Introduction

Systemic lupus erythematosus (SLE) is a chronic and life-threatening autoimmune disease that can affect any organ, such as the skin, kidneys, joints, central nervous system and cardiovascular system.¹ Its global prevalence has been estimated at 9-241 per 100,000 person-years, with an incidence of 0.3-23.2 per 100,000 person-years, ranging from 2.4 to 7.4 in Europe to 7.9 in Central and South America and 8.1 in Asia.² It is more common in women, with a ratio of women to men of 7-15:1 for adults and 3-5:1 for pediatric patients.² However, in the latter group and in men, SLE appears to be much more aggressive, with a greater presence of clinical manifestations and higher severity.^3,4

In Colombia, between 2012 and 2016, 431,834 cases of SLE were identified, with an unadjusted prevalence of 91.9 cases per 100,000 inhabitants; in those over 18 years of age, the prevalence was 126.3 per 100,000 inhabitants.⁵ There is evidence of greater aggressiveness in the clinical presentation of SLE in men, who present with greater lung involvement, longer hospital stays and high readmission and lethality rates.⁶

In the pharmacological treatment of SLE, antimalarials, corticosteroids, immunosuppressants and biotech drugs stand out; these drugs are used to induce remission or reduce disease activity and prevent flare-ups and target organ damage.⁷ Management is individualized and must take into account the clinical manifestations, complications, severity and degree of disease activity.^7-9 However, these drugs are not innocuous and maybe associated with adverse reactions, such as increased susceptibility to infections, increased risk of cardiovasculardiseases, diabetes and cancer, and decreased bone density.^1,10

The existing information on the prescription patterns of drugs used in rheumatological diseases is scarce in Colombia, and an understanding of these patterns can help direct informed interventions that focus on the appropriate use of these drugs. Such prescription patterns vary among countries and may also differ among their different regions; therefore, we sought to determine the pharmacological management of patients with SLE and to identify the differences between women and men and among age groups and geographical regions in Colombia.

Materials and methods

This was a cross-sectional study on the prescription patterns of drugs used for patients diagnosed with SLE; the data were obtained from a population database for drug dispensing that collects information from approximately 8.5 million people affiliated with the Colombian Health System in six health insurance companies, corresponding to approximately 30.0% of the active population covered by the contributory or paid regime and 6.0% covered by the state-subsidized regime, which together comprise 17.3% of the Colombian population.

Patients were identified using International Classification of Diseases (ICD-10) codes related to SLE (M321, M328, M329) in the period from July 1,2019, to June 30,2020. Patients of any age and sex who were seen as outpatients and were under pharmacological management for SLE were selected. Patients with a concomitant diagnosis of cutaneous lupus erythematosus or drug-induced lupus and those diagnosed with SLE without pharmacological treatment were excluded.

Based on the information on the prescription of drugs to the affiliated population, systematically obtained by a dispensing company (Audifarma SA), a database was designed that allowed the collection of the following groups of patient variables:

1. Sociodemographic: sex, age (groups: <20 years, 20-39 years, 40-64 years and >65 years), health system regime affiliation (contributory or subsidized) and city of dispensation;

2. Drugs for SLE management:

3. Comedications (grouped into the following categories): (a) antidiabetics (oral and subcutaneous), (b) antihy-pertensives and diuretics, (c) lipid-lowering drugs; (d) antiulcers, (e) antidepressants, (f) anxiolytics and hypnotics (benzodiazepines and Z-drugs), (g) thyroid hormone, (h) antipsychotics (typical and atypical), (i) antiepileptics, (j) antiarrhythmics, (k) antihistamines, (l) antidementia drugs, (m) opioid analgesics, (n) nonopioid analgesics, (o) bronchodilators and inhaled corticosteroids, (p) antiplatelet agents, (q) anticoagulants (oral and parenteral), (r) antipsychotics, and (s) hormonal contraceptives.

4. Comorbidities: The main cardiovascular, endocrine, rheumatic, urological, kidney, psychiatric, neurological, digestive, respiratory and neoplastic diseases were identified from the reported ICD-10 diagnostic codes. Autoimmune rheumatological diseases (rheumatoid arthritis, Sjogren's syndrome, vasculitis, polymyalgia rheumatica, psoriatic arthritis, ankylosing spondylitis and systemic sclerosis) and non-immune diseases (fibromyalgia, osteoatrosis, osteoporosis and gout) were included.

Ethical considerations

The protocol was approved by the Bioethics Committee of Universidad Tecnológica de Pereira in the "research without risk" category (approval number: 02-051020). The principles established by the Declaration of Helsinki were respected.

The data were analyzed with the statistical package SPSS Statistics, version 26.0 for Windows (IBM, USA). A descriptive analysis was performed; qualitative variables are presented as frequencies and proportions, and quantitative variables are presented as measures of central tendency and dispersion, depending on the normality of the data, as established by the Kolmogorov-Smirnov test. Quantitative variables were compared using Student's t-test or the Mann-Whitney U test, and categorical variables were compared using the X ² test or Fisher's exact test. A statistical significance level of p < 0.05 was adopted.

Results

A total of 4307 patients with a diagnosis of SLE were identified, distributed in 135 different cities or municipalities. Of these, 89.4% (n = 3851) were women, and the median age was 44.2 years (interquartile range: 32.4-55.8 years; range:6.7-90.8 years); the age group distributions were as follows: <20 years (n = 198; 4.6%), 20-39 years (n = 1560; 36.2%), 40-64 years (n = 2120; 49.2%) and ≥65 years (n = 429; 10.0%). Most patients lived in the Bogotá-Cundinamarca Region (n = 1491; 34.6%), followed by the Caribbean Region (n = 1036; 24.1%), Central Region (n = 878; 20.4%), Pacific Region (n = 753; 17.5%), and Eastern Region and Amazonia (n = 149; 3.5%). A total of 91.2% (n =3926) of the patients participated in the contributory regime, and 8.8% (n =381) belonged to the subsidized regime.

Most patients with SLE were treated with DMARDs (n =3901; 90.5%), with apredominance of chloroquine (n =2343; 54.4%), azathioprine (n = 1579; 36.6%), hydroxychloroquine (n =1267; 29.4%) and methotrexate (n =758; 17.6%). Corticosteroids, especially prednisolone (n = 2915; 67.6%), were prescribed to 76.1% (n =3278) of patients; among immuno-suppressants, mycophenolate was the most prescribed (n = 719; 16.7%). The cyclophosphamide was prescribed to 36 patients (0.8%) while DMARDb were prescribed to 11 patients (0.3%).

A total of 71.4% (n= 3076) of all patients had some chronic disease. Of these, 67.8% (n = 2084) had one or two diseases, 23.6% (n=726) had three or four diseases, and 8.6% (n = 266) had five or more diseases. The 10 most common comorbidities were hypertension (n = 2012; 46.7%), rheumatoid arthritis (n = 532; 12.4%), hypothyroidism (n = 513; 11.9%), Sjögren's syndrome (n =489; 11.4%); diabetes (n =383; 8.9%); chronic kidney disease (n =277; 6.4%); depressive disorders (n =226; 5.2%); chronic pain (n = 207; 4.8%); anxiety disorders (n = 179; 4.2%) and fibromyalgia (n = 157; 3.6%). Polyautoimmunity (systemic lupus erythematosus and rheumatoid arthritis and Sjögren's syndrome) occurred in 21.6% (n =932) of patients. In addition, of all patients, 2.7% (n =115) had a diagnosis related to thrombosis. A total of 18.4% (n =792) of the patients had an infection-related diagnosis, including urinary tract infection (n =189; 4.4%), tuberculosis (n =47; 1.1%), pneumonia (n =39; 0.9%), meningitis (n =7; 0.2%) and sepsis (n=7; 0.2%).

The most common comedications were nonopioid analgesics (n = 2854; 66.3%), followed by antiulcers (n = 2391; 55.5%), antihypertensives and diuretics (n = 2234; 51.9%), antiplatelet agents (n=1210; 28.1%), lipid-lowering drugs (n = 1078; 25.0); antidepressants (n =1017; 23.5%), antihistamines (n =1002; 23.3%), opioid analgesics (n =949; 22.0%), thyroid hormone (n =922; 21.4%), antiepileptics (n =621; 14.4%), anticoagulants (n = 525; 12.2%), bronchodilators and inhaled corticosteroids (n =382; 8.9%), antidiabetics (n =262; 6.1%)), antipsychotics (n =68; 3.9%) and hormonal contraceptives (n =97; 2.3%).

Comparison among geographical regions

The median age was higher in the Pacific Region than in the other regions. The proportion of patients with chronic comorbidities was lowest in the Caribbean Region, and cardiovascular diseases predominated in all regions. Patients in the Pacific Region had fewer prescriptions for corti-costeroids and mycophenolate but more prescriptions for chloroquine, azathioprine and methotrexate than did the other regions. Deflazacort was widely used in the Central Region. The use of anticoagulants was predominant in the Bogotá-Cundinamarca Region (Table 2).

Table 2 Comparison of sociodemographic, clinical and pharmacological variables according to geographic region for 4307 patients diagnosed with systemic lupus erythematosus in Colombia, 2019-2020. 

IQR: Interquartile range. DMARDs: synthetic disease-modifying antirheumatic drugs. DMARDb: biological disease-modifying antirheumatic drugs.

Discussion

The prescription patterns of drugs used for patients diagnosed with SLE of any age and sex in a group of Colombian patients were identified, with characterizations of the differences or similarities in treatment according to age group, sex, and geographical region. SLE occurred more frequently in women, consistent with what is widely documented in other studies,^11-15and the mean age (44.2 years) was in agreement with what was found in the United States (44.5-46.0 years)^12,16 Switzerland (44.8 years)¹³ and Jamaica (45.1 years)¹⁷ but higher than previously described in Canada (40.6 years)¹⁸ and Colombia (38.4 years).⁶

Overall, corticosteroids were prescribed to 76.1% of patients, a similar rate to that found in the United States (78.8%)¹⁶ but much more frequent than that reported for Canada (53.2%),¹⁸ Puerto Rico (50.9%)¹⁴ and Switzerland (48.0%).¹³ Among DMARDs, the use of chloroquine stood out, differing markedly from reports in other studies where the use of hydroxychloroquine was predominant (43.4-91.3%).^11,13,16,17 The probable explanation for this result is that in Colombia, hydroxychloroquine is not included in the health benefits plan and has a higher cost compared to chloroquine, which can make access to it difficult.¹⁹ However, chloroquine has been associated with greater adverse drug reactions and a worse safety profile, especially regarding the risk of retinopathy.²⁰ It was found that mycophenolate was prescribed more frequently in Colombia than in countries such as Korea (2.2%)²¹ and the United States (3.3%),¹¹ whereas DMARDb were rarely used, similar to what has been reported in the literature.^11,21 In general, our hypothesis is that the differences and similarities found could be explained by the degree of activity and severity of SLE because in mild and moderate cases, antimalarials and corticosteroids are the recommended drugs, whereas when the disease progresses or is refractory, higher doses of corticosteroids, immunosuppressants or DMARDb are necessary^7-9; however, we do not know the degree of disease activity in this cohort.

In addition to the above, cyclophosphamide that is indicated in patients with lupus nephritis²² has been prescribed in 10.5% of patients in a Colombian cohort⁶ and in 29.2% of patients in a Latin American cohort,²³ in marked contrast to the minimum proportion of patients who had it prescribed in this study. This is probably because cyclophosphamide when administered intravenously will require it to be performed in the hospital, and as the drug dispensing information used in this research only involves outpatient medications, it is highly likely that the patients who received this medication have not been fully identify them.

The prescription pattern for these drugs showed important differences when comparing women and men. There is limited information available in this regard; however, a study by Santamaría-Alza et al. that included 200 patients with SLE found significant differences in the use of cyclophosphamide between women and men (8.3% vs. 22.6%, respectively, p = 0.017),⁶ a result that was not found in this study. In addition, they found no differences in the proportion of use of other drugs used to manage SLE,⁶ which contrasts with the results found in the present study, i.e., corticosteroid, chloroquine, azathioprine and mycophenolate prescriptions were predominant for men, whereas azathioprine and hydroxychloroquine prescription were predominant for women. These findings are related to the clinical differences and the frequency of complications between women and men; renal disease, serositis and thrombocytopenia are predominant in the latter.²⁴ In the United States, Pelletier et al. compared patients with and without lupus nephritis and found that the use of corticosteroids, immunosuppressants and antimalarials was more frequent in patients with kidney disease.²⁵

Different clinical pictures of SLE have been observed in children and adults. In children, the disease is more active, with a much more aggressive progression and with more complications, which affects the type of treatment used.^3,15,26 In this analysis, prescription patterns were investigated according to age group, showing greater use of chloroquine, immunosup-pressants and corticosteroids in patients under 20 years of age. A study conducted in Canada compared pediatric and adult patients with SLE and reported similar findings; i.e., corticosteroids (97.0% vs. 70.0%; p < 0.0001) and immunosuppressants (66.0% vs. 37.0%; p = 0.0001) were prescribed more to children, whereas methotrexate was prescribed more to adults (31.0% vs. 9.0%; p = 0.009).¹⁵ In Hungary, it was found that mycophe-nolate was prescribed much more frequently to children than to adults (15.2% vs. 5.3%; p = 0.0056),²⁶ results that are similar to those for the United States (28.1 vs. 13.0; p<0.001)²⁷ and in agreement with our results.

The drugs used for the management of SLE were prescribed differently among the different geographical regions of Colombia. This pattern had already been evidenced in another pharmacoepidemiological study that compared the prescription of ambulatory antibiotics in the different regions of the country²⁸ and is probably due to the prescribing habits of physicians, to the academic training of physicians, to the variability in the availability of drugs, to the influence of the pharmaceutical industry and to the health system affiliation regime. Thus, for example, in Argentina, cyclophosphamide was administered more frequently in the public health system than in the private system, while the use of corticosteroids, antimalarials and immunosuppressants was similar between the two systems.²⁹

Finally, it is important to highlight that, during the study period, the first months of the mandatory preventive isolation (confinement) caused by the pandemic of Coronavirus Disease 2019 (COVID-19) in Colombia were included, without affecting the dispensing of drugs in patients with systemic lupus erythematosus present in the study.

Some limitations in the interpretation of the results are recognized because access to medical records was not obtained to verify the patients' diagnoses and their hospitalizations and the activity, severity and complications of the disease or comorbidity such as antiphospholipid antibody syndrome. In addition, the drugs prescribed outside the health system or that were not delivered by the dispensing company and the drug induction cycles that the patients may have received are unknown. However, the sample included many patients distributed throughout most of the Colombian territory and both the contributory and subsidized regimes.

Conclusions

Given the above findings, it can be concluded that there are differences in the prescription of drugs used for patients with SLE according to age group, sex and geographical regions of the country. Patients with SLE in Colombia are treated with DMARDs, especially chloroquine and azathioprine, with corticosteroids, particularly prednisolone, and with immunosuppressants such as mycophenolate, and infrequently receive DMARDb. In addition, they frequently have cardiovascular, rheumatologic and endocrine comorbidities and, in addition to drugs for the management of SLE, are prescribed nonopioid analgesics, antiulcers and antihyperten-sives. Knowledge of the differences in management found can be useful to guide treating physicians and health plan administrators to manage the resources necessary to meet the needs of this important group of patients.