Functional Disability and Its Determinants in Ecuadorian Patients with Rheumatoid Arthritis

Introduction

Rheumatoid arthritis (RA) is an autoimmune disease with a prevalence that ranges from 0.5% to 1.0%.¹ This disease is the result of a chronic inflammatory state that progresses and in advanced stages causes severe limitation of the patient. Although there have been many advances in the treatment of RA, functional disability remains a serious problem for these patients. Sokka et al² established that patients with RA have a risk of disability seven times greater than the general population. Factors that are related to a higher disability rate in patients with RA include advanced age, female sex, higher disease activity indices and higher inflammatory markers.³

Disability leads to several repercussions such as loss of workdays and greater use of health resources. It has been shown that up to 70% of patients with RA develop impairment at work after 10 years of disease progression.⁴ Huscher et al⁵ estimated that the annual direct medical costs for patients with RA with disabilities are three times higher than those for patients without disability; also, annual indirect costs that include days lost at work and disability pensions are significantly higher in patients with disability. Likewise, disability affects personal relationships and lifestyles of patients and their families and is one of the most important predictors of depression and quality of life. Ji et al⁶ demonstrated a significant association between functional disability in RA and depression. Another study showed that functional disability and limitations in daily activities in RA have a negative impact on quality of life.⁷

Several authors have studied the prevalence of disability in RA and its determinants, among them, Krause et al⁸ in the United States, Sokka et al² in Finland and Öken et al⁹ in Turkey. Regarding Latin America, Cordeiro et al¹⁰ described the prevalence of disability in RA and associated factors in Brazil while Schneeberger et al³ in Argentina and Durán et al¹¹ in Chile. Studies in Ecuador regarding AR are limited and there are no studies having a disability as the main outcome or describing its determinants in Ecuadorian patients.

Due to the regional differences in the severity of rheumatic diseases and the available treatments, and the impact of disability in patients with RA, the aim of this study was to study the prevalence of disability and its determinants in Ecuadorian population. By doing this, physicians might be able to recognize disability and its determinants in the clinical practice and intervene in them early to reduce the burden of the disease on functional capacity.

Materials and Methods

A cross-sectional study was carried out to determine the prevalence of disability and its determinants in RA during a period of 6 months: January 2019–June 2019. The study was carried out at a Rheumatology Center in the city of Guayaquil, Ecuador.

This study was in accordance with the Declaration of Helsinki. The patients signed a written informed consent, which reflects the approval of the clinical study according to article 361 of the Political Constitution of the Republic of Ecuador, article 7 of the Organic Health Law, and article 15–16 of the Code of Medical Ethics of Ecuador. The study was approved by the Ethics and Teaching Committee of the Centro de Reumatología y Rehabilitación (CERER) with a registration number: No.005/2019; Folio 01: Book of Acts No.1. Patient anonymity of data was respected throughout the entire investigative process, and it was explained to the participants that the data collected would be used solely for the study.

The population consisted of patients with a pre-established diagnosis of RA according to the criteria of the American College of Rheumatology (ACR) 1987 which includes morning stiffness for at least 1 hour, arthritis of three or more joints, arthritis of hand joints, symmetric distribution of the disease, rheumatoid nodules, being seropositive for rheumatoid factor and radiographic changes compatible with RA such as erosions and periarticular osteopenia.¹² Patients with RA that showed to the clinic either for a first consult or a follow-up were considered for the study, which represents a convenience sampling as patients were selected from the center due to its easy accessibility. Only those that had at least four ACR criteria out of the seven possible for more than 6 weeks were 18 years or older and gave written consent to participate in the study, were included. Also, as the surveys were in Spanish, only Spanish speakers were included. Patients with other connective tissue diseases or autoimmune diseases were excluded, as well as patients with mental disabilities or physical limitations that prevented them from understanding, communicating or completing the questionnaires. Patients were asked for their authorization prior to their enrollment after an adequate explanation of the study, its purpose and their roles as participants.

Data collection was based on surveys and questionnaires. Functional disability was measured using the validated Spanish version of the Health assessment questionnaire disability index (HAQ-DI).¹³ This questionnaire has proved to be a good predictor of mortality, joint replacement, work disability and economic losses, in addition to being sensitive to changes in the patient’s condition over time.¹⁴ All the questionnaires were in the official language of the patient (Spanish) to facilitate understanding. On the day the patient was seen in the consult, the following five activities were carried out: filling out the basic information survey, filling out the HAQ-DI questionnaire, assessment by the rheumatologist, checking the laboratories in the medical record and calculation of the disease activity score DAS-28:

1. The basic information survey was filled out by a previously trained investigator and included:
   1. Demographics: age, sex, race, marital status, work, smoking.
   2. Characteristics of the disease: years with the disease, type of articular involvement, visual analogue scale (VAS) of pain (0–10) and presence of extra-articular symptoms (Raynaud’s phenomenon, fever, fatigue, myalgia, weight loss, xerophthalmia, xerostomia).
   3. Comorbidities: hypertension, diabetes mellitus, dyslipidemia, thyroid disease, depression, gastrointestinal disease. This was determined by asking the patient if he/she had been previously diagnosed or receives treatment for any of these diseases.
   4. Current treatment: non-steroidal anti-inflammatory drugs, corticosteroids, hydroxychloroquine, methotrexate, sulfasalazine, leflunomide, anti-TNF drugs and other biologics.
2. The HAQ-DI questionnaire was filled out by the patient after an explanation of it. In this case, a researcher did not participate in order to avoid bias. Disability was defined as a HAQ-DI score greater than 1.25.
3. A rheumatologist assessed the patients to determine the number of swollen joints (SJC) and the number of tender joints (TJC) according to the 28 joint count for arthritis.
4. Medical records were accessed to obtain the most recent values for erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) and rheumatoid factor (RF).
5. Using the information about tender joint count, swollen joint count, ESR and patient global health, the DAS28-ESR was calculated online at: https://www.rheumakit.com/en/calculators/das28. The patient was considered in remission if the score was less than 2.6, with a mild activity if it was 2.6 to 3.2, moderate activity 3.2 to 5.1 and high activity if greater than 5.1.

Data were analyzed using the statistical program SPSS v22. Categorical variables were described with percentages. Continuous variables were described with mean and standard deviation. The Kolmogorov–Smirnov test was used to examine for normality of distribution of continuous variables. We compared characteristics between patients with and without disability using Student’s t-test (independent samples, two-tailed) for continuous variables and chi-square for categorical variables. With the variables that were significant predictors of disability in the univariate analysis, we conducted a multiple logistic regression model for functional disability. The a-priori level used was 0.05 for all analyses.

Results

We included 395 patients, 87.8% female and 12.2% male with a mean age of 51.4±12 years. The mean duration of the disease was 13.8±7 years and 72.4% had a positive rheumatoid factor. Table 1 shows the demographics and clinical characteristics of the population.

Table 1 Characteristics of the Population

The mean DAS-28 score was 3.4±1.5; 33.7% of patients were in remission, 17.0% had low activity, 34.1% moderate activity and 15.2% high activity. The mean for TJC was 5±6 and for SJC 4±5 joints, with a mean VAS of pain of 4±3. The mean ESR was 31.9±8.9 mm/h and CRP 18.2±8.1 mg/L. The mean HAQ-DI was 0.8±0.9, with a prevalence of disability of 26.6%. The disability rate in women was 28.5% and in men 12.5%. When stratifying patients according to their age, the disability rate in those <65 years of age was 26% and in those ≥65 years of age 30.4%.

Table 2 shows the comparison between patients with and without disabilities. We found a statistically significant relationship between disability and female sex (p=0.018), age (p=0.020), presence of extra-articular manifestations (p=0.008), myalgias (p<0.001), xerostomia (p=0.023), fever (p<0.001), fatigue (p<0.001), presence of comorbidities (p<0.001), number of comorbidities (p<0.001), depression (p<0.001), hypertension (p=0.001), diabetes (p=0.030) and gastric disease (p<0.001). As for markers of disease activity, functional disability was associated with ESR (p<0.001), CRP (p<0.001), TJC (p<0.001), SJC (p<0.001), VAS of pain (p<0.001) and DAS-28 (p<0.001). In addition, patients with disability had a lower employment rate (26.7%) compared to those without disability (45.5%, p=0.001). Likewise, patients with disability reported greater sexual impairment (19%) compared to patients without disability (10.3%, p=0.022). No statistically significant associations were found between disability and duration of the disease (p=0.058), RF positivity (p=0.441), smoking (p=0.463), Raynaud phenomenon (p=0.728), xerophthalmia (p=0.069), weight loss (p=0.149), hypothyroidism (p=0.827), dyslipidemia (p=0.355) or any treatment.

Table 2 Comparison Between Patients with and Without Disability

In the multivariate logistic regression analysis, seen in Table 3, only depression (p=0.029), diabetes (p=0.003), SJC (p=0.001) and VAS of pain (p=0.004) were significantly related to functional disability. Using the Hosmer–Lemeshow test, this model had a good fit (χ2 = 11.160, P = 0.193). The Nagelkerke’s R² suggested that the model explained 56% of the variation in the outcome.

Table 3 Multivariate Logistic Regression Analyses Predictive of Disability

Discussion

We found a prevalence of disability of 26.6%, similar to that found in other studies by Krause et al⁸ (28%) and Myasoedova et al¹⁵ (26%).

Among the predictors of disability, female gender and age have been described in many studies,¹⁶ which is consistent with our findings. This may be because women usually have a more aggressive disease and higher disease activity.¹⁷ In addition, aging is related to many anatomical and functional changes that can contribute to disability.

We did not find an association between disability and duration of the disease; however, in a study by Wolfe,¹⁸ it was seen that the HAQ-DI score increased at a rate of 0.020 units per year and that in 53% of patients, the grade of disability worsened with time, which can be related to the progression of the disease. In this aspect, we need to consider the effect of treatment in controlling the disease and limiting disability. Ward et al¹⁹ showed that the rate of progression of disability in patients treated regularly was 0.008 disability index units per year compared to 0.020 disability index units per year in those treated intermittently. Nowadays, biologics have shown to decrease the rate of disability by 1–3% per year, according to a study by Krishnan et al.²⁰ Other studies have also found that treatment with biologic agents increases employability rates and decreases sick leaves.^21,22 This is related to the control of the inflammation and joint damage that is achieved with biologic DMARDs. In this study, there was no significant difference in the rate of biological therapy between patients with and without disability, which could be related to the low overall use of biologics. In our study, only 9.1% of the patients used biologics which is significantly lower than the rate of biological therapy for RA in the United States (26%) described by Yazici et al.²³

In this study, patients with disabilities had higher painful joints count and greater VAS of pain. Wolfe¹⁸ showed that a change of 0.5 units in the level of pain is associated with changes of 0.25 units in the HAQ-DI. One possible explanation for this relationship is that chronic pain can interfere with daily activities. Also, another study²⁴ found that pain is not only related to disability but also to greater depression and fatigue which in turn can contribute to disability. In addition, Jeong et al²⁵ found a greater impact on functional capacity in patients who have a compromise of the joints of the foot or ankle, which can produce changes in gait and the need for orthotic devices.

The presence of comorbidities was another determinant of disability. Patients with disabilities had a greater number of comorbidities compared to those without a disability. Similar results were found in other studies.^26,27 Likewise, Radner et al²⁶ reported a lower HAQ-DI in patients without comorbidities, which increased according to the number of comorbidities. In the study by Michaud et al,²⁷ the comorbidities with the greatest influence on HAQ-DI were cardiac, pulmonary and psychiatric diseases. In this study, depression was significantly related to disability, as well as hypertension, diabetes and gastric disease but the only one that showed significance in the logistic regression analysis was depression; thus, highlighting the importance of addressing depression in patients with RA.

The prevalence of depression in our study was high (51%), consistent with other studies that have shown depression rates of 40%²⁸ and 43%.²⁹ The impact that depression has on the disease is important since it is related to greater disability, worse quality of life and higher mortality.^29,30 Even in one study, it was found that patients with depression despite being in clinical remission, had scores of quality of life and disability similar to patients with high disease activity.³¹ This relationship between depression and disability in RA has been shown to be bidirectional. A study found that a 10% reduction in the ability to perform activities that an individual considers important, such as visiting the family or going on vacation, is followed by a sevenfold increase in depression.³² On the other hand, changes of two to three units in the depression questionnaires generate significant changes in the HAQ-DI.¹⁸ This association could be due to the fact that depression affects the threshold and the way of coping with pain and pain is also a determinant of disability.

This study also showed that patients with disabilities had a higher DAS-28, SJC, ESR and CRP. Regarding the association between disability and disease activity scores, Karpouzas et al¹⁶ found that patients with high disease activity had significantly higher HAQ-DI scores than those with low disease activity. In the study by Hakkinen et al,³³ it was shown that the number of swollen joints in the lower extremities was related to the sub-dimensions of the HAQ-DI that involved walking and common activities of daily living that need weight support, while the number of swollen joints of the upper extremities influenced the sub-dimensions that require reach or grip. Also, Wolfe¹⁸ established that changes in the ESR generate significant changes in the HAQ-DI score.

Other factors that have been associated with disability are global patient evaluation, morning stiffness, grip strength and seropositivity of the rheumatoid factor.³⁴ None of these findings were seen in this study.

The main limitation of this study is that it is cross-sectional so no causal relationships can be established between disability and its determinants. Also, we did not address radiographic damage or joint mobility which are related to disability. Another factor to consider is that sampling was done by convenience and not randomly, which means that our sample could differ from the whole population of patients with RA in Ecuador. This is important as there might be differences in the HAQ-DI score based on the mean age and the access to treatment of the patients included. However, when comparing our findings with a study by Ríos et al³⁵ about the clinical and serological characteristics of Ecuadorian patients with RA, the description of the samples are similar. Moreover, one of the strengths of the study is that it was done with a large number of patients which might mean that the results can be generalized to the Ecuadorian population.

Conclusion

Disability affects a quarter of patients with RA. Older patients and women are more affected. Pain is one of the main determinants of disability, as well as inflammatory markers and the level of disease activity. Patients with extra-articular manifestations and comorbidities also had worse functional capacity. For this reason, the diagnosis and implementation of an early treatment in patients with RA is essential to reduce the impact of the disease on functional capacity. It should also be emphasized that depression is an important comorbidity that contributes to disability and worsens the quality of life of patients with RA, so interventions aimed at it could be an alternative to improve functional capacity and quality of life in patients with RA.