Respiratory Medicine 202 (2022) 106942 Contents lists available at ScienceDirect Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed Original Research Is asthma control more than just an absence of symptoms? An expert consensus statement Giorgio Walter Canonica a, *, Antonio Spanevello b, c, Luis Pérez de Llano d, Christian Domingo Ribas e, John D. Blakey f, g, Gabriel Garcia h, Hiromasa Inoue i, Margareth Dalcolmo j, Dong Yang k, Soniya Mokashi l, 1, Abhishek Kurne m, 1, Aman Kapil Butta n, 1 a Humanitas University, Personalized Medicine Asthma & Allergy Clinic-Humanitas Research Hospital IRCCS, Milan, Italy b Dipartimento di Medicina e Chirurgia, Università degli Studi dell’Insubria, Varese, Italy c Dipartimento di Medicina e Riabilitazione Cardio-Respiratoria, Istituti Clinici Scientici Maugeri, IRCCS Tradate, Varese, Italy d Pneumology Service, Hospital Universitario Lucus Augusti, EOXI Lugo, Cervo e Monforte, Spain Grupo BIOCHUS, Instituto de Investigación Sanitaria Santiago de Compostela (IDIS), Spain e Pulmonary Service, Corporació Sanitària Parc Taulí, Universitat Autònoma de Barcelona (UAB), Sabadell, Spain f Respiratory Medicine, Sir Charles Gairdner Hospital, Nedlands, Western Australia, Australia g Medical School, Curtin University, Perth, Australia h Pneumology, Pulmonary Service, Hospital R Rossi, La Plata, Argentina i Department of Pulmonary Medicine, Graduate School of Medical and Dental Sciences, Kagoshima University, Kagoshima, Japan j Reference Center Hélio Fraga, Fundação Oswaldo Cruz (Fiocruz)/Ministry of Health, Rio de Janeiro, Brazil k Department of Pulmonary Medicine, Zhongshan Hospital, Fudan University, Shanghai, China l GSK Brentford, Middlesex, UK m Global Medical Affairs, GSK, Asia House, Singapore n GSK, Asia House, Singapore A R T I C L E I N F O A B S T R A C T Keywords: Purpose: Definitions and measures of asthma control used in clinical trials and in clinical practice vary consid­ Asthma erably. There is also misalignment between patients and healthcare professionals (HCPs) in terms of under­ Asthma control standing and managing asthma control. This study aimed to progress towards a consensus definition of asthma Delphi consensus control, and evaluate disparities between HCP and patient perspectives. Exacerbations Quality of life Basic procedures: A two-stage Delphi questionnaire involving asthma specialists sought to identify areas of Symptom control consensus on aspects of asthma control in clinical practice. Results were compared with those of a structured literature review to assess if existing guidance and measures of asthma control used in studies correlated with practice. Eighty-two panelists took part in the Delphi questionnaire. The structured literature review included 185 manuscripts and 31 abstracts. Main findings: Panelists agreed that there was no standard definition of asthma control, confirmed by a total of 19 different composite consensus/guideline definitions and/or validated measures of control being identified across the Delphi study and literature review. Panelists agreed on the positive associations of well-controlled asthma with patient outcomes, but not on the components or thresholds of a working definition of control. Principal conclusions: A universally accepted definition and measure of asthma control that is utilized and un­ derstood by patients, HCPs, and researchers is required. Abbreviations: ACQ, Asthma Control Questionnaire; ACT, Asthma Control Test; ATS/ERS, American Thoracic Society/European Respiratory Society; GINA, Global Initiative for Asthma; HCP, Healthcare professional; SABA, Short-acting β2-agonist. * Corresponding author. E-mail address: canonica.gw@gmail.com (G.W. Canonica). 1 At time of study conduct. https://doi.org/10.1016/j.rmed.2022.106942 Received 6 May 2022; Received in revised form 8 July 2022; Accepted 21 July 2022 Available online 4 August 2022 0954-6111/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 1. Introduction and disparities in HCP perceptions on patient opinion and understand­ ing. The Delphi survey method is a recognized group facilitation tech­ The goal of asthma management is to achieve and maintain symptom nique designed to transform opinion into group consensus [13]. While control, and reduce future risk of adverse outcomes, including exacer­ the authors acknowledge that patient opinion is also extremely impor­ bations, lung function decline, and side-effects of medications, irre­ tant, circulation of the Delphi survey to patients was outside of the scope spective of asthma severity [1]. Periodic assessment and ongoing asthma of the current study due to regulatory-imposed constraints. We were control monitoring can confirm treatment optimization or alert the therefore limited to HCP perceptions of patient opinion and under­ healthcare professional (HCP) to the need for treatment adjustment, standing only. Secondly, a structured literature review examined mea­ further investigations, or patient education. Asthma control has been sures of asthma control used in clinical trials and real-world studies. described as having two distinct components: one patient-focused, based Findings from both the Delphi study and literature review were on symptoms, and one HCP-focused, based on objective measures of compared to provide a consensus regarding challenges and approaches pathophysiological abnormalities [2]. Well-controlled asthma is asso­ in achieving asthma control in clinical practice. Both the Delphi study ciated with good symptom control and maintenance of normal activity and literature study were carried out by a research team at Ashfield levels [3], yet many patients and HCPs overestimate their level of MedComms, an Inizio company. symptom control [4], with patients potentially tolerating symptoms and reduced quality of life, and HCPs accepting a sub-optimal level of control 2.1. Delphi survey method [5]. Indeed, the lack of a universally accepted definition of asthma control [6] can potentially result in HCPs underestimating the level of 2.1.1. Questionnaire development control. HCPs often define well-controlled asthma as minimal symptoms The Delphi method consisted of two rounds of questionnaires, which and impact upon activities, while patients often define it as effective were developed, reviewed, and approved by all authors. The first and management of symptoms that potentially exceed their second questionnaires were disseminated to panelists in March 2021 individually-defined threshold [7,8]. and April 2021, respectively. Further details on the questionnaires are Following the American Thoracic Society (ATS)/European Respira­ provided in the Online Repository. tory Society (ERS) task force’s recommendations [6], a combined approach to assessing asthma control from both symptom control and 2.1.2. Panelists future risk is widely accepted. Yet, although measures and definitions Nine of the authors are practicing clinicians (GWC, AS, LPdL, CDR, for asthma control exist, there is no “gold standard” definition [6]. The JDB, GG, HI, MD, DY) and completed the questionnaire. Seventy-three Global Initiative for Asthma (GINA) and the National Asthma Education additional HCPs with a professional interest in asthma were also and Prevention Program provide definitions of control that appear to invited to complete the questionnaire. This provided an additional eight encompass both the patient’s clinical disease characteristics and limi­ panelists per one practicing clinician author to offer representative in­ tations, including diurnal and nocturnal symptoms, activity limitation, formation and greater generation of data. Panelists who were invited to and use of reliever medication [3,9]. One of the long-term goals of participate through SERMO (a third-party, centralized database of asthma management as stipulated by GINA is “to minimize future risk of HCPs) received an honorarium which was compliant with GSK’s fair asthma-related mortality, exacerbations, persistent airflow limitation market value evaluation for each country; authors did not. All panelists and side-effects of treatment”, which is often overlooked in short-term provided written informed consent to participate. Further details on questionnaires. There are also several standardized questionnaires, e. recruitment and screening are available in the Supplementary Methods g., the Asthma Control Questionnaire (ACQ) [10] and the Asthma within the Online Repository. Control Test (ACT) [11]. However, it should be noted that HCP bias may have influenced the development of these tools. For example, symptoms 2.1.3. Analysis and interpretation of responses identified in the ACQ were selected based upon HCP opinion of those Descriptive summaries were produced for closed questionnaires. most important [10]. During ACT development, stepwise regression Responses to Likert scale-based questionnaires were grouped into three methods were used to select a subset of items that showed the greatest categories: disagree, neither agree nor disagree, and agree. Consensus on discriminant validity regarding the HCP’s rating of asthma control, for an item was reached when ≥66% of panelists voted within the agree or which no accepted system of defining control in relation to these goals disagree category. Further details on coding of responses are available in has been articulated [11]. In addition to clinical disease characteristics the Online Repository. and limitations, some definitions/measures of asthma control consider exacerbations and pulmonary function [11,12], highlighting disparities 2.2. Structured literature review between definitions and measures. Different measures of asthma control can also be useful for various functions, such as quickly screening for A structured literature review was undertaken to identify measures poor control [12], highlighting when an intervention is needed [3], or and guidelines for asthma control in clinical publications. The review longitudinal evaluation of the efficacy of an intervention [10]. sought to identify all relevant publications since the Gaining Optimal Given these disparities between definitions and expectations of Asthma ControL study in 2004 by Bateman et al. [14] until search end in asthma control, there is a need for an expert consensus regarding chal­ March 2021. Congress abstracts from 2018 to 2021 were included as it lenges and approaches in achieving well-controlled asthma. The current was presumed that abstracts published prior to this would have been study aimed to progress towards reaching a consensus on asthma con­ published as full manuscripts. PubMed, Cochrane, and Embase data­ trol, by examining the validity and utility of current measures of asthma bases were searched for relevant articles. All titles and abstracts were control, evaluating which factors additional to those included in stan­ manually screened by a reviewer from the research team. Publications dard definitions of asthma control should be considered, and providing were included if they met pre-defined criteria, namely: (1) adult study expert recommendations on the future assessment of asthma control in population (aged ≥18 years); (2) control used as a study endpoint; and clinical practice. (3) an appropriate study design was used (i.e. randomized controlled trial, systematic review, meta-analysis, and/or real-world evidence). 2. Methods Following initial screening, the following details were extracted from the included publications to quantify the results: (1) trial design; (2) The study was composed of two parts. Firstly, a two-stage Delphi control as primary/secondary endpoint; (3) population; (4) total N; (5) method was used to reach expert consensus on definitions of asthma center type; (6) country; (7) measure of control used (guidelines, vali­ control used in clinical practice, HCP understanding of asthma control, dated measure [e.g. ACT, ACQ], symptomatic measure of control [e.g. 2 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 frequency of exacerbations, use of rescue medication], patient-reported 3.3. Outcome measure indicative of well-controlled asthma measures of control [e.g. sleep disturbances, impact on day-to-day ac­ tivities]); and (8) any further details. Details were extracted and recor­ Panelists agreed that there is no agreement that there is a clear, ded from the publications as written; no assumptions were made by the published definition of asthma control (Table 2). This was confirmed by research team (full details of the data extraction are available in Table the identification of 19 different composite consensus/guideline defi­ E1 in the Online Repository). Results of the structured literature nitions or validated measures of control across the Delphi study and research were reported descriptively. literature review (Table 3). Further details on the literature review, including search criteria and Notwithstanding the above finding, panelists agreed that the strategy, inclusion/exclusion criteria for relevant papers, and informa­ currently accepted definition of asthma control is appropriate (Table 2). tion on secondary and tertiary reviewers, are provided in the Online When asked how asthma control is defined in clinical practice, panelists Repository. noted that they largely used GINA, ATS/ERS, ACT, or ACQ definitions. Panelists agreed that GINA, the most commonly cited document by 3. Results panelists, clearly defines well-controlled asthma symptoms (Table 2). When the definitions cited by panelists were compared with the litera­ 3.1. Delphi survey method ture, their use was more varied, with asthma control defined using multiple other measures, such as other guidelines, symptomatic mea­ In total, 82 panelists, including the nine authors, completed the first- sures, or patient-reported outcomes (Table 3 and Figure E2> in the round Delphi questionnaire. Overall, 63 panelists, including the nine Online Repository). authors, took part in the second-round Delphi questionnaire (19 pan­ Consensus was reached by Delphi panelists that asthma control elists were lost in follow up). Panelist demographics are detailed in should be considered as a continuum, yet panelists also agreed that Table 1. asthma control should be considered as a categorical variable, with level of control defined as poor, good, or total (Table 2). When asked what level of outcome measure they consider indicative of well-controlled 3.2. Structured literature review asthma, the majority of panelists reported that they use ACT values (59% of panelists; Figure E3 in the Online Repository). Of those that The literature research identified 664 manuscripts and 89 abstracts. provided a specific score, the majority stated they defined well- Following exclusions, 185 manuscripts and 31 abstracts were included controlled asthma as an ACT score of >19 or >20 (40 and 33%, in the full data extraction. A PRISMA flow diagram of included and respectively). In the literature, only 23% of included publications used excluded publications, stratified by study design, is detailed in Figure E1 ACT as a measure of asthma control; of those that noted specific values, in the Online Repository. Asthma control was a primary endpoint for 57% stated that an ACT score ≥20 was indicative of well-controlled 36% and secondary endpoint for 17% of included publications. Nearly asthma. The second most commonly used outcome measure was ACQ half of publications used a single measure to define/measure asthma (16% of panelists, 27% of included publications), and of those that control (49%), 20% used two measures, and 29% used 3–6 measures. utilized ACQ, 73% of panelists advised a score of <0.75 as being indicative of well-controlled asthma, while 95% of publications advised Table 1 a score of ≤0.75. Delphi questionnaire respondent demographics. When asked specifically about the frequency of short-acting β2- Demographic First-round Second-round agonist (SABA) reliever medication in a 4-week period that they would questionnaire (N = 82) questionnaire (N = 63)a consider indicative of poorly controlled asthma in clinical practice, re­ Primary medical specialty, n sponses from the surveyed HCPs varied widely (see Figure E4 in the Allergist 14 8 Online Repository). Only seven of the included publications included General practitioner/ 30 20 rescue medication (SABA [n = 4] or other [n = 3]) as a measure of primary care physician control, all with different definitions for SABA use indicative of poorly Pulmonologist 38 35 Years qualified in specialty,b n controlled asthma. >10 18 11–15 17 3.4. Most important measures of asthma control from the HCP and 15–20 22 patient perspective 20–25 23 25–30 17 30+ 14 Consensus was reached by panelists that when categorizing the level Mean time spent on activities, %b of control, both the HCP and the patient’s perspective is considered Actively treating patients 82% (Table 2). When asked to report which measures of asthma control they Academic/research 11% Admin/other 7% thought were most important from both their perspective and the pa­ Number of patients treated in a typical monthb,c tient’s perspective, most panelists reported the number/frequency of 0–30 17 exacerbations (HCP 40% vs. patient 39%) and severity of symptoms 31–60 26 (HCP 33% vs. patient 36%) as the most important (see Figure E5 in the 61–90 10 Online Repository). Despite this apparent close alignment of the patient 91–120 12 121+ 17 and HCP, as perceived by the panelists, it was agreed that there is a Location, n misalignment between the patient’s most important measures of control Argentina 12 9 and those of the HCP (Table 2). Australia 11 9 When asked specifically, consensus was reached by panelists that Brazil 11 8 China 12 6 symptoms represent an important component of the assessment of Italy 12 11 asthma control in clinical practice (Table 2), symptom burden is directly Japan 11 9 related to asthma control, and asthma control is the same as symptom Spain 13 11 control (Table 2). Consensus was also reached by panelists that airway a All respondents also took part in the first-round questionnaire. obstruction represents an important component of the assessment of b Question asked in first-round questionnaire only. asthma control in clinical practice (Table 2). However, few publications c Patients were asked to give their response as it was pre-COVID-19 pandemic. appear to explore symptom severity, exacerbations, or airway 3 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 Table 2 Level of agreement to statements presented to healthcare professionals. Statement Proportion of panelists, %a Consensus reached? Net Net neither agree Net agree nor disagree disagree Outcome measure indicative of well-controlled asthma There is a clear, published definition of asthma control (1) 65 28 7 X The currently accepted definition of asthma control is appropriate (1) 71 24 5 ✓ GINA clearly defines well-controlled asthma symptoms (in the past four weeks: daytime asthma symptoms less 95 2 3 ✓ than twice a week, no night waking due to asthma, SABA reliever use for relief of symptoms two or fewer times/week, no activity limitation due to asthma) (2) Asthma control is best considered as a continuum (1) 85 7 7 ✓ Categories of control should be defined as poor/good/total (1) 67 21 12 ✓ Most important measures of asthma control from the HCP and patient perspective When categorizing the level of control, both the physician and patient’s perspective is considered (1) 87 7 6 ✓ Physicians and patients share the treatment goal of good asthma control (1) 85 9 6 ✓ There is misalignment between the patient’s most important measures of control and those of the physician (2) 68 11 21 ✓ Symptoms represent an important component of the assessment of asthma control in clinical practice (1) 98 2 0 ✓ Symptom burden is directly related to asthma control (2) 98 0 2 ✓ Asthma control is the same as symptom control (2) 71 8 21 ✓ Airway obstruction represents an important component of the assessment of asthma control in clinical practice 85 12 2 ✓ (1) Poorly controlled asthma and future risk Patients understand the importance of asthma control, and the link between good asthma control and future 66 16 18 ✓ risk (1) Well-controlled asthma is associated with improved quality of life (2) 95 3 2 ✓ Well-controlled asthma is associated with reduced future risk (of exacerbations, loss of control, unscheduled 94 3 3 ✓ healthcare use, and long-term adverse outcomes) (2) Well-controlled asthma is associated with improved quality of life and reduced future risk, but there are many 84 10 6 ✓ confounding factors (2) Well-controlled asthma is associated with ability to perform normal day-to-day activities (including minimal 98 0 2 ✓ sleep disturbances) (2) Patients who achieve good control of asthma, according to a guideline-based composite measure (e.g. the 95 4 1 ✓ patient has good asthma control according to GINA/ATS/GEMA etc. definition of control), also achieve greater improvements in quality of life (1) Asthma control and asthma exacerbations There is a clear, published definition of asthma exacerbations (1) 79 13 7 ✓ A reduction in frequency of exacerbations is indicative of improved asthma control (2) 94 5 2 ✓ The importance of HCP and patient education It is important to educate patients on the significance of good asthma control on current and future risk and 100 0 0 ✓ health outcomes (1) It is important to educate healthcare professionals on the significance of good asthma control on current and 95 5 0 ✓ future risk and health outcomes (1) Well-controlled asthma is associated with a good patient-doctor relationship (2) 81 17 2 ✓ Asthma control and adherence Patients understand that they should continue to adhere to their medication regimen, even when they are not 63 12 24 X experiencing any symptoms of asthma (1) Adherence to therapy should be considered as part of the assessment of a patient’s level of asthma control (2) 89 5 6 ✓ Well-controlled asthma is associated with minimal use of as-needed reliever medication (SABA use) (2) 90 3 6 ✓ Well-controlled asthma is associated with good adherence with the minimal effective treatment dose of inhaled 87 10 3 ✓ corticosteroids (2) Well-controlled asthma is associated with using the most cost-effective treatment (2) 84 13 3 ✓ Net agreement defined as total percentage of panelists answering “somewhat agree” or “strongly agree”. Net disagreement defined as total percentage of panelists answering “strongly disagree” or “somewhat disagree”. Consensus on an item was reached when ≥66% of panelists voted within the agree or disagree category. Number in brackets indicates Delphi questionnaire round in which statement was posed to panelists. ATS, American Thoracic Society; GEMA, Spanish Guideline on the Management of Asthma; GINA, Global Initiative for Asthma; HCP, healthcare professional; SABA, short-acting β2-agonist. a Due to rounding, totals may not equal exactly 100%. obstruction as measures of asthma control. When the individual do­ adherence to therapy as a specific measure when assessing study par­ mains of asthma control were analyzed, nearly all definitions and ticipant’s level of asthma control. measures cited by panelists or in literature included aspects on diurnal Panelists were asked to score on a scale of 1–10 (1 being not at all and nocturnal symptoms, but only seven (35%) cited measures of con­ important and 10 being very important) how important each of the trol included exacerbations as a considered domain, and none included factors were for achieving well-controlled asthma. NET important was airway inflammation (Table 3). defined as the total percentage of panelists answering 8–10, Net neutral was defined as the total percentage of panelists answering 4–7, and Net not important was defined as the total percentage of panelists answering 3.5. Asthma control and adherence 1–3. Consensus on an item was reached when ≥66% of panelists voted within the important or not important category. Panelists did not agree that patients understood the importance of adhering to their medication regimen, even when not experiencing any symptoms of asthma (Table 2). It was suggested that there was a need for 3.6. Asthma control and asthma exacerbations better patient education on the importance of adherence, even when not symptomatic (Table 4). None of the included publications assessed There was consensus that there is a clear, published definition of 4 G.W. Canonica et al. Table 3 Domains and components of guideline/consensus definitions and validated measures of control. Cited Number of Number of Recall period Domains and components measured measure of panelists publications Nocturnal Diurnal Activity Use of Exacerbations Patient self- Spirometry/ ED visits or Use Eosinophil control noting citing measure, symptoms symptoms limitation reliever assessment pulmonary hospitalizations of testing measure, n n (%) medication function OCS (%) Guidelines/ GINA 61 (84) 47 (13) 4 weeks ✓ ✓ ✓ ✓ recommendations ATS/ERS 42 (58) 8 (2) Not specified ✓ ✓ ✓ ✓ ✓ ✓ taskforce GEMA 5 (7) 0 (0) ✓ ✓ ✓ ✓ ✓ ✓ ✓ SBPT 3 (4) 0 (0) 4 weeks ✓ ✓ ✓ ✓ ✓ ✓ Australian 1 (1) 0 (0) 4 weeks ✓ ✓ ✓ ✓ Asthma Handbook EPR-3 0 (0) 8 (2) 1 week (symptoms), ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ since last visit (exacerbations and hospitalizations) CDC’s 0 (0) 1 (0) 1 year for ✓ ✓ ✓ ✓ ✓ ✓ National exacerbations, ED/ Asthma hospitalizations and Survey activity limitation Validated measures ACT 30 (41) 84 (23) 4 weeks ✓ ✓ ✓ ✓ ✓ ACQ 23 (32) 97 (27) ACQ-7 87 (24) 1 week ✓ ✓ ✓ ✓ ✓ ACQ-6 8 (2) 1 week 5 ✓ ✓ ✓ ✓ ACQ-5 4 (1) 1 week ✓ ✓ ✓ PCAQ-6 0 (0) 4 (1) ✓ ✓ ✓ ✓ ATAQ 0 (0) 3 (1) 1 week ✓ ✓ ✓ ✓ ✓ ACSS 0 (0) 1 (0) 1 week ✓ ✓ ✓ ✓ ✓ ✓ CARAT 0 (0) 1 (0) ✓ ✓ ✓ ✓ SASCQ 0 (0) 1 (0) 30 days ✓ ✓ ✓ ✓ ✓ AIRQ 0 (0) 1 (0) 2 weeks for ✓ ✓ ✓ ✓ ✓ ✓ ✓ impairment measures and 12 months for risk measures JACS 0 (0) 1 (0) 4 weeks ✓ ✓ ✓ ✓ ✓ RCP3 0 (0) 1 (0) 1 month ✓ ✓ ✓ ACQ, Asthma Control Questionnaire; ACSS, Asthma Control Scoring System; ACT, Asthma Control Test; AIRQ, Asthma Impairment and Risk Questionnaire; ATAQ, Asthma Therapy Assessment Questionnaire; ATS/ERS, American Thoracic Society/European Respiratory Society; CARAT, Control of Allergic Rhinitis and Asthma Test; CDC, Centers for Disease Control; ED, emergency department; EPR, Expert Panel Report; GEMA, Spanish Guidelines on the Management of Asthma; GINA, Global Initiative for Asthma; JACS, Japan Asthma Control Survey Questionnaire; OCS, oral corticosteroids; PCAQ-6, Six-item Perceived Control of Asthma Questionnaire; RCP3, Royal College of Physicians 3-item asthma questionnaire; SASCQ, Seattle Severity and Control Questionnaire; SBPT, Sociedade Brasileira de Pneumologia e Tisiologia. Respiratory Medicine 202 (2022) 106942 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 asthma exacerbations (Table 2). Whilst the authors acknowledge that and general practitioners, coupled with assessment and analysis of reductions in exacerbation frequency in a clinical trial setting often does published literature, demonstrated that there is a need for a genuine not correlate with asthma control, the panelists agreed that a reduction consensus definition on asthma control with high penetration. With in frequency of exacerbations is indicative of improved asthma control respect to the validity and utility of current measures of asthma control, in a clinical setting (Table 2). When asked about the frequency of ex­ while HCPs may believe they are using a universally accepted measure acerbations in a 12-month period that they would consider indicative of of asthma control, this study highlights that no such universally poorly controlled asthma in clinical practice, responses were varied, accepted measure exists. The identification of 19 different composite with 74% of panelists stating 1–3 exacerbations per year (see Figure E6 consensus/guideline/expert paper definitions or validated measures of in the Online Repository). Fifteen of the included publications noted control identified both in the literature published since 2004, and ac­ frequency of exacerbations as an outcome measure. Of these, the ma­ cording to expert clinical opinion, demonstrates the disparities and jority calculated the total number of exacerbations during the study difference between clinical trials and clinical practice. It is possible that period, while only one of the publications included specified that poor many more outcome measures exist and are used in clinical practice asthma control could be indicated by ≥2 exacerbations in the past year. globally. Some validated measures were used more frequently in clinical practice and in clinical trials than others, namely the ACT and ACQ. Yet 3.7. Poorly controlled asthma and future risk among these measures, and despite published guidance, there was no standardized, specific value at which the literature or expert panel Panelists concurred that, in their opinion, patients understand the considered asthma to be well-controlled, and research suggests these importance of asthma control, and the link between good asthma control measures may not be used as standard in practice [15]. The GINA def­ and future risk of worsening asthma symptoms (Table 2). They agreed initions of asthma control were the most frequently selected document that well-controlled asthma is associated with reduced future risk and by panelists. GINA highlights that there are many available systems for improved quality of life, including the ability to perform normal day-to- the assessment of asthma symptom control, which may correlate with day activities and minimal sleep disturbances, despite there being many one another but are not identical [3]. Panelists agreed that there is no confounding factors (Table 2). When asked specifically, panelists re­ clear, published definition of asthma control. However, they agreed that ported that severe exacerbations, decline in lung function, and loss of the currently accepted definition of asthma control is appropriate. It asthma control were the most important future risk factors associated could be suggested that HCPs have confidence in the measure of control with poorly controlled asthma (see Figure E7 in the Online Repository). they feel is most appropriate for their individual practice. To reach a universal definition of asthma control, it is important to 3.8. The importance of HCP and patient education examine which factors should be considered with those already included in standard definitions of asthma control, while addressing the needs of Panelists considered it important that both patients and HCPs are both HCPs and patients. With respect to the needs of HCPs, while pan­ educated on the significance of well-controlled asthma on current/ elists strongly agreed that asthma control should be considered as a future risk of worsening asthma and health outcomes (Table 2). It was continuum, they also suggested that a universally accepted definition of suggested that it is important to increase patient understanding of the asthma control would require some binary categorization to ensure its correlation between asthma control and improved outcomes, and the usefulness in clinical decision-making. For example, for well-controlled impact that treatment and lifestyle choices can have on control asthma, patients could continue to be monitored and managed using (Table 4). Panelists suggested that setting functional goals with the their current treatment regimen, whereas patients with asthma that is patient could also help to improve asthma control (Table 4). When asked not well controlled may require treatment adjustment or further inves­ how the importance of asthma control could be best communicated to tigation. Concepts of “partial” asthma control may not be as appropriate the patient, panelists felt that good communication and the patient- for use in clinical practice as binary concepts of “well-controlled” or doctor relationship were very important, as were materials to share “poorly-controlled” asthma. This suggestion is in line with other studies with the patient (Tables 2 and 4). HCP panelists did not feel it was as stating that asthma severity categorization could result in improved care important to increase consultation time or endeavor for more effective [16]. Furthermore, the semantics of current measures of asthma control history and note taking in consultations, nor use media outlets such as may introduce misunderstanding and error: the term “control” may be social media for disease education (Table 4). understood differently by patients compared with HCPs [3]. Although outside the scope of this study, any new universally accepted measure of 4. Discussion asthma control should ideally include a numerical scoring system, with values used in a more dichotomous manner to indicate either a need for The present expert opinion of practicing pulmonologists, allergists, further evaluation/treatment or that the patient is on adequate Table 4 Healthcare professional rating of importance of statements presented in Question 3, Round 2. Statement Proportion of panelists, % Consensus reached? Net Net Net not important neutral important The importance of HCP and patient education Improving the patient’s understanding of the correlation between asthma control and improved outcomes 90 8 2 ✓ (symptoms, severity, exacerbations, quality of life, etc.) Increasing patient understanding of treatment and lifestyle choices and the impact upon control 89 11 0 ✓ Setting of functional goals with the patient (e.g. to go on a bike ride) 75 25 0 ✓ Improving communication between the patient and healthcare professional 86 14 0 ✓ Improving patient education, including materials which can be shared with the patient 81 19 0 ✓ Use of media outlets including social media to communicate the importance of asthma control 52 44 3 X Increasing consultation time 54 43 3 X More effective history and note taking in consultations (e.g. occupational history, social context) 65 32 3 X Asthma control and adherence Better patient education on the need for adherence, even when not symptomatic 83 13 5 ✓ HCP, healthcare professional. 6 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 treatment and can continue to be monitored, in order to improve us­ quantify and distinguish a patient’s reliever medication use from ability and reduce misunderstanding between the patient and HCP. controller medication use, and simple frequency of reliever medication GINA documents state that assessment of both symptom control and use may not be appropriate as a measure of asthma control. future risk of adverse outcomes should be described separately [3]. We There are several strengths and limitations to this study. A strength suggest that, if a patient has good symptom control, but has an increased of the Delphi study is that experts with extensive experience in asthma future risk of adverse outcomes, they should not be classified as management across different healthcare systems, cultures, and lan­ “well-controlled”, as to be at future risk indicates they require treatment guages were recruited using strict criteria; thus, their individual re­ modification and should be classified as well-controlled with future risk sponses and opinions reflect current knowledge and perceptions within based on therapeutic goals. the field of asthma, which were transformed into a group consensus To develop a universally accepted definition of asthma control that [13]. In addition, sample size was relatively large, providing represen­ also meets patients’ needs, it is important to recognize that patient needs tative information and limiting any potential influence of the authors on differ from those of HCPs, thus HCPs need to fully understand factors of the consensus opinion (authors constituted a minority of the sample asthma control that are important to the patient. Although there may be size). A limitation of the Delphi study design is that there is no univer­ some degree of overlap in patient and HCP perceptions of asthma con­ sally accepted proportion of panelists indicative of consensus [13]. In trol, they are not consistently matched [17], with patients potentially this study, a proportion of ≥66% agreement was considered to be a having their own specific personal meanings and measures to gauge consensus, per a recently published Delphi consensus study of inhaled their individual asthma control [8]. Therefore, to facilitate an accurate therapy in asthma experts [21]. A further limitation is that concerns and objective assessment of asthma control, future measures of control about the frequency of as-needed inhaled corticosteroids plus should consider the patient’s perspective and recognize the potential for rapid-onset β2-agonist as reliever and the assessment of symptom control patient under-reporting of symptoms [18]. Important factors to consider were not covered in this study. The results of the Delphi survey were include levels of control among patients in clinical practice, which may made more robust by comparing with the results of a comprehensive, be significantly lower than in clinical trials [4], and patients may not structured literature review. A limitation of the literature review is the recognize their symptoms as indicators of poor control [4]. potential for bias, which was not assessed. A qualitative study of 55 adults with asthma reported that patients evaluate their own level of asthma control by considering their level of 5. Expert opinion on future goals of asthma control and respiratory symptoms, exacerbations, activity limitations, fatigue, conclusions panic/fear relating to symptoms, sleep disturbances, and use of rescue medication [17]. While the panelists agreed that these measures of It is the authors’ expert opinion that: asthma control are considered by both HCPs and patients when assessing asthma control in clinical practice, this does not occur in commonly used ➢ There is a need for a universally-accepted measure of asthma control assessments of control, and they suggested misalignment between pa­ to permit alignment between clinical practice, clinical studies, and tients’ most important measures of control and those of HCPs. The patient/HCP understanding. The lack of a universally accepted panelists suggested that increasing patient understanding of asthma measure of asthma control should be acknowledged and addressed control could improve outcomes. This could be achieved through greater by HCPs. communication and patient–doctor relationships, setting of functional o Expert response: While the findings of the current study may not goals, and discussion of the importance of treatment and lifestyle be unexpected, it is critical that this is acknowledged so that we choices. Since the panelists concurred that consultation time should not can move forward. A greater collaboration between international be increased, potential ways to educate patients and establish functional thoracic societies could help promote a universally accepted goals whilst ensuring optimized use of the time available in consulta­ measure of asthma control, utilized across clinical trials, and in tions should be identified, for example the use of educational videos in which HCPs could have confidence. Another interesting target for the waiting room before the patient’s visit. It may also be appropriate to future development in the clinical management of asthma is include patients with asthma in future research to help define and create minimal clinically important difference (MCID), defined as the measures of asthma control. smallest change or difference in an outcome measure that is Another key insight gained from this study is that the impact of perceived as beneficial and would lead to a change in the patient’s treatment adherence on asthma control is not considered as standard in medical management. An expert consensus report on literature either clinical trials or in validated measures of asthma control. evidence about MCID for endpoints to monitor asthma control has Although it is possible that patients can be well controlled and non- been published [22]. adherent, lack of therapeutic compliance can cause poor control. In ➢ HCPs should be mindful that patients may have their own measures clinical practice, non-adherence can be high, with low adherence asso­ to gauge their asthma control, and should recognize the potential for ciated with a higher risk of exacerbations [19]. Evaluation and patients under-reporting symptoms. Treatment adherence and measuring of adherence should be more greatly recognized as an inhaler competence should be evaluated and measured when important factor in well-controlled asthma [19]. When providing the assessing a patient’s level of control. patient with further information and education on the need for o Expert response: Since patients did not contribute to the current well-controlled asthma, as highlighted within this study’s results, it study, we were only able to include HCP interpretations of patient could be appropriate to provide the patient with information and edu­ values. Future studies, with help from patient advocacy groups, cation on treatment adherence, even when they believe they are not should seek to understand asthma control domains that are experiencing symptoms with asthma. A further issue highlighted by this important to patients and their concerns around outcomes, as well study with regard to treatment is the disparity in use of reliever medi­ as compare existing measures or develop and validate new mea­ cation as a measure of asthma control. In our study, per available sures of asthma control which encompass both the patient’s measures/definitions [3,10], panelists were asked to quantify SABA perception of their asthma control as well as adherence and reliever medication use indicative of poorly-controlled asthma in clin­ inhaler competence. As improved asthma medication adherence is ical practice. However, over-use of SABA therapy is associated with an a mediator of asthma control, it is also important to match a pa­ increased risk of exacerbations [20]. GINA has recently recommended tient’s control measures to outcomes important to the patient (i.e., as-needed inhaled corticosteroid plus rapid-onset long-acting β2-agonist a patient-centered approach), and to address real-world diffi­ as preferred reliever therapy [3]. As this recommendation becomes culties in measuring medication adherence in patient care. implemented in clinical practice, it may become more complex to 7 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 ➢ A future validated measure of asthma control should use a stan­ were GSK employees at the time of study conduct and hold shares in dardized numerical system to indicate whether or not asthma is well- GSK. controlled. CRediT authorship contribution statement There are disparities in the assessment of asthma control among specialists, and between clinical practice and research. Although outside Giorgio Walter Canonica: Conceptualization, Methodology, Formal the scope of this study, there is a need for future research, collaboration, analysis, Investigation, Data curation, Writing – review & editing. and recommendations to develop a universal metric to measure asthma Antonio Spanevello: Conceptualization, Methodology, Formal anal­ control in clinical practice. ysis, Investigation, Data curation, Writing – review & editing. Luis Pérez de Llano: Conceptualization, Methodology, Formal analysis, Investi­ Ethics approval and consent to participate gation, Data curation, Writing – review & editing. Christian Domingo Ribas: Conceptualization, Methodology, Formal analysis, Investigation, This study complied with all applicable laws regarding participant Data curation, Writing – review & editing. John D. Blakey: Conceptu­ privacy. No direct participant contact, or primary collection of indi­ alization, Methodology, Formal analysis, Investigation, Data curation, vidual human subject data occurred. Study results were in tabular form Writing – review & editing. Gabriel Garcia: Conceptualization, Meth­ and aggregate analyses omitted participant identification, therefore odology, Formal analysis, Investigation, Data curation, Writing – review informed consent, ethics committee, or Institutional Review Board & editing. Hiromasa Inoue: Conceptualization, Methodology, Formal approval were not required. analysis, Investigation, Data curation, Writing – review & editing. Margareth Dalcolmo: Conceptualization, Methodology, Formal anal­ Trademarks and copyright ysis, Investigation, Data curation, Writing – review & editing. Dong Yang: Conceptualization, Methodology, Formal analysis, Investigation, Trademarks are owned by or licensed to their respective owners Data curation, Writing – review & editing. Soniya Mokashi: Concep­ (ACT [QualityMetric Inc.], SERMO [Sermo, Inc.]). tualization, Methodology, Formal analysis, Investigation, Data curation, Writing – review & editing. Abhishek Kurne: Conceptualization, Funding declaration Methodology, Formal analysis, Investigation, Data curation, Writing – review & editing. Aman Kapil Butta: Conceptualization, Methodology, This was a collaborative study funded by GSK. SM, AK, and AB were Formal analysis, Investigation, Data curation, Writing – review & GSK employees at the time of study conduct and hold shares in GSK and, editing. as per all authors, were involved in the study conception/design, data acquisition, and data analysis and/or interpretation, and in writing/ References critical review of draft versions of this manuscript and approval of the final version to be submitted for publication. Delphi questionnaire [1] C.E. Jia, H.P. Zhang, Y. Lv, R. Liang, Y.Q. Jiang, H. Powell, et al., The asthma control test and asthma control questionnaire for assessing asthma control: panelists received honoraria compliant with GSK’s fair market evalua­ systematic review and meta-analysis, J. Allergy Clin. Immunol. 131 (2013) tion for each country. Panelists were blinded to the name of the study 695–703. sponsor until completion of the first questionnaire. Medical writing [2] J.D. Blakey, A.J. Wardlaw, What is severe asthma? Clin. Exp. Allergy 42 (2012) 617–624. support for the development of this manuscript, under the direction of [3] Global Initiative for Asthma, Global strategy for asthma management and the authors, was provided by Rosie Robson MSc., of Ashfield Med­ prevention, Available at: https://ginasthma.org/wp-content/uploads/2022 Comms, an Inizio company, and was funded by GSK. /05/GINA-Main-Report-2022-FINAL-22-05-03-WMS.pdf, 2022. (Accessed 4 May 2022). [4] D. Price, M. Fletcher, T. van der Molen, Asthma control and management in 8,000 Author disclosures European patients: the REcognise Asthma and LInk to Symptoms and Experience (REALISE) survey, NPJ Prim. Care Respir. Med. 24 (2014), 14009. [5] L. Prieto, C. Badiola, J.R. Villa, V. Plaza, J. Molina, E. Cimas, Asthma control: do A. Spanevello reports consulting fees from GSK; payment or hono­ patients’ and physicians’ opinions fit in with patients’ asthma control status? raria from GSK, Chiesi and AstraZeneca; and participation on a data J. Asthma 44 (2007) 461–467. safety monitoring board or advisory board for GSK. L. Pérez de Llano [6] H.K. Reddel, D.R. Taylor, E.D. Bateman, L.P. Boulet, H.A. Boushey, W.W. Busse, et al., An official American Thoracic Society/European Respiratory Society statement: reports grants or contracts from AstraZeneca, Esteve, FAES Pharma, and asthma control and exacerbations: standardizing endpoints for clinical asthma Teva, consulting fees from AstraZeneca, GSK, Novartis, GEBRO, Gilead, trials and clinical practice, Am. J. Respir. Crit. Care Med. 180 (2009) 59–99. MSD and Sanofi; payment or honoraria from AstraZeneca, GSK, [7] P.M. O’Byrne, C. Jenkins, E.D. Bateman, The paradoxes of asthma management: time for a new approach? Eur. Respir. J. 50 (2017), 1701103. Novartis, Chiesi, Sanofi, Esteve, MSD and Leo-Pharma; support for [8] N. Bidad, N. Barnes, C. Griffiths, R. Horne, Understanding patients’ perceptions of attending meetings and/or travel from Novartis, FAES Pharma, Chiesi, asthma control: a qualitative study, Eur. Respir. J. 51 (2018), 1701346. GSK and AstraZeneca; and participation on a data safety monitoring [9] National asthma education prevention program, Expert panel report 3 (EPR-3): board or advisory board for AstraZeneca. C. Domingo Ribas has received guidelines for the diagnosis and management of asthma-summary report 2007, J. Allergy Clin. Immunol. 120 (2007) S94–S138. funding for travel or speaker fees from ALK, Almirall, AstraZeneca, [10] E.F. Juniper, P.M. O’Byrne, G.H. Guyatt, P.J. Ferrie, D.R. King, Development and Boehringer-Ingelheim, Chiesi, Esteve, Ferrer, GSK, Menarini, Novartis, validation of a questionnaire to measure asthma control, Eur. Respir. J. 14 (1999) Stallergenes, Takeda and Pfizer, and declares no specific conflicts of 902–907. [11] R.A. Nathan, C.A. Sorkness, M. Kosinski, M. Schatz, J.T. Li, P. Marcus, et al., interest to report regarding this paper. J.D. Blakey reports grants or Development of the asthma control test: a survey for assessing asthma control, contracts from AstraZeneca, Novartis and GSK; consulting fees from J. Allergy Clin. Immunol. 113 (2004) 59–65. GSK, Chiesi and Boehringer Ingelheim; payment or honoraria from [12] G. Hoskins, B. Williams, C. Jackson, P.D. Norman, P.T. Donnan, Assessing asthma control in UK primary care: use of routinely collected prospective observational AstraZeneca, GSK and Chiesi; support for attending meetings and/or consultation data to determine appropriateness of a variety of control assessment travel from GSK, Boehringer Ingelheim and AstraZeneca; receipt of models, BMC Fam. Pract. 12 (2011) 105. medical writing from GSK and Teva; payment to their institution for [13] F. Hasson, S. Keeney, H. McKenna, Research guidelines for the Delphi survey technique, J. Adv. Nurs. 32 (2000) 1008–1015. advisory work from Asthma Australia; and, unpaid advisory work from [14] E.D. Bateman, H.A. Boushey, J. Bousquet, W.W. Busse, T.J. Clark, R.A. Pauwels, et Asthma WA. H. Inoue declares research grants from Boehringer- al., Can guideline-defined asthma control be achieved? The Gaining Optimal Ingelheim, payment or honoraria for lectures and advisory committees Asthma ControL study, Am. J. Respir. Crit. Care Med. 170 (2004) 836–844. [15] K.R. Chapman, D. Hinds, P. Piazza, C. Raherison, M. Gibbs, T. Greulich, et al., from AstraZeneca, Boehringer-Ingelheim, Kyorin, GSK, Novartis and Physician perspectives on the burden and management of asthma in six countries: Sanofi. G.W. Canonica, D. Yang, M. Dalcolmo and G. Garcia declare that the Global Asthma Physician Survey (GAPS), BMC Pulm. Med. 17 (2017) 153. they have no disclosures of interest. S. Mokashi, A. Kurne, and A. Butta 8 G.W. Canonica et al. Respiratory Medicine 202 (2022) 106942 [16] G.L. Colice, Categorizing asthma severity: an overview of national guidelines, Clin. [20] B.I. Nwaru, M. Ekström, P. Hasvold, F. Wiklund, G. Telg, C. Janson, Overuse of Med. Res. 2 (2004) 155–163. short-acting β2-agonists in asthma is associated with increased risk of exacerbation [17] K. Rudell, A. Hareendran, N. Bonner, R. Arbuckle, C. Burbridge, L. Abetz, Patients’ and mortality: a nationwide cohort study of the global SABINA programme, Eur. experience of asthma control and clinical guidelines: perspectives from a Respir. J. 55 (2020), 1901872. qualitative study, Respir. Med. 106 (2012) 909–911. [21] A.L. Valero, J.A. Trigueros, V. Plaza, Multidisciplinary consensus on inhaled [18] Y.A. Alzahrani, E.A. Becker, Asthma control assessment tools, Respir. Care 61 therapy in asthma, Expet Rev. Respir. Med. 15 (2021) 425–434. (2016) 106–116. [22] M. Bonini, M. Di Paolo, D. Bagnasco, I. Baiardini, F. Braido, M. Caminati, et al., [19] M. Engelkes, H.M. Janssens, J.C. de Jongste, M.C. Sturkenboom, K.M. Verhamme, Minimal clinically important difference for asthma endpoints: an expert consensus Medication adherence and the risk of severe asthma exacerbations: a systematic report, Eur. Respir. Rev. 29 (156) (2020 Jun 3), 190137. review, Eur. Respir. J. 45 (2015) 396–407. 9