Short-acting β₂-agonist prescriptions are associated with poor clinical outcomes of asthma: the multi-country, cross-sectional SABINA III study

Discussion

Our study describing asthma medication prescriptions by PCPs and specialists for patients with mild or moderate-to-severe asthma in 24 countries with a wide global representation confirms high levels of SABA prescriptions, with 38% of patients being prescribed ≥3 SABA canisters in the past 12 months. Use of ≥3 SABA canisters per year is considered undesirable since it indicates overreliance on SABA for the management of persistent symptoms [28], usually related to the underuse of ICS and other controllers. GINA-defined controlled or partly controlled asthma specifies that SABA reliever use should not be >2 doses per week, which equates to <2 standard SABA canisters per year. In support of this threshold, in our study, even after adjusting for known confounding factors, SABA prescriptions for >2 canisters per year were associated with an increasing rate of severe exacerbations and a lower likelihood of satisfactory symptom control. More than half of the patients with mild asthma receiving SABA alone for as-needed use were prescribed ≥3 canisters and almost one-third were prescribed ≥10 canisters, suggesting that a majority should have been considered for additional maintenance treatment with controllers. Among patients prescribed controller treatment, >60% received ≥3 SABA canister prescriptions in the past year and almost one-third received ≥10 SABA canister prescriptions, suggesting overuse of SABA instead of optimisation of controller treatments. Although there were differences in prescribing between PCPs and specialists, the pattern of SABA overreliance was common to both. SABA monotherapy for mild asthma was more commonly prescribed by PCPs and in higher numbers, and SABA prescriptions for moderate-to-severe asthma by both categories of prescriber were similarly high.

In some of the countries surveyed, SABA may be obtained without a prescription, increasing the potential for SABA overuse [29, 30]. Overall, one-fifth of patients in our study reported obtaining SABA in this way, of whom one-half purchased ≥3 canisters in the past year. In most cases (77%), these canisters were in addition to those prescribed by their physician. The potential for overuse by patients receiving SABA from two sources is suggested by the fact that among such patients, 70% also received prescriptions for ≥3 SABA canisters in the past year and 35% also received prescriptions for ≥10 SABA canisters.

Overall, these findings are similar to what has been observed in the SABINA I and II studies in Europe. Across the UK, Germany, Spain, Sweden and Italy [22], SABA prescription/possession of ≥3 SABA canisters per year (33%) was slightly lower than that in SABINA III, although differences were observed between countries. SABA overuse ranged from 38% in the UK to 9% in Italy. Subsequently, it was confirmed that in Italy, SABA overuse was higher (>50%) when SABAs dispensed by pharmacists, including those purchased without a prescription, were included [25].

The findings in our study confirming the association between SABA prescriptions and poor asthma outcomes contribute to the growing evidence that SABA overuse in asthma needs to be addressed if further reductions in asthma morbidity and mortality are to be achieved. An association between SABA prescription/possession and severe exacerbations [22, 24, 25], and even asthma deaths [24], has been reported in SABINA I (UK) and II (Sweden and Italy). Similar findings have been observed in other studies of SABA use in asthma [28], with high SABA overuse, which may occur even on symptom-free days [31], being associated with worse asthma control.

The growing concern about the negative effects of SABA use on global efforts to improve asthma outcomes has led to research into alternative treatment options for providing quick relief from asthma symptoms either for occasional symptom relief or, more importantly, when breakthrough symptoms herald an approaching severe exacerbation. Foremost has been examining the potential of ensuring that use of a rapid-onset bronchodilator is always accompanied by use of an ICS to ensure that the underlying airway inflammation is also addressed at these critical times. The single-inhaler maintenance and reliever approach was initially trialled with formoterol, a long-acting bronchodilator with a rapid onset of action like that of a SABA, combined with budesonide. Most research has been focused on this combination, but efficacy has also been shown for the combination of formoterol with beclomethasone [7] and is currently being evaluated for combinations of a SABA with an ICS [8, 9]. The anti-inflammatory reliever approach has been shown to be highly effective in mild asthma [15–18], where it may be used without maintenance dosing, and in moderate-to-severe asthma (with fixed daily dosing of the same combination as maintenance treatment) [7]. Consequently, an anti-inflammatory reliever approach has become the preferred option in both the GINA report and the recently published updated report of the National Asthma Education and Prevention Program in the USA, and in other national guidelines and formularies [13, 19, 32]. The 2019 World Health Organization Model List of Essential Medicines, which represent “minimum medicine needs for a basic healthcare system”, includes budesonide/formoterol for use in asthma [33].

Given the entrenched and time-honoured position of SABAs in asthma care spanning >50 years, SABINA and similar studies provide potentially useful information about the magnitude of the problem relating to SABA use that may be used to assess the gains that are possible if this alternative reliever strategy were to be introduced globally. The current study is focused on data from several countries, most of which do not have national databases from which to gauge SABA use. Although not fully representative of asthma care within each country and biased towards specialist services for asthma patients, it provides a snapshot of the situation in a range of countries, including some with limited healthcare insurance or national provision of care. Our findings reveal overuse of SABAs by both PCPs and specialists, and although the assessment of asthma control was not as poor as that reported in many cross-sectional surveys [34, 35], most patients were not optimally controlled, asthma attacks remained common and both were associated with SABA use. On the other hand, 37.8% were recorded as having no SABA prescriptions, a proportion similar to that seen in SABINA I in the UK [22]. Although some had obtained SABA OTC, it is likely that many such patients, 89.2% of whom had been prescribed an ICS/LABA combination, may have already been switched to ICS/formoterol as reliever. Unfortunately, the size of this group could not be accurately assessed in our study.

In strategising how to encourage the use of the preferable reliever option, several approaches are needed. First, OTC SABA purchases may need to be better regulated in some countries as part of the education process on reliever use and limits should be put in place. Entrenched prescribing habits in well-resourced health services, such as automatic repeat prescriptions, may result in high and unnecessary SABA prescriptions [36]. Although easier to enact in developed countries, such a limitation will be difficult in poorer nations where the relatively low cost and accessibility of SABAs are relied on for short-term benefit despite the fact that they may help to entrench poor asthma care. In such settings, the bias towards using relievers rather than more costly controller medications is likely to be greater. Access to affordable combination medications should be a key priority as it is likely that in these countries, the single-inhaler maintenance and reliever approach will be of greatest benefit in view of its strong effects on reducing asthma worsening and attacks, which pose an avoidable high burden on health services [1, 37]. These approaches will need to be accompanied by national initiatives targeting patients, physicians and other stakeholders such as pharmacists to increase awareness of updated treatment guidelines. Creation of national asthma programmes based on current evidence-based asthma guidelines and tailored to the context of clinical practice and local resources can play a critical role in this endeavour. National or regional asthma programmes have been shown to be more effective than conventional treatment guidelines in improving asthma care [38]. Patient involvement is also crucial and patient advocacy groups can play a significant role in disseminating appropriate treatment information [39]. Besides these measures, the current move away from a SABA as reliever to be replaced by an ICS-containing rapid-onset reliever for all treatment steps, as now proposed by GINA, may, in some countries, represent the most significant step towards addressing overreliance and overuse of SABAs. This trend was already evident in our study in the high proportion of patients who received prescriptions for ICS/LABA and no provision for a SABA.

As a limitation, first, it is recognised that SABA prescriptions may not necessarily reflect actual usage, which is likely to be lower. However, overprescribing, particularly in poorly resourced settings, may result in medications being passed on to family and friends, increasing the potential for mis-assessments and haphazard treatment. Second, the assignment of asthma severity based on GINA treatment steps appeared to be poorly adhered to by investigators, as evidenced by the large proportion of patients with mild asthma who received an ICS/LABA prescription. It is possible that instead of assigning severity according to the 2017 GINA classification [26], a later version that proposed as-needed ICS/formoterol for mild asthma may have been followed. In our study, the nonrandom selection of sites with a majority representing specialist care resulted in the enrolment of more patients with moderate-to-severe asthma. In view of this bias, we have avoided comparisons of data obtained from participants enrolled by specialists with those recruited in primary care. However, these data from different contexts and of differing severities of asthma permit broad generalisations. Furthermore, this cross-sectional study does not permit an assessment of a causal link between SABA prescriptions and asthma outcomes, and does not discount reverse causality; the results simply represent an association. Our aim to include data from a large number of countries and practices with different methods of recording clinical data necessitated acceptance of limitations in methods of collecting source data, such as reliance on patient recall for some questions and limiting the scope of the questionnaire. For example, data on comorbidities and the number and type of all maintenance medications was not obtained. Lastly, the basis for an asthma diagnosis in each participant was not requested. However, misdiagnosis is not likely to have had an impact on the main findings of this study.

Our SABINA III findings demonstrating that 38% of patients in 24 countries in five continents are overprescribed SABAs (≥3 canisters per year) extend the data from the SABINA studies in Europe [22–25]. Although drivers for SABA prescribing may differ by country, SABA overprescription results in an unnecessary burden of poor asthma symptom control and severe asthma exacerbations with their attendant risks. These findings support the need for continued efforts to improve asthma care in these countries, particularly relating to the prescribing of SABAs and the need to switch to combination medications that provide both quick symptom relief and an anti-inflammatory effect.