Association between disease duration and comorbidity burden in systemic lupus erythematosus: a multicentre study from Pakistan

Abstract

Objective

This study aimed to evaluate the prevalence and predictors of cardiovascular disease (CVD), chronic kidney disease (CKD) and osteoporosis among patients with systemic lupus erythematosus (SLE) in Punjab, Pakistan.

Methods

A cross-sectional analysis of a prospective registry was conducted using data from the Punjab Lupus Registry between January 2020 and December 2024. Adults (≥18 years) fulfilling the 2019 European League Against Rheumatism/American College of Rheumatology SLE classification criteria were included. Of 536 registered patients, 482 with complete data were analysed. Comorbidities were physician-confirmed, and disease activity was assessed using Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K). Associations between disease duration and comorbidities were examined using χ² tests for trend, while predictors of comorbidity burden were identified using multivariable Poisson regression with robust error variance.

Results

Among 482 patients (mean age 39.6±12.4 years; 67.2% female), 43.6% had at least one comorbidity and 15.1% had multimorbidity. The prevalence of CVD, CKD and osteoporosis increased with disease duration: 12.3%, 10.6% and 8.2% in <5 years versus 39.6%, 28.9% and 25.4% in >10 years (p<0.001). Older age (adjusted prevalence ratios (aPR) 1.31 per 10 years), longer disease duration (aPR 1.22 per 5 years), corticosteroid use (aPR 1.33), smoking (aPR 1.26) and low socioeconomic status (aPR 1.38) were independent predictors, while hydroxychloroquine use was protective (aPR 0.78). Patients with comorbidities had higher disease activity (SLEDAI-2K 6.9 vs 5.6, p=0.009) and greater organ damage (Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index 2.4 vs 1.3, p<0.001).

Conclusion

Comorbidities are common in SLE and increase with age and disease duration, underscoring the need for early, integrated management to improve outcomes.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Systemic lupus erythematosus (SLE) increases the risk of chronic comorbidities such as cardiovascular disease, kidney disease and osteoporosis due to systemic inflammation and long-term corticosteroid use.

• Evidence from South Asia is limited, despite regional differences in genetic and socioeconomic factors.

WHAT THIS STUDY ADDS

• This study provides the first multicentre, registry-based data from Pakistan on comorbidity burden in SLE.

• It shows that comorbidity prevalence increases with older age, longer disease duration and higher corticosteroid exposure.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Findings highlight the need for earlier screening and targeted management of high-risk SLE patients.

• This study provides baseline data to inform future longitudinal research and guide prevention efforts in South Asia.

Introduction

Systemic lupus erythematosus (SLE) is a chronic, multisystem autoimmune disease characterised by loss of immune tolerance, autoantibody production and immune complex deposition, leading to inflammation and tissue damage. Advances in diagnosis and immunosuppressive therapy have improved survival in SLE, but this success has also shifted the clinical burden towards long-term comorbidities and treatment-related complications. The accumulation of comorbid conditions in SLE is now recognised as a major determinant of morbidity, mortality and reduced quality of life.¹

Among these comorbidities, cardiovascular disease (CVD), chronic kidney disease (CKD) and osteoporosis are the most prevalent and clinically significant. Patients with SLE have approximately two to three times higher risk of developing CVD compared with the general population, attributable to both traditional risk factors (such as hypertension, dyslipidaemia and smoking) and SLE-specific factors including chronic inflammation, endothelial dysfunction and corticosteroid exposure.² A large meta-analysis reported that SLE patients have a 2.35-fold increased risk of cardiovascular events overall, including ischaemic heart disease and stroke.³ Even in young women, who are otherwise at low cardiovascular risk, premature atherosclerosis is well documented in SLE, highlighting the profound influence of systemic inflammation on vascular health.⁴

Renal involvement is another common and serious manifestation of SLE. Lupus nephritis develops in approximately 30%–60% of patients during the disease course and can progress to CKD or end-stage renal disease.⁵ Recent cohort studies report that up to half of SLE patients may develop CKD during long-term follow-up, and that both disease activity and cumulative exposure to nephrotoxic medications contribute to this risk.⁶ Disease duration is also an important predictor of renal damage; patients with longer disease duration show higher rates of CKD and irreversible renal dysfunction.⁷ Notably, CKD itself is a well-established and independent risk factor for CVD, contributing to accelerated atherosclerosis, vascular calcification and endothelial dysfunction even in non-SLE populations.⁸ This inter-relationship underscores the compounded cardiovascular burden among patients with SLE who develop renal impairment.

Osteoporosis and bone mineral density (BMD) loss are equally important complications, arising from both the inflammatory process and chronic glucocorticoid use. A meta-analysis including over 3000 SLE patients found a pooled osteoporosis prevalence of approximately 16%, while reduced BMD was present in nearly 45% of patients.⁹ Other studies have demonstrated that longer disease duration, higher cumulative steroid dose and postmenopausal status are independently associated with low BMD and fracture risk in SLE.¹⁰ These findings emphasise the need for early prevention, screening and treatment strategies targeting skeletal health in patients with SLE.

Importantly, as patients live longer with SLE, comorbidities often cluster, resulting in multimorbidity, the coexistence of two or more chronic conditions, which compounds the burden of disease and complicates management.¹¹ Several studies have suggested that disease duration is a major driver of comorbidity accumulation, reflecting the cumulative impact of inflammation, medication exposure and irreversible organ damage.¹² Moreover, prior studies have used ORs, which can overestimate associations for common outcomes, rather than prevalence ratios, which are more appropriate for cross-sectional data.¹³

Given these gaps, the present study aimed to quantify the prevalence of major comorbidities, specifically CVD, CKD and osteoporosis among patients with SLE enrolled in our registry and to examine their association with disease duration after adjusting for key confounders such as age, sex, body mass index (BMI), smoking and medication use. By identifying patterns of comorbidity accumulation, this study seeks to provide evidence to inform screening priorities and preventive care strategies in SLE, particularly in resource-limited settings.

Methods

Study design and setting

This was a cross-sectional analytical study based on data conducted among patients with SLE who were registered in 15 major government and private hospitals of Punjab, Pakistan Lupus Registry between January 2020 and December 2024. The registry is a prospective observational database maintained by the Department of Rheumatology, which systematically records demographic, clinical, serological and treatment-related information of patients fulfilling the 2019 European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) classification criteria for SLE.¹⁴

Data are updated during routine follow-up visits every 3–6 months. For this analysis, data were extracted at a single time point corresponding to the most recent clinic visit within the study period. The registry operates within a tertiary referral centre serving a large, ethnically diverse lupus population, allowing for comprehensive longitudinal data collection through electronic medical records (EMR). All data are verified by trained clinical research coordinators and audited annually to maintain quality assurance.

The primary objective was to quantify the burden of comorbidities, specifically CVD, CKD and osteoporosis, and to determine their association with disease duration among patients with SLE.

Written informed consent was obtained from all participants at registry enrolment, allowing the use of anonymised data for research purposes.

Study population

All adults aged 18 years or older with a confirmed diagnosis of SLE according to the 2019 EULAR/ACR criteria were eligible for inclusion.¹⁵ Patients were required to have at least 12 months of follow-up within the registry to ensure adequate data completeness and reliability of comorbidity documentation. We excluded individuals with overlap syndromes such as mixed connective tissue disease or rheumatoid arthritis SLE overlap as well as pregnant women at the time of data collection, given that pregnancy induces physiological changes in renal function and bone metabolism that could confound study results.¹⁶ Patients with incomplete data for any of the key study variables, disease duration, comorbidities or medication history were also excluded.

A priori sample size estimation was performed using G*Power. The calculation was based on detecting a difference in the prevalence of comorbidities (CVD, CKD or osteoporosis) between patients with shorter disease duration (<5 years) and those with longer duration (>10 years). Assuming proportions of 30% versus 45%, a two-tailed test, an α level of 0.05 and a statistical power of 0.95, the required minimum sample size was estimated to be 536 participants.

Handling of missing data

Initially, data were available for 536 registered patients in the Lupus Registry. After applying the inclusion and exclusion criteria, participants with incomplete or inconsistent data regarding comorbidities, disease duration or medication history were excluded. Missing data were assessed for randomness using Little’s Missing Completely at Random (MCAR) test, which indicated that data were missing completely at random (p>0.05). Given the low proportion and random pattern of missingness, listwise deletion was applied to ensure analytic integrity.

Data collection and variables

Data were extracted from the Lupus Registry and validated using hospital EMRs, laboratory databases and pharmacy records to ensure accuracy. Trained data managers performed cross-verification and resolved discrepancies through direct chart review.

Sociodemographic and lifestyle variables

Sociodemographic information included age, sex and self-reported ethnicity. Age was analysed as a continuous variable and also categorised into four groups (18–29, 30–44, 45–59 and ≥60 years) for descriptive comparison. BMI was calculated as weight in kilograms divided by height in metres squared (kg/m²) and classified according to the WHO criteria (WHO, 2021) as underweight (<18.5), normal (18.5–24.9), overweight (25.0–29.9) or obese (≥30.0). Smoking status was categorised as current, former or never smoker.

Socioeconomic status (SES) was assessed using a composite index derived from registry and hospital data, including education level, occupation, approximate household income and residential area (urban vs rural). Where income data were unavailable, SES classification was inferred from employment type and healthcare facility type (public vs private). Each variable was scored from 1 (low) to 3 (high), and a total SES score (range 4–12) was computed. SES was then categorised as low (4–6), middle (7–9) or high (10–12). This approach was adapted from the modified Kuppuswamy socioeconomic scale validated for use in South Asian populations.¹⁷

Disease-related variables

Disease duration was defined as the number of years between the date of SLE diagnosis (confirmed by the treating rheumatologist) and the date of registry data extraction. For some analyses, duration was categorised as <5 years, 5–10 years, or >10 years to explore duration–response trends.

Disease activity at the time of data extraction was assessed using the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K),¹⁸ and cumulative organ damage was evaluated using the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SLICC/ACR-DI).¹⁹

Serological variables included complement levels (C3 and C4), anti-dsDNA antibody status and presence of antiphospholipid antibodies, all measured within ±3 months of the registry visit. Laboratory parameters such as serum creatinine, estimated glomerular filtration rate (eGFR; using the CKD Epidemiology Collaboration (CKD-EPI) equation), lipid profile and fasting glucose were extracted from the hospital’s laboratory information system.

Medication use

Medication data were reviewed for both current and past use based on physician documentation and pharmacy dispensing records. Corticosteroid exposure was categorised as current use (yes/no) and quantified by daily dose (prednisone equivalent, mg/day) and cumulative duration. Long-term corticosteroid use was defined as a daily dose ≥5 mg for ≥6 months.²⁰ Daily glucocorticoid doses were further stratified into three groups (≤5 mg/day, >5–7.5 mg/day and >7.5 mg/day; prednisone-equivalent) to assess potential dose-related differences in comorbidity prevalence.

Medication data were reviewed for both current and past use based on physician documentation and pharmacy dispensing records. Corticosteroid exposure was categorised as current use (yes/no), quantified by daily dose (prednisone equivalent, mg/day) and cumulative duration. Long-term corticosteroid use was defined as a daily dose ≥5 mg for ≥6 months. Hydroxychloroquine (HCQ) use was recorded as yes/no, with duration in months where available. Immunosuppressants included azathioprine, mycophenolate mofetil, methotrexate and cyclophosphamide, coded as ever versus never use. Biologic therapies such as belimumab and rituximab were recorded similarly. Data on bone-protective treatments, including calcium/vitamin D supplementation and bisphosphonate use, were also collected, particularly among patients with long-term corticosteroid exposure.

Medication variables were chosen based on their potential impact on cardiovascular, renal and bone health in SLE.²¹

Comorbidity assessment

Comorbidities were physician-diagnosed and confirmed through clinical and laboratory data.

• CVD: defined as a history of ischaemic heart disease, angina, myocardial infarction, stroke or peripheral vascular disease, supported by clinical records or imaging reports.

• CKD: defined according to the KDIGO 2021 criteria as eGFR <60 mL/min/1.73 m² or the presence of structural or functional abnormalities persisting ≥3 months.²²

• Osteoporosis: defined as a BMD T-score ≤−2.5 at the lumbar spine, femoral neck or total hip on dual-energy X-ray absorptiometry or a history of fragility fracture-confirmed radiologically.²³

Multimorbidity was defined as the coexistence of two or more of the above comorbidities.²⁴

Outcome measures

The primary outcome was the presence of at least one major comorbidity (CVD, CKD or osteoporosis). Secondary outcomes included the prevalence of multimorbidity (≥2 comorbidities) and the relationship between disease duration and individual comorbidities.

Statistical analysis

All data were analysed using R. Descriptive statistics were used to summarise baseline demographic, clinical and socioeconomic characteristics of patients with SLE. Continuous variables were expressed as mean±SD or median (IQR), depending on data distribution assessed by the Shapiro-Wilk test. Categorical variables were presented as frequencies and percentages.

Comparisons between patients with and without comorbidities were performed using the independent samples t-test or Mann-Whitney U test for continuous variables, and the χ² test or Fisher’s exact test for categorical variables. Cohen’s d effect sizes with 95% CIs were calculated to estimate the magnitude of group differences, interpreted as small (0.2–0.49), medium (0.5–0.79) and large (≥0.8).

The relationship between disease duration and the prevalence of specific comorbidities CVD, CKD (CKD) and osteoporosis was analysed using the χ² test for trend to evaluate dose–response patterns across duration categories (<5 years, 5–10 years >10 years).

To identify independent predictors of comorbidity burden, a multivariable Poisson regression model with robust error variance was used. Adjusted prevalence ratios (aPRs) and 95% CIs were calculated for variables including age, sex, BMI, disease duration, smoking status, corticosteroid use, HCQ use, immunosuppressant use and SES. Variables with p<0.10 in univariate analysis were included in the multivariable model.

To explore inter-relationships among continuous variables (eg, age, disease duration, corticosteroid dose, SES, SLEDAI-2K, number of comorbidities), Pearson’s correlation coefficients (r) were computed and displayed as a correlation matrix. Correlation strength was interpreted as weak (r=0.10–0.29), moderate (0.30–0.49) or strong (≥0.50).

Comparisons of disease activity and organ damage (SLEDAI-2K and SLICC/ACR-DI) between patients with and without comorbidities were assessed using independent t-tests. Inflammatory markers (erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), complement C3 and anti-dsDNA) were similarly analysed to evaluate associations between comorbidity burden and systemic inflammation.

All statistical tests were two-tailed, and a p value <0.05 was considered statistically significant.

Results

Figure 1 illustrates the systematic process of patient selection for the study, starting from the initial screening of 536 SLE patients. A total of 54 patients were excluded due to missing key data, which included incomplete disease duration (n=18), incomplete comorbidity information (n=14) and missing laboratory or disease activity data (SLEDAI-2K/SLICC-ACR DI) (n=22). The remaining 482 patients formed the final analysed sample. This cohort had complete demographic, clinical and medication data and was subsequently included in regression, correlation and subgroup analyses for the study.

Figure 1. Patient selection flowchart for the study of systemic lupus erythematosus (SLE) patients.

Table 1 presents the baseline demographic, clinical and socioeconomic characteristics of 482 patients with SLE. The mean age of participants was 39.6±12.4 years, with a female predominance (67.2%). Patients with at least one comorbidity were significantly older (46.2±11.6 vs . 34.8±10.9 years, p=0.005) and had longer disease duration (10.9 vs 5.9 years, p=0.001) compared with those without comorbidities. They also demonstrated higher disease activity (SLEDAI-2K score: 6.9 vs 5.6, p=0.009) and greater cumulative organ damage (SLICC/ACR-DI score: 2.4 vs 1.3, p=0.003).

Table 1. Baseline demographic, clinical and socioeconomic characteristics of participants (N=482).

Characteristic	Total (N=482)	With≥1 comorbidity (n=210)	Without comorbidity (n=272)	P value	Cohen’s d (95% CI)
Age (years)	39.6±12.4	46.2±11.6	34.8±10.9	0.005	0.92 (0.71 to 1.13)
Male, n (%)	158 (32.8)	70 (33.3)	88 (32.4)	0.77	0.03 (–0.14 to 0.20)
Female, n (%)	324 (67.2)	140 (66.7)	184 (67.6)	–	–
BMI (kg/m²)	24.9±4.6	25.9±5.1	24.3±4.1	0.004	0.34 (0.13 to 0.55)
Current smoker, n (%)	85 (17.6)	47 (22.4)	38 (14.0)	0.02	0.25 (0.05 to 0.45)
Disease duration (years)	8.3((4.2–13.5)	10.9((6.9–16.0)	5.9((2.8–9.5)	0.001	0.87 (0.65 to 1.09)
SLEDAI-2K score	6.2±4.4	6.9±4.8	5.6±4.1	0.009	0.30 (0.09 to 0.51)
SLICC/ACR-DI score	1.8±1.1	2.4±1.2	1.3±0.9	0.003	0.91 (0.69 to 1.13)
Employment status				0.002	0.36
Employed	198 (41.1)	69 (34.8)	129 (47.4)	–	–
Unemployed/housewife	213 (44.2)	104 (48.8)	109 (40.1)	–	–
Retired	71 (14.7)	37 (52.1)	34 (12.5)	–	–
Residence				0.04	0.25
Urban	354 (73.4)	139 (39.3)	215 (79.0)	–	–
Rural	128 (26.6)	71 (55.5)	57 (21.0)	–	–
Socioeconomic status (SES)				0.005	0.44
Low SES	132 (32.0%)	75 (44.1%)	57 (23.6%)	–	–
Middle SES	193 (46.8%)	78 (45.9%)	115 (47.5%)	–	–
High SES	87 (21.1%)	17 (10.0%)	70 (28.9%)	–	–

BMI, body mass index; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000; SLICC/ACR-DI, Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index.

Regarding lifestyle and socioeconomic factors, current smoking was more frequent among patients with comorbidities (22.4% vs 14.0%, p=0.02). Employment and residence were also associated with comorbidity burden; patients with comorbidities were more often retired and resided in rural areas (55.5% vs 21.0%, p=0.04). SES was significantly associated with comorbidities, with a higher proportion of patients with low SES having at least one comorbidity (44.1% vs 23.6%, p=0.005).

These findings indicate that older age, longer disease duration, higher disease activity, rural residence and lower SES are linked to an increased likelihood of comorbidities among patients with SLE.

Table 2 outlines medication use among 482 patients with SLE and its relationship to comorbidities. Corticosteroids were prescribed to 344 patients, with 110 on ≤5 mg/day, 150 on >5–7.5 mg/day and 84 on >7.5 mg/day; those with comorbidities received higher mean daily doses (9.2 mg/day) and more frequent long-term therapy (72.6%) compared with patients without comorbidities. HCQ was used by 420 participants, while 280 received immunosuppressants, particularly azathioprine and cyclophosphamide, which were more common among patients with comorbidities. Bone-protective agents such as calcium with vitamin D and bisphosphonates were also more frequently used in patients with comorbidities, indicating preventive management of corticosteroid-related bone loss. Biologic therapy was administered to 38 patients, with slightly greater use in those with comorbid conditions. Overall, greater corticosteroid exposure, specific immunosuppressant use and bone-protective therapy were associated with the presence of comorbidities in this cohort.

Table 2. Medication use and associations with comorbidities.

Medication variable	Total N (%)	With≥1 comorbidity % (95% CI)	Without % (95% CI)	P value	Cohen’s d
Corticosteroid use	344 (71.4)	81.9 (77.0 to 86.1)	63.2 (57.3 to 68.8)	0.005	0.46
≤5 mg/day	110	35.5 (28.0 to 43.7)	64.5 (55.8 to 73.2)	0.02	0.35
>5–7.5 mg/day	150	49.8 (41.8 to 58.2)	50.0 (41.8 to 58.2)	0.01	0.40
>7.5 mg/day	84	64.2 (55.1 to 73.8)	35.0 (26.2 to 44.9)	0.003	0.50
Mean daily dose (mg/day)	7.8±4.6	9.2±4.9	6.7±4.0	0.003	0.56
Long-term corticosteroid (≥ 6 months)	266 (55.2)	72.6 (66.6 to 78.0)	44.8 (38.7 to 51.1)	0.005	0.60
Hydroxychloroquine use	420 (87.1)	82.9 (77.6 to 87.1)	90.4 (86.1 to 93.5)	0.02	–0.24
Immunosuppressant (any)	280 (58.1)	64.8 (58.2 to 70.9)	52.9 (47.1 to 58.6)	0.01	0.25
Azathioprine	134 (27.8)	35.7 (28.1 to 44.1)	21.7 (16.8 to 27.4)	0.003	0.31
Mycophenolate mofetil	96 (19.9)	21.0 (15.3 to 28.2)	18.9 (14.3 to 24.6)	0.55	0.06
Methotrexate	72 (14.9)	17.1 (11.7 to 24.4)	13.2 (9.1 to 18.8)	0.29	0.12
Cyclophosphamide	54 (11.2)	15.7 (10.0 to 23.8)	8.5 (5.3 to 13.3)	0.03	0.25
Biologic therapy (belimumab/rituximab)	38 (7.9)	11.0 (6.2 to 18.7)	5.9 (3.3 to 10.2)	0.08	0.20
Bone-protective therapy (any)	152 (31.5)	41.9 (35.1 to 49.0)	24.6 (19.6 to 30.4)	0.002	0.38
Calcium+vitamin D	142 (29.5)	40.0 (33.3 to 47.1)	22.4 (17.5 to 28.3)	0.004	0.39
Bisphosphonate	54 (11.2)	18.1 (12.3 to 25.8)	6.3 (3.7 to 10.4)	0.004	0.34

As shown in table 3, the overall prevalence of comorbidities increased progressively with longer disease duration. Among patients with disease duration <5 years, any comorbidity was present in 12.3% (95% CI 7.2 to 19.9), and multimorbidity (≥ 2 comorbid conditions) in 4.9% (95% CI 2.0 to 11.3). These proportions rose to 26.2% and 11.3%, respectively, in patients with 5–10 years of disease and reached 39.6% (95% CI 31.8 to 47.9) and 20.8% (95% CI 14.7 to 28.6) among those with >10 years of disease duration. A significant trend towards higher comorbidity burden with increasing disease duration was observed (χ²=14.8, p<0.001).

Table 3. Comorbidity prevalence by disease duration and demographic subgroups.

Subgroup	Any comorbidity % (95% CI)	Multimorbidity % (95% CI)	P value	Interpretation
Disease duration <5 years (n=142)	12.3 (7.2 to 19.9)	4.9 (2.0 to 11.3)	<0.001*	Lowest prevalence
5–10 years (n=161)	26.2 (19.5 to 34.2)	11.3 (6.7 to 18.2)
>10 years (n=179)	39.6 (31.8 to 47.9)	20.8 (14.7 to 28.6)		Trend ↑ with duration
Male (n=158)	47.5 (39.6 to 55.5)	17.7 (12.0 to 25.4)	0.24	Higher but NS
Female (n=324)	41.9 (36.6 to 47.3)	13.5 (10.0 to 18.0)
Age≥45 years (n=200)	63.9 (57.0 to 70.4)	23.5 (17.7 to 30.5)	0.006	Strong age effect
Low SES (n=186)	51.6 (44.2 to 59.0)	21.0 (15.5 to 27.8)	0.001	SES gradient
Corticosteroid users (n=344)	50.0 (44.7 to 55.2)	17.3 (13.5 to 21.8)	0.004	Drug-related risk

χ² for trend across disease duration=14.8, p<0.001.

^*Highly statistically significant result.

NS, not significant; SES, socioeconomic status.

When analysed by demographic factors, males showed a slightly higher comorbidity prevalence compared with females (47.5% vs 41.9%), although this difference did not reach statistical significance (p=0.24). Older age (≥ 45 years) was strongly associated with the presence of comorbidities (63.9%, p=0.006) and multimorbidity (23.5%). Similarly, patients of low SES demonstrated higher rates of any comorbidity (51.6%, p=0.001) and multimorbidity (21.0%), indicating a clear socioeconomic gradient.

Corticosteroid users exhibited a significantly greater prevalence of comorbidities (50.0%, p=0.004) compared with non-users, suggesting that prolonged corticosteroid exposure contributes to increased comorbidity burden. Collectively, these findings demonstrate that comorbidity frequency and clustering in SLE patients are influenced by disease duration, age, SES and corticosteroid use.

Table 4 summarises the multivariable Poisson regression, identifying independent predictors of comorbidity burden in patients with SLE. Increasing age and longer disease duration were the strongest predictors of comorbidity, with each 10-year increase in age associated with a 31% higher prevalence (aPR 1.31, 95% CI 1.19 to 1.45, p=0.002) and each 5-year increase in disease duration conferring a 22% rise (aPR 1.22, 95% CI 1.13 to 1.32, p<0.001). BMI, smoking and corticosteroid use were also significant contributors, with corticosteroid therapy showing a medium effect size (Cohen’s d=0.46). Conversely, HCQ use was associated with a 22% lower prevalence of comorbidities (aPR 0.78, 95% CI 0.64 to 0.95, p=0.014), suggesting a protective effect. Low SES independently increased comorbidity risk by 38% (aPR 1.38, 95% CI 1.15 to 1.65, p=0.005), emphasising the influence of social determinants of health. Female sex and immunosuppressant use were not significantly associated with comorbidity burden after adjustment.

Table 4. Predictors of comorbidity burden in patients with systemic lupus erythematosus (SLE).

Variable	Adjusted prevalence ratio (aPR)	95% CI	P value	β (estimate)	SE	Cohen’s d	Interpretation
Age (per 10 years)	1.31	1.19 to 1.45	0.002	0.27	0.05	0.85	Strong positive association
Female sex	0.95	0.77 to 1.18	0.68	—	—	—	No significant effect
Disease duration (per 5 years)	1.22	1.13 to 1.32	<0.001	0.19	0.04	0.72	Moderate association
BMI (per 5 kg/m²)	1.07	1.02 to 1.13	0.008	—	—	—	Weak positive association
Current smoker	1.26	1.03 to 1.54	0.03	—	—	—	Positive association
Corticosteroid use	1.33	1.10 to 1.61	0.003	0.14	0.05	0.46	Medium effect
Hydroxychloroquine use	0.78	0.64 to 0.95	0.014	–0.11	0.04	–0.35	Small protective effect
Low socioeconomic status	1.38	1.15 to 1.65	0.005	0.16	0.06	0.44	Medium effect, risk factor
Immunosuppressant use	1.09	0.89 to 1.33	0.31	—	—	—	Not significant

β represents standardised regression coefficients; Cohen’s d values interpreted as small (0.2–0.49), medium (0.5–0.79) and large (≥0.8). Variables with p<0.05 were considered statistically significant. Multivariable Poisson regression model adjusted for age, sex, BMI, disease duration, smoking, corticosteroid use, hydroxychloroquine use, immunosuppressant use and socioeconomic status. Variables with p<0.05 were considered statistically significant. Missing β or SE values indicate that the variable was not retained in the final model.

BMI, body mass index.

As shown in online supplemental table S1, the number of comorbidities correlated moderately with age (r=0.49) and disease duration (r=0.52), indicating that older patients and those with longer disease courses had greater comorbidity burden. Corticosteroid dose was also positively correlated (r=0.33*), while SES showed a negative correlation (r=–0.38*), suggesting higher comorbidity frequency among lower SES patients.

Clinically, patients with comorbidities exhibited significantly greater disease activity and damage. Mean SLEDAI-2K scores were higher (6.9±4.8 vs 5.6±4.1, p=0.009), and SLICC/ACR-DI scores nearly doubled (2.4±1.2 vs 1.3±0.9, p<0.001). Inflammatory markers such as ESR and CRP were elevated, while complement C3 levels were reduced (p≤0.004). Although anti-dsDNA levels trended higher, the difference was not significant (p=0.08).

Overall, greater comorbidity burden in SLE was associated with advancing age, longer disease duration, lower SES and increased disease activity.

Discussion

In this multicentre cohort of 482 patients with SLE, comorbidities were highly prevalent and strongly associated with age, disease duration, corticosteroid exposure and socioeconomic disadvantage. Our findings are consistent with recent international studies and provide novel insight into comorbidity patterns in a South Asian context, particularly in Pakistan, where socioeconomic and healthcare access disparities remain major determinants of disease outcomes.

Comorbidities increased significantly with advancing age and longer disease duration. Older patients (≥45 years) exhibited markedly higher rates of CVD, CKD and osteoporosis.

In the general Pakistani population, the prevalence of chronic comorbidities also increases with age, but at a slower and later trajectory than observed in our SLE cohort. Data from the Pakistan National Health Survey (PNHS) and the WHO STEPwise Approach to Noncommunicable Disease Surveillance indicate that hypertension prevalence rises from approximately 18%–20% in adults aged <40 years to over 45% among those aged ≥45 years, while diabetes prevalence nearly doubles across the same age range.^{25 26} In a large South Asian urban cohort, the prevalence of multimorbidity (≥2 chronic conditions) was about 10% in adults aged <60 years and increased to approximately 37% in those aged ≥60 years.^{27 28} Recent population-based data from Pakistan indicate that cardiometabolic multimorbidity (≥2 chronic conditions) affects approximately 15% of adults, with prevalence rising sharply in older age groups.²⁹ Our SLE patients developed cardiovascular, renal and bone comorbidities much earlier in adulthood and at substantially higher frequencies, suggesting that autoimmune inflammation, corticosteroid exposure and disease chronicity accelerate comorbidity accumulation beyond the effects of normal ageing. This pattern aligns with multiple large SLE cohorts, including LUpus in MInorities: NAture versus nurture (LUMINA) studies, which demonstrated age as a consistent determinant of damage accrual and multimorbidity.^{30 31} Longer disease duration contributes to cumulative inflammation, prolonged medication exposure and end-organ stress, all pathways linked to higher comorbidity risk.³²

Lower SES emerged as a significant independent predictor of comorbidity burden (aPR=1.38; p=0.005). This finding reflects how social determinants of education, income, rural residence and healthcare access shape disease control and long-term outcomes. Previous study from the USA reported similar findings, with low SES linked to greater organ damage and mortality.³³ Our data extend these observations to a low and middle-income country context, suggesting that limited access to continuous care, medication costs and late presentation amplify comorbidity risk among Pakistani SLE patients.

Corticosteroid use was widespread (71.4%), and higher doses were significantly associated with the presence of ≥1 comorbidity. Patients with comorbidities had a mean daily dose of 9.2 mg/day and more frequent long-term use (≥6 months). These findings mirror global concerns about glucocorticoid toxicity. In a recent real-world analysis of over 3000 SLE patients, it was reported that even moderate daily prednisone doses (>5 mg/day) were associated with higher odds of CVD, osteoporosis and diabetes.³⁴ Similarly, a comprehensive review highlighted the persistent challenge of steroid dependency and its role in long-term organ damage despite modern immunosuppressive therapies.³⁵

HCQ use (87.1%) was associated with a lower prevalence of comorbidities (aPR=0.78; p=0.014). This protective association is biologically plausible and supported by recent epidemiologic studies. Grimaldi et al demonstrated that current HCQ exposure was linked to a 37% reduction in cardiovascular events in SLE (OR=0.63; 95% CI 0.49 to 0.80).³⁶ Similarly, another previous study found that HCQ reduced thromboembolic events and improved lipid and glycaemic profiles. The mechanism likely involves inhibition of platelet aggregation, lipid modulation and endothelial protection, reinforcing HCQ’s role beyond immunomodulation as a cardiometabolic protector.³⁷

Immunosuppressant use (58.1%), particularly azathioprine and cyclophosphamide, was more common among patients with comorbidities. While this may reflect treatment for more severe disease, it also raises concern for drug-related toxicities (eg, infections, cytopenias). Similar findings were noted in the Cohort, where exposure to multiple immunosuppressants correlated with increased infection-related hospitalisations.³⁸ Biologic therapy (belimumab/rituximab) was used in only 7.9% of patients, slightly higher among those with comorbidities, consistent with selective use in refractory or complicated disease.

Patients with comorbidities had significantly higher disease activity (mean SLEDAI-2K=6.9 ± 4.8) and greater cumulative organ damage (SLICC/ACR-DI=2.4 ± 1.2) compared with those without comorbidities. Elevated ESR and CRP levels and lower complement C3 further support an association between persistent inflammation and comorbidity accrual. Chronic low-grade inflammation contributes to vascular injury, endothelial dysfunction and accelerated atherosclerosis, as shown in previous study.³⁹ Thus, comorbidities in SLE are both consequences and amplifiers of systemic inflammation.

In this study, the correlation matrix underscores the interconnectedness of age, disease duration, corticosteroid dose and SES, all showing moderate correlations (r=0.38–0.52) with comorbidity count. These findings emphasise that comorbidities arise not from a single determinant but from a multifactorial interplay of biological and socioeconomic stressors.

Our findings emphasise the need for targeted interventions in high-risk SLE populations, particularly older patients, those with low SES and long-term corticosteroid users. Early introduction of steroid-sparing agents, continued HCQ therapy and proactive screening for cardiovascular and bone health may reduce cumulative comorbidity risk. In addition, addressing socioeconomic barriers and improving access to rheumatologic care are essential steps towards achieving equitable outcomes.

Limitation

This study has several limitations. Medication adherence was not assessed, which may have influenced the associations between treatments and comorbidity outcomes. The cross-sectional design prevents establishing causal relationships between disease duration, medication use and comorbidities. Additionally, as data were derived from major hospitals in Punjab, Pakistan, findings may not be generalisable to other regions or community-based settings. Residual confounding from unmeasured factors such as diet, physical activity and genetic predisposition may also have influenced results.

Conclusion

Comorbidities such as CVD, CKD and osteoporosis were common among patients with SLE and increased with longer disease duration. Factors like older age, corticosteroid use and low SES contributed to higher comorbidity risk, while HCQ use appeared protective. Early screening and comprehensive management are essential to improve long-term outcomes in SLE.

Data availability statement

Data are available upon reasonable request.