%) were men. Racial composition of the study participants was as follows: 8% African Americans, 37% Caucasian, 14% Hispanic, 40% Asian and 1% others. The mean (SD) physician global assessment (PhGA) and SELENA-SLEDAI were 0.6 (0.7) and 3.4 (4.0) respectively. Mean (SD) SDI was 0.9 (1.5). Descriptives for QoL for the whole group are shown in Table 1. The domains most affected were cognition, pain-vitality, emotional health, desires-goals and coping.

Table 1.

Description of the study group

Variable	Value
Age (Years)	40.8, 13.1
Ethnicity (%)
African American	8.00
Caucasian	37.00
Hispanic	14.00
Asian	40.00
Other	1.00
Disease Features
Physician Global Assessment (n 920)	0.6, 0.7
SELENA-SLEDAI (n 1240)	3.4, 4.0
SDI (n 1247)	0.9, 1.5
SFI-Yes (%)	25
LupusPRO HRQOL
Lupus Symptoms	70.3, 24.9
Cognition	63.8, 28.9
Lupus Medications	68.3, 29.5
Procreation	77.9,32.0
Physical Health	76.7,27.0
Pain-Vitality	63.3,27.8
Emotional Health	58.3,27.8
Body Image	71.0,29.1
LupusPRO non HRQOL
Desires-Goals	65.6, 28.6
Social Support	66.0,31.4
Coping	65.5, 26.0
Satisfaction with Treatment	66.5, 32.6

All values are mean, SD unless specified

There were no differences by age, ethnicity (not shown), or disease activity among men and women (Table 2). Damage tended to be greater among men than women (Median (IQR): 1.0 (2.0) vs. 0.0 (1.0), p=0.03).

Table 2.

Comparisons between men and women with SLE

		Men (n 122)	Women (n 1681)
Variable	Unit	Values		P value
Age (Years)	Mean, SD	39.5, 14.1	40.9,13.0	0.70
Physician Global Assessment	Median, IQR, n	1.0,1.2, 55	0.4,0.9,865	0.15
SELENA-SLEDAI	Median, IQR, n	2.0,4.0,77	2.0, 4.0,1163	0.83
SDI	Median, IQR, n	1.0, 2.0, 72	0.0,1.0,1175	0.03
LupusPRO-HRQOL
Lupus Symptoms	Median, IQR	83.3,33.3	75.0,33.3	0.06
Cognition	Median, IQR	75.0,37.5	62.5,37.5	0.14
Lupus Medications	Median, IQR	62.5,50.0	75.0,50.0	0.22
Procreation	Median, IQR	100,25.0	100,37.5	0.08
Physical Health	Median, IQR	90.0,25.0	85.0,40.0	0.18
Pain-Vitality	Median, IQR	75.0,45.0	65.0,50.0	0.12
Emotional Health	Median, IQR	58.3,45.8	58.3, 45.8	0.66
Body Image	Median, IQR	80.0,40.0	80.0, 50.0	0.49
LupusPRO-non HRQOL
Desires-Goals	Median, IQR	68.8,46.9	68.8, 50.0	0.87
Social Support	Median, IQR	62.5,37.5	75.0,50.0	0.02
Coping	Median, IQR	58.3,33.3	66.7,33.3	0.40
Satisfaction with Treatment	Median, IQR	75.0, 56.3	75.0,56.3	0.24

Among men, the domains with median scores ≤ 65 were lupus medications, emotional health, social support and coping. Among women, the domains with median scores of ≤ 65 were cognition, pain-vitality and emotional health (Table 2). In the social support domain, men had worse median scores (62.5, IQR=37.5) as compared to women [75.0 (IQR = 50.0) (p=0.02)]. There were trends noted with worse performance among women in lupus symptoms [(Median (IQR): 75.0 (33.3) vs. 83.3 (33.3), p=0.06)]. Trends were also noted for worse functioning among women in the cognition, physical health and pain-vitality domains (Table 2). Comparison of QOL by gender is shown in the spydergram in Figure 1.

Figure 1.

QOL Spydergram for men and women with SLE.

When men were age-matched and compared with women in the reproductive age group (age ≤ 45 years), greater damage was again noted among men with a median of 1.0 (IQR=2) as compared to women that had a median of 0.0 (IQR=1) (p <0.001). Women scored significantly lower than men on lupus symptoms, cognition, and procreation domains (Table 3) with trends for lower scores on physical health and pain-vitality. Domains with median scores ≤ 65 among men were lupus medications, emotional health, coping, and social support (Table 3). Domains with median scores ≤ 65 among women were cognition and pain-vitality and emotional health.

Table 3.

Comparison between men and women with SLE, aged ≤ 45 years

		Men (n 117)	Women (n 1080)
Variable	Unit	Values		P value
Age (Years)	Mean, SD	38.2, 12.9	32.9,7.2	0.09
Physician Global Assessment	Median, IQR, n	1.0,1.2,54	0.4,0.8,552	0.60
SELENA-SLEDAI	Median, IQR, n	2.0,4.0,73	2.0,5.0,747	0.9
SDI	Median, IQR, n	1.0,2.0,70	0.0,1.0,747	<0.001
LupusPRO-HRQOL
Lupus Symptoms	Median, IQR	83.3,33.3	75.0,41.7	0.01
Cognition	Median, IQR	75.0,37.5	62.5,50.0	0.04
Lupus Medications	Median, IQR	62.5,50.0	75.0,50.0	0.64
Procreation	Median, IQR	100,25.0	75.0,50.0	0.008
Physical Health	Median, IQR	90.0,25.0	85.0,45.0	0.10
Pain-Vitality	Median, IQR	75.0,45.0	65.0,45.0	0.08
Emotional Health	Median, IQR	58.3,45.8	58.3,45.8	0.93
Body Image	Median, IQR	75.0,40,0	75.0,50.0	0.59
LupusPRO-non HRQOL
Desires-Goals	Median, IQR	68.8,46.9	68.8,43.8	0.42
Social Support	Median, IQR	62.5,37.5	75.0,50.0	0.01
Coping	Median, IQR	58.3,31.3	66.7,33.3	0.47
Satisfaction with Treatment	Median, IQR	75.0,56.3	75.0,50.0	0.3

Discussion

There are gender specific QOL issues in SLE, and it is important for physicians, and health care systems to be aware of these while providing individualized medical care to patients with SLE. We found poor social support to be a key QOL issue among men with SLE, irrespective of their age. This may be plausible theoretically, as SLE is traditionally known to be a “woman’s disease,” due to its 9: 1 predilection for women over men. Men may not feel comfortable discussing their having SLE or matters related to that with their social support circle. It is possible that there are gender-based differences in both the receiver and provider of social support. Men with SLE may not be similar to women in their ability regarding communicating their need for support or reaching out for social support as readily or effectively, or receiving help. It is well known that men and women differ on a wide variety of behavioral, cognitive and affective dimensions (29). Men may perceive the need for social support and/or emotional fragility as indicative of weakness or hypo-masculinity, due to their traditional social role of being the provider or protector (34). One of the features of male gender role in society is defined by strength and it includes emotional toughness, courage, self-reliance and rationality (34). Any deviation from this expected role may subject men to harassment and discrimination. Women, on the other hand, are more likely to seek out social support (30–33). Another possibility is paucity of gender appropriate informational resources available for men with SLE as the disease is more common among women. Social support is important in SLE, and its relationship with health in SLE recognized (37–38).

Among women with SLE, main QOL issues that differed from men were related to procreation, especially among the younger patients. This is plausible as the stereotypical roles assigned by the society are still somewhat gender centric (35). The woman’s role has traditionally included child bearing/rearing, care of the family members and home (35). Thus, ongoing or new lupus symptoms, unpredictable disease flares and effects on cognition may interfere with planning, scheduling or partaking in these “role performance” functions. This may explain the poorer QOL among women as compared to men (≤ 45 years) in the lupus symptoms, cognition and procreation domains. Of note the emotional health domain was the most adversely affected in both the gender.

Damage was greater among men as compared to women overall and in the reproductive age group in our study. This finding hypothetically could be attributable to disease duration and age. On further analysis (not shown) male gender remained an independent predictor of greater damage overall and in the reproductive age group, even after accounting for age and disease duration. Greater damage among men has previously been noted by others as compared to women with SLE (1–5).

The study findings have great potential application and relevance to clinical practice. Addressing procreational concerns among women of reproductive age and social support among men, in addition to an overall evaluation of QOL, especially emotional health for both genders, is important in SLE. Discussing side effects of medications, especially those on procreation, among women with SLE within the reproductive age group, may help allay some concerns and anxiety. In addition, physicians need to be comfortable asking for patients’ immediate, intermediate and long term procreational plans, and accordingly, offer an individualized management plan for SLE, that includes indications, relative benefits, harms and alternatives of various treatment options.

There are several limitations to our study. We included only 122 men given the relatively rarity of SLE among men. The only other SLE study that included a comparison of QOL by gender in its analysis included 54 men and 54 women (36). They used the Short Form 36 tool for HRQOL and reported better functioning on vitality and mental health domains among men than women with SLE. Another limitation was lack of availability of complete data on disease activity and damage from all patients. These data were available on nearly 1250 patients; 25% of these had an ongoing flare at the time of the study. In addition, 40% of our participants were Asians, and only 8% were African Americans. This may affect the generalizibility of the results. Lastly, we did not have any data on fibromyalgia. As QOL is affected by fibromyalgia, and up to 1/3 of SLE patients may concomitantly have fibromyalgia, it would have been of interest to adjust for this covariate when comparing QOL domains between genders. Despite these limitations, this study addresses HRQOL in men with SLE, an area which has not been previously addressed.

Strengths of our study include large numbers of participants from varied geographic and racial backgrounds. Use of a disease targeted QOL tool, which includes a non-HRQOL construct, allowing for these comparisons is another advantage as these domains are not routinely represented in generic and other disease targeted tools. Some of the domains that did show gender differences in QOL in SLE are not included in generic QOL instruments and some of the other disease specific HRQOL tools. These unique domains include lupus symptoms, cognition, lupus medications, procreation and social support. Furthermore, as the LupusPRO was developed using feedback from both men and women with SLE, and uses a gender-neutral language, it was possible to administer the tool to patients of either gender and assess all SLE related pertinent aspects of their QOL.

In summary, we found gender differences in damage and QOL among patients with SLE. These were especially evident in the reproductive age groups and may be partly attributable to traditional gender roles. A focused evaluation and discussion of effects of SLE on a patient’s QOL should also include gender specific QOL concerns, and these should be considered when developing a patient’s management plan.

Significance and Innovation.

1. Quality of life concerns in Lupus can be different based on patient’s gender and age. These have never been systematically studied previously.

2. Awareness of gender pertinent quality of life concerns among Lupus patients can assist physicians in providing personalized care.

3. This study utilized disease specific patient reported outcome to delineate gender specific quality of life concerns in an ethnically heterogeneous large group of patients with Lupus from various countries.