Background

Asthma is a chronic inflammatory disorder of the airways [1] and its prevalence is estimated to range from 0.2 to 21.0% globally [2]. It is one of the most common chronic diseases among children [3] and young adults [2], and it is a significant cause of disability, high health resource utilization, and poor quality of life for those who are affected [1]. Thus, it accounts for considerable healthcare costs and loss of work productivity [4, 5].

Approximately 2–10% have some form of severe asthma (SA) [6, 7]. According to ERS/ATS 2014 guidelines [8], SA is defined as asthma that requires step four or five treatment (e.g. high-dose Inhaled Corticosteroids (ICS)/Long-acting beta2-agonist (LABA)) to prevent it from becoming ‘uncontrolled’, or asthma that remains ‘uncontrolled’ despite this treatment. SA is a heterogeneous disease with high variability in clinical presentation, physiological characteristics, and disease manifestation [6, 9].

In Latin America, there is a lack of information related to SA epidemiology and the burden of SA following implementation of the ERS/ATS 2014 criteria. Essentially, most of the Latin American studies that have assessed the SA population in the region had also included untreated patients [10, 11] or focused on the, uncontrolled asthma population [12]. Therefore, by using the ERS/ATS 2014 criteria the aims of this study were to compare the demographic, socioeconomic, and clinical characteristics, as well as treatment courses and use of healthcare resources between SA patients and non-severe asthma (NSA) patients in Argentina, Colombia, Chile, and Mexico.

Methods

Study design and population

The Asthma Control in Latin America (ASLA) study was a cross-sectional multisite study. Patients were consecutively enrolled between December 2013 and December 2015. The 16 recruiting sites were located in Argentina (five sites); Chile (five sites), Colombia (three sites), and Mexico (three sites) [13, 14].

The study included 594 diagnosed and treated asthma patients who were recruited during routine care at specialized outpatient sites, aged over 12 years old, under pneumologists follow up, with at least one prescription of asthma medication and one medical visit for asthma within the last 6 months. Individuals were excluded if they were participating in a clinical trial at the time of the study or were unable or not willing to comply with the study requirements. This study was conducted in accordance with the amended Declaration of Helsinki [15]. Written informed consent was obtained from all the participants. Minors younger than 18 years old signed the assent form and parents or legal guardians provided their written informed consent. Institutional review boards in each country approved consent forms and procedures.

Measurements

During the course of a scheduled medical visit, the participating patients were asked to complete the Asthma Control Test (ACT) [16], which was a self-administered questionnaire containing five items that can each be rated on a five-point Likert scale. Based on the ACT score, subjects were categorized as controlled (ACT score ≥ 20) or uncontrolled (ACT score < 20). The physician conducted an interview, which included questions on socio-demographic data [gender, age, skin colour (white, African-Latin American, Native American, multiple and other skin colours that were not listed)]; healthcare utilization in the last year; treatment used; and comorbidities.

Asthma severity was operationally defined using an algorithm based on answers to questions relating to the ERS/ATS 2014 criteria. SA was defined as asthma that required step four or five treatment (i.e.: high ICS doses plus a second controller or use of OCS (oral corticosteroid) regardless of ICS doses). Four treatment categories did not fulfil the ERS/ATS 2014 criteria for SA and NSA and were reviewed by a panel of three pneumologist experts on asthma to allocate a group (Table 1). Clinically significant asthma exacerbations were based on patients self-reported healthcare resource utilization, and were defined as any emergency room (ER) visit or hospitalization due to asthma.

Table 1 Respirologist-adjudicated decisions about patients’ asthma classifications and treatments, for cases where ERS/ATS 2014 guidelines were unclear
Full size table

Data Analysis

A descriptive analysis was carried out, comparing socio-demographic characteristics, clinical factors, asthma control, nutritional status, healthcare utilization in the last year, treatment used, and comorbidities of SA patients compared with NSA patients.

Body Mass Index (BMI) was calculated in adults as weight divided by the square of height, and the nutritional status categorized as: underweight < 18.5 Kg/m2; eutrophic = 18.5–24.9 Kg/m2; overweight = 25–29.9 Kg/m2; and obesity ≥30Kg/m2. BMI in adolescents was calculated following the indications described elsewhere [17] and the nutritional status classified as underweight (Z-score < − 3 and < − 2); eutrophic (Z-score ≥ − 2 and ≤ 1); overweight (score-Z > 1 and ≤ 2); and obese (Z-score > 2). For demographic and clinical outcome variables, Chi-square test or the Fisher exact test (as appropriate) and Mann Whitney test were used to assess associations between SA patients and NSA patients. Values of p < 0.05 were considered statistically significant. Data analysis was performed using Stata™13 College Station,TX: StataCorpLP and SPSS™ version 24.

Results

A total of 594 asthma patients were studied: 154 subjects from Argentina, 154 from Chile, 163 from Mexico and 123 from Colombia. Overall 72.7% were women and 50.8% white, with a median age of 47 years old at the time of the study visit and an age of 25 at the time of asthma diagnosis. According to the ERS/ATS 2014 criteria, 31.0% of the patients had SA (Table 2). Among the SA patients, 88.0% were classified as SA by the use of high ICS dose plus a second controller (Table 4). Of the 12% remaining, about 11.4% used ICS in doses that were not high but were classified as SA due to the use of anti-IgE, tiotropium, anti-IgE and tiotropium at the same time or OCS. The last 0.6% used OCS without ICS in combination and due to that it was also considered as SA.

Differences in socio-demographic and clinical factors by SA and NSA are shown in Table 2. NSA patients were younger than the SA patients at study visit and at diagnosis. 72.1% of the SA patients were overweight or obese. A higher proportion of SA patients reported having performed at least one PEF test in the last year compared with NSA patients (35.9% vs. 16.6%, p < 0.001).

Table 2 Sociodemographic and clinical factors by severe asthma status in Argentina, Chile, Colombia, and Mexico 2013–2015
Full size table

Figure 1 shows that more than a half of SA (64.1%) and NSA (53.2%) patients were uncontrolled, and this proportion was higher in SA when compared to NSA (p = 0.013).

Fig. 1
figure 1

Proportions of controlled and uncontrolled asthma patients in Argentina, Chile, Colombia, and Mexico. 2013–2015

Full size image

Table 3 shows healthcare resources used and the number of clinically significant asthma exacerbation suffered during the last year by SA and NSA patients. Overall, SA patients reported a statistically significant two times higher proportion of both hospital admission (8.7% vs. 3.7%, p = 0.011) and almost two times higher proportion for ER visits due to asthma (37.0% vs. 21.7%, p < 0.001)) in the last year, compared with NSA patients.

Table 3 Asthma exacerbation by severe asthma status in Argentina, Chile, Colombia, and Mexico, 2013–2015
Full size table

Regarding the hospital admission and emergency visits due to other causes, SA patients also had a significant higher proportion of both events than NSA patients (6.5% vs 2.4%, p = 0.015 and 14.1% vs 7.3%, p = 0.009). The same pattern was observed when considering the mean number of clinically significant asthma exacerbations (mean 1.88 vs 0.56, p < 0.001).

When comparing use of current asthma medication by asthma severity classification (Table 4), the most common medication used in SA patients was ICS (99.5%), even in NSA patients (88.0%). Nevertheless, when comparing combination therapies (Table 5), the most prescribed was ICS/LABA, followed by ICS only in the overall population (49.0 and 17.3%, respectively). The most controller therapy used was ICS/LABA for both SA (47.8%) and NSA (49.5%) patients.

Table 4 Self-reported medication used by severe asthma status in Argentina, Chile, Colombia, and Mexico, 2013–2015
Full size table
Table 5 Self-reported combination therapies used by severe asthma status in Argentina, Chile, Colombia, and Mexico, 2013–2015
Full size table

Finally, when comparing SA with NSA patients regarding self-reported comorbidities (Table 6), the most reported among the SA patients were chronic rhinitis (52.7%) followed by hypertension/hypertension syndrome (32.1%) and gastroesophageal reflux disease (27.2%). On the other hand, among the NSA patients the most frequent reported comorbidities were chronic rhinitis (56.8%), followed by chronic sinusitis and/or rhinosinusitis (22.2%) and gastroesophageal reflux (17.1%). Nevertheless, hypertension, COPD, psychological disturbances (as depression and anxiety), gastroesophageal reflux disease and obesity were higher in SA and this difference was statistically significant.

Table 6 Patient-reported comorbidities by severe and non-severe asthma patients in Argentina, Chile, Colombia, and Mexico, 2013–2015
Full size table

Discussion

To our knowledge, there are no previous description of the SA population in Latin America countries following the updated 2014 ERS/ATS criteria. Results from the present cross-sectional study showed that overall prevalence of SA is 31.0% among patients from outpatient specialized sites in Argentina, Chile, Colombia and Mexico. We found that patients reflecting the ERS/ATS steps 4 and 5 tended to be diagnosed at a later age, had higher proportion of hypertension, COPD, psychological disturbances, gastroesophageal reflux disease, obesity, and experienced increased healthcare utilization.

In our study, we observed that SA patients were older at the time of study and at the time of asthma diagnosis when compared with NSA patients - which is in alignment with other studies [18, 19]. This might be explained by the generalized decline in lung function in older patients, leading to more severe patients [20]. However, our study is limited to reinforce this idea due to lack of a longitudinal analysis and a multivariable analysis controlled by possible confounders.

It is worth noting from our results that despite the use of high doses of ICS in the SA group, more than 60% of them were uncontrolled, proportion that was higher when compared with NSA. Even in the overall population attending those specialized sites, 56.6% were uncontrolled. This was consistent with other studies, where SA patients had an increased number of exacerbations, as well as higher rates of hospital admissions and ER attendances [19].

Current asthma management guidelines acknowledge that PEF monitoring during exacerbations of asthma help determine the severity of these flares and can be useful to guide therapeutic decisions [8]. However, in our study less than 40% in both SA and NSA patients reported using PEF meter at least once in the last year. This low proportion is in accordance with the AIRLA study [10], which evaluated the impact of asthma in Latin America and showed that 54% of asthma patients ever had no spirometry and 96% had not ever performed a PEF test, according to patient self-reported surveys. Although there are differences in methodologies between our study and AIRLA study, which is population-based, both data reinforce that spirometry is not common in Latin America region and the reasons for that should be further evaluated.

As other studies have described that SA patients often display a high number of comorbidities [12, 21], our results found relevant differences between SA and NSA patients with regards to obesity, hypertension, COPD, psychological disturbances and gastroesophageal reflux disease. The SA Research Program (SARP) also showed that hypertension, obesity and gastroesophageal reflux disease are associated with SA [22]. Specially for hypertension, it is a significant predictor for asthma severity, but this relationship was found predominately among the white population, rather than in black patients [23]. It is important to take into account that aging could be a confounding factor in the association found between SA and comorbidities observed in our study. For psychological disturbances, the literature already described higher levels of anxiety and depression in SA patients [24, 25], which is in line with the results found in our study. Finally, COPD has been recently described to be higher in SA group by other authors [26, 27].

The external validity of the current study is limited, as most patients were recruited from specialized asthma clinics and the sample size for ASLA study was not calculated to be representative for each country. Another limitation of our study is the cross-sectional evaluation of medication use, as we did not capture the treatment step up and step down through the past years. The comparison between our result and other studies is difficult, mainly due to the fact that most part of the studies refers to the prevalence of refractory asthma and not to SA. Our analysis may also be subject to recall bias.

Conclusion

The results of our study may support the identification of SA in our region contributing to the better management of these patients. What is more, this study highlights the need to improve the asthma control in these patients that are already being treated with high ICS doses. Furthermore, complexity of SA phenotypes would require deeper investigations to determine specific phenotypes to be able to prescribe the precise therapy to increase the level of control in SA patients.