Abstract
Background
Personalised medicine targets treatable traits with tailored therapy. Extrapulmonary traits, modifiable and related to asthma, were analysed for their association with hospitalisation rates.
Methods
We conducted an observational study of a general population cohort in Sweden, based on the Respiratory Health in Northern Europe (RHINE) and Global Allergy and Asthma European Network (GA2LEN) studies. Participants completed questionnaires in 2008–2010, including respiratory symptoms, smoking habits and education level. Data were linked with national health registers to obtain information on hospitalisations with asthma.
Results
Of 31 000 participants, 2341 had current asthma. Asthma patients exhibited a higher prevalence of airway comorbidities, such as rhinitis, allergic rhinitis and chronic rhinosinusitis, which were associated with increased hospitalisation rates. Obesity, insomnia and snoring were also significant risk factors for hospitalisation. A clear trend of increasing hospitalisations was observed with a higher number of treatable traits, especially in asthma patients aged <60 years, who had significantly increased odds of hospitalisation when presenting with three or more traits (adjusted OR 1.88, 95% CI 1.03–3.42). Chronic rhinosinusitis contributed the most to hospitalisations (10.7% population attributable fraction), followed by obesity (8.6% population attributable fraction). On average, each increase in number of treatable traits was associated with a 13% increased risk of hospitalisation (HR 1.13, 95% CI 1.02–1.25, p=0.02). Smoking and damp/mould exposure had a negligible contribution to the overall burden of hospitalisations with asthma.
Conclusion
Recognising extrapulmonary traits like obesity and chronic rhinosinusitis is vital in asthma management, as they increase the risk of future hospitalisations.
Shareable abstract
Identifying and addressing treatable traits that significantly influence asthma outcomes is crucial for personalised asthma management, as these traits are associated with a heightened risk of future hospitalisations https://bit.ly/4fMScgo
Introduction
Asthma is a chronic respiratory disease characterised by variable airflow obstruction, airway hyperresponsiveness and underlying inflammation. Despite treatment advances and management strategies, asthma remains a significant public issue, affecting approximately 300 million worldwide [1]. Hospitalisations due to asthma exacerbations are a critical concern, contributing to healthcare costs, morbidity and mortality [2]. Identifying treatable traits ‒ specific, measurable characteristics that can be targeted with therapy ‒ is important for improving patient outcomes [3].
The concept of treatable traits is an emerging paradigm that aims to personalise treatment by addressing the heterogeneity of the disease. Unlike traditional approaches that rely on broad classifications, mainly on severity or phenotypes, the treatable traits strategy focuses on identifying and targeting distinct clinical, physiological, and molecular features that contribute to asthma exacerbations and poor disease control [4]. This approach recognises asthma as a multifaceted disease with diverse underlying (not fully known) mechanisms, necessitating a more tailored treatment plan for each patient [5].
Several traits have been identified as potentially modifiable factors that influence the course of asthma and its associated healthcare utilisation. These include pulmonary, extrapulmonary, and even psychosocial, behavioural and lifestyle factors [6]. Addressing these traits through specific interventions has shown promise in reducing the frequency and severity of asthma exacerbations [7, 8]. However, it is not clear how these traits combine to affect hospitalisations, or which are most influential.
In this study, we aimed to explore the association between treatable traits and hospitalisations in those with asthma. By analysing a cohort of patients with asthma, we seek to identify which traits are most strongly linked to increased risk of hospitalisation.
Methods
This observational, general population cohort study is based on two general population samples, the Swedish cohorts from the Respiratory Health in Northern Europe (RHINE) and Global Allergy and Asthma European Network (GA2LEN) studies, collectively called REGAL (RHINE+GA2LEN). Participants in both cohorts answered similar detailed questionnaires in 2008–2010, covering respiratory symptoms, smoking habits, education level and more [9, 10]. We linked the data with national health registers, including the Swedish Patient Register, to obtain comprehensive health information including hospitalisations.
Cohort and current asthma definition
This study cohort consists of 31 000 REGAL participants that answered questions on asthma. At the time of the questionnaire, participants were adults in Sweden, aged 16–76 years, randomly selected from the general population in Gothenburg, Stockholm, Uppsala and Umeå, as previously described in detail [9, 10]. Current asthma was defined as having either reported an asthma attack in the previous 12 months or currently using asthma medication [11].
Treatable traits
Treatable traits were defined at baseline based on questionnaire reports to identify traits known by the participant. We defined the following to be treatable traits: chronic rhinosinusitis, current smoking, obesity, allergic rhinitis, snoring, insomnia, damp or mould exposure in the home, traffic-related air pollution and lack of exercise.
Chronic rhinosinusitis was defined as reporting at least two of the following symptoms for >12 weeks in the last 12 months: nasal congestion; pressure in the forehead, around the eyes or nose; nasal secretion; and loss of smell, with at least one of the symptoms being nasal congestion or nasal secretion [12]. Obesity was defined as having a body mass index ≥30 kg·m−2, calculated from self-reported height and weight. Snoring was defined as reporting loud snoring at least three times a week. Insomnia was defined as reporting at least one of the following three insomnia symptoms at least three times per week: difficulties falling asleep, waking up frequently and early morning waking. Physical activity was estimated by asking participants how often each week they exercised to the point of breathlessness or sweating. Reporting never to engage in physical activity was defined as having “no physical exercise”.
The other traits mentioned above were defined from single-item yes/no questions. Allergic rhinitis was defined by an affirmative answer to the following question: “Do you have any nasal allergies including allergic rhinitis?”. Traffic-related air pollution from traffic was assessed by two questions: “Do you hear traffic noise in your bedroom?” (where possible answers were: 1) no; 2) yes, a street with little traffic; 3) yes, a street with moderate traffic; and 4) yes, a street with much traffic; responses 3 and 4 were defined as “having troublesome traffic near home”) and “How annoyed are you by fumes from traffic in your residential area?” (where possible answers were: 1) none/a little (none), 2) somewhat, and 3) very much, where reporting “very much” was defined as “having troublesome traffic near home”). Dampness was defined as a positive response to at least one of the following questions: “Have any of the following been noted in your home during the last 12 months: 1) water or moisture damage on indoor walls, floors or ceiling (water damage); 2) dented plastic mats, yellowed plastic mats or blackened parquet (floor dampness); or 3) visible moulds on walls, floors or ceiling (visible mould)?” We included insomnia, damp/mould exposure (previously labelled as water damage) and traffic-related air pollution (previously labelled as troublesome traffic) as treatable traits based on previous studies that demonstrate their association with poor asthma control and their potential modifiability [13, 14]. While not traditionally categorised as treatable traits in clinical guidelines, these exposures and symptoms are modifiable through behavioural and environmental interventions, aligning with the expanded treatable traits framework.
Hospitalisations
Data on hospitalisations were collected from the Swedish Patient Register, which contains information from public Swedish hospitals, including all hospitalisations. Using these data, linked through personal identification numbers, we identified all hospitalisations where asthma was listed among the discharge diagnoses (all-cause hospitalisation) and where asthma was the main discharge diagnosis (asthma-caused hospitalisation), according to the International Classification of Diseases codes J45–J45.9, J46 and J46.0. Hospitalisations in the 3 years before the date the questionnaires were answered and those occurring after it, up to the end of 2019, were grouped separately. Participants with any hospitalisations for asthma in the respective time periods were then defined as having been hospitalised before or after entering the study.
Statistical analysis
All statistical analyses were performed using STATA/MP 18.0 (StataCorp, College Station, TX, USA). Descriptive statistics were calculated for the total cohort, stratified by the presence or absence of current asthma. The frequency of hospitalisations with asthma as a diagnosis was summarised both before and after completion of the baseline questionnaires. Hospitalisations were registered in two ways: 1) any hospitalisation with asthma listed as a diagnosis (all-cause hospitalisation), and 2) hospitalisations with asthma listed as the primary diagnosis (asthma-related hospitalisation). We analysed associations between treatable traits and any hospitalisations in two ways: 1) through logistic regression models (any hospitalisation versus no hospitalisation as the main outcome), and 2) a time-to-event analysis for hospitalisations by the number of treatable traits. This is described further below.
Differences in hospitalisations were assessed by the presence or absence of individual treatable traits and by the number of coexisting treatable traits (grouped as 0, 1, 2 or ≥3 traits). Regression analyses were adjusted for age group (16–40, 40–60 and 60–75 years) and sex.
The time-to-event analysis was conducted using Kaplan–Meier curves to visualise time to first all-cause hospitalisation, stratified by the number of treatable traits. Cox proportional hazards regression was used to examine the association between the number of traits and time to first hospitalisation, adjusted for age and sex. Hazard ratios (HRs) with 95% confidence intervals were reported. The proportional hazards assumption was verified before interpretation.
Population attributable fractions were calculated for each treatable trait in relation to all-cause hospitalisations, based on both trait prevalence and its relative risk of hospitalisation, as previously described [15, 16]. The population attributable fraction represents the estimated proportion of hospitalisations that can be attributable to each trait, assuming all confounders have been adjusted for.
Results
The study included 31 000 participants, of whom 2341 had current asthma. The asthma group had a younger median age (44.0 years) compared with the non-asthma group (45.9 years) and a higher proportion of women (60% versus 54%). Rhinitis and allergic rhinitis were significantly more prevalent in the asthma group, and obesity was also more common (16% versus 10%). All-cause hospitalisation rates were significantly higher among individuals with asthma, both before study entry (5% versus 0.1%) and during follow-up (18% versus 0.7%) (table 1).
TABLE 1.
Population characteristics
| Without asthma (n=28 659) | Current asthma (n=2341) | |
|---|---|---|
| Age (years), median (IQR) | 45.9 (32.1–58.0) | 44.0 (30.7–56.2) |
| Female | 15 395 (53.7%) | 1402 (60.0%) |
| BMI (kg·m−2), median (IQR) | 24.3 (22.1–26.9) | 24.8 (22.5–28.3) |
| Smoking history | ||
| Non-smoker | 16 963 (60.7%) | 1355 (60.0%) |
| Ex-smoker | 7096 (25.4%) | 608 (26.9%) |
| Current smoker | 3903 (14.0%) | 297 (13.1%) |
| Education level | ||
| Primary | 4324 (15.2%) | 375 (16.1%) |
| Secondary | 9971 (35.1%) | 855 (36.8%) |
| University or higher | 14 140 (49.7%) | 1095 (47.1%) |
| Damp or mould exposure in home | 1990 (7.0%) | 166 (7.2%) |
| Traffic-related air pollution | 1076 (3.8%) | 127 (5.4%) |
| No physical exercise | 3039 (10.6%) | 223 (9.6%) |
| Chronic rhinosinusitis | 2002 (7.1%) | 483 (20.9%) |
| Allergic rhinitis | 6437 (23.2%) | 1506 (66.7%) |
| Snoring | 8323 (30.1%) | 796 (35.0%) |
| Insomnia | 5596 (19.9%) | 639 (27.6%) |
| Obesity (BMI>30 kg·m−2) | 2848 (10.1%) | 376 (16.3%) |
| Any hospitalisation 3 years prior | 20 (0.1%) | 117 (5.0%) |
| Any hospitalisation after | 202 (0.7%) | 422 (18.0%) |
| Any asthma hospitalisation 3 years prior | 2 (0.0%) | 15 (0.6%) |
| Any asthma hospitalisation after | 26 (0.1%) | 46 (2.0%) |
Data are presented as n (%), unless otherwise stated. Percentages are calculated among respondents with available data for each variable; therefore, totals may not exactly correspond to the overall column total. IQR: interquartile range; BMI: body mass index.
All-cause hospitalisations
Within the asthma group, several treatable traits were associated with increased hospitalisations. Asthma patients with chronic rhinosinusitis, obesity (body mass index >30 kg·m−2), snoring, insomnia and no physical activity were more frequently hospitalised (table 2). These associations remained significant for chronic rhinosinusitis and obesity after adjustment for age group and sex in logistic regression models, as shown in figure 1.
TABLE 2.
All-cause hospital admissions with asthma diagnosis among people with asthma (n=2341) by presence of treatable traits
| Hospitalisations with asthma | p-value # | p-value ¶ | |
|---|---|---|---|
| No treatable traits (n=185) | 24 (13%) | ||
| Any treatable trait (n=2115) | 388 (18%) | 0.07 | 0.07 |
| Current smoker (n=297) | 48 (16%) | 0.32 | 0.34 |
| Damp or mould exposure in the home (n=166) | 19 (11%) | 0.02 | 0.66 |
| Traffic-related air pollution (n=127) | 27 (21%) | 0.34 | 0.05 |
| No physical exercise (n=223) | 53 (24%) | 0.02 | 0.006 |
| Chronic rhinosinusitis (n=483) | 123 (25%) | <0.001 | <0.001 |
| Allergic rhinitis (n=1506) | 258 (17%) | 0.24 | 0.15 |
| Snoring (n=796) | 173 (22%) | <0.001 | 0.007 |
| Obesity (BMI>30 kg·m−2) (n=376) | 110 (29%) | <0.001 | <0.001 |
| Insomnia (n=639) | 142 (22%) | 0.001 | 0.006 |
Data are presented as n (%). Bold font indicates statistical significance. BMI: body mass index;: each condition compared with all asthma patients without named condition; ¶: each condition compared with asthma patients with no treatable traits.
FIGURE 1.
Hazard ratios for all-cause asthma hospitalisations. Shown are all-cause hospital admissions among participants with asthma (n=2341), by presence of treatable traits, adjusted for sex and age group. The points represent adjusted odds ratios and the horizontal lines indicate the 95% confidence intervals for each treatable trait.
A cumulative effect for multiple treatable traits was also observed: hospitalisations increased with a higher number of coexisting treatable traits. As shown in figure 2, this trend suggests a dose-response relationship. When adjusted for age group and sex, the odds of all-cause hospitalisation rose with the number of traits (for ≥3 traits, OR 1.65, 95% CI 1.02–2.67, p=0.04). Among participants <60 years this association was even stronger (for ≥3 traits, OR 1.88, 95% CI 1.03–3.42), indicating a stronger impact of multiple traits in younger adults (table 3).
FIGURE 2.
All-cause hospital admissions (blue) and asthma-caused hospital admissions (green) among participants with asthma by the number of treatable traits.
TABLE 3.
Hospital admissions among participants with asthma by the number of treatable traits among those <60 years, adjusted for age and sex
| One treatable trait (n=771) | Two treatable traits (n=623) | ≥3 treatable traits (n=721) | p-value# | |
|---|---|---|---|---|
| Hospitalisations | 1.43 (0.78–2.61) | 1.29 (0.70–2.40) | 1.88 (1.03–3.42) | <0.001 |
| Asthma hospitalisations | 0.47 (0.12–1.89) | 0.79 (0.21–3.03) | 1.08 (0.30–3.90) | 0.25 |
Data are presented as odds ratios (95% confidence intervals). Values in bold indicate statistically significant associations (p<0.05).: number of treatable traits analysed as a continuous variable.
The time-to-event analysis further demonstrated that a higher number of treatable traits was associated with a shorter time to first all-cause hospitalisation. After adjustment for age and sex, each additional trait was associated with a 13% increased risk (HR 1.13, 95% CI 1.02–1.25, p=0.02), as seen in figure 3.
FIGURE 3.
Time (in years) to first hospital admission for asthma, by the number of treatable traits, among patients with asthma at baseline. The numbers below the graph indicate how many patients in each treatable trait group remained at risk (i.e. had not yet experienced a hospital admission) at each time interval.
Population attributable fraction analysis indicated that chronic rhinosinusitis contributed most to all-cause hospitalisations (10.7%, range 5.8–15.3%), followed by obesity (8.6%, range 4.1–12.8%) and allergic rhinitis (6.6%, range −5.4–17.2%). Interestingly, smoking and damp/mould exposure were associated with negative contributions to the burden of asthma-related hospitalisations (−2.2% each; table 4).
TABLE 4.
Population attributable factors for hospitalisations for asthma among participants with current asthma at baseline, adjusted for age and sex
| Estimate (range) | |
|---|---|
| Chronic rhinosinusitis | 10.7% (5.8–15.3%) |
| Obesity (BMI>30 kg·m−2) | 8.6% (4.1–12.8%) |
| Allergic rhinitis | 6.6% (−5.4–17.2%) |
| Snoring | 2.9% (−4.7–9.9%) |
| Insomnia | 2.0% (−4.1–7.7%) |
| No physical exercise | 0.5% (−2.7–3.5%) |
| Traffic-related air pollution | 0.1% (−2.1–2.3%) |
| Smoking | −2.2% (−5.4–0.9%) |
| Damp or mould exposure | −2.2% (−4.2– −0.2%) |
Values are presented as point estimates with corresponding minimum–maximum ranges; 95% confidence intervals were not calculated. BMI: body mass index.
Asthma-caused hospitalisations
Among the 2341 individuals with asthma, 46 (2.0%) were hospitalised primarily due to asthma (table 3). Asthma-caused hospitalisations were more common in individuals with certain treatable traits. Asthma patients with chronic rhinosinusitis were more often hospitalised for asthma (3.1%, p=0.05), as were those with obesity (3.5%, p=0.03) and insomnia (3.0%, p=0.04), as illustrated in table 5.
TABLE 5.
Asthma-caused hospital admissions among people with asthma (n=2341) by presence of treatable traits
| Primary asthma hospitalisations | p-value# | p-value¶ | |
|---|---|---|---|
| No treatable traits | 3 (1.6%) | ||
| Any treatable trait | 43 (2.0%) | 0.70 | 0.70 |
| Current smoker (n=297) | 5 (1.7%) | 0.72 | 0.96 |
| Damp/mould exposure in home (n=166) | 4 (2.4%) | 0.66 | 0.60 |
| Traffic-related air pollution (n=127) | 4 (3.2%) | 0.33 | 0.37 |
| No physical exercise (n=223) | 5 (2.2%) | 0.72 | 0.65 |
| Chronic rhinosinusitis (n=483) | 15 (3.1%) | 0.05 | 0.29 |
| Allergic rhinitis (n=1506) | 26 (1.7%) | 0.28 | 0.92 |
| Snoring (n=796) | 20 (2.5%) | 0.22 | 0.47 |
| Obesity (BMI>30 kg·m−2) (n=376) | 13 (3.5%) | 0.03 | 0.22 |
| Insomnia (n=639) | 19 (3.0%) | 0.04 | 0.32 |
Percentages are calculated within each subgroup (e.g. among asthma patients with or without each treatable trait, or with no treatable traits); therefore, totals may not correspond to the overall asthma cohort (n=2341). Data are presented as n (%). Bold font indicates statistical significance. BMI: body mass index; ¶: each condition compared with all asthma patients without named condition; ¶: each condition compared with asthma patients with no treatable traits.
A pattern similar to that observed for all-cause hospitalisations was observed: asthma-related hospitalisation rates increased with the number of coexisting treatable traits. However, the association was not statistically significant in the full asthma cohort (for ≥3 traits, OR 1.62, 95% CI 0.48–5.52).
In individuals <60 years of age, the same trend persisted, although it remained statistically non-significant, probably due to the small number of asthma-caused hospitalisations in this age group (n=31; table 3).
Discussion
In this observational general population study, we found that chronic rhinosinusitis and obesity were the traits that were independently and significantly associated with an increased risk of asthma hospitalisations in all analyses. While some traits, such as smoking, damp/mould exposure in the home, traffic-related air pollution and allergic rhinitis, were frequently observed, they did not independently raise the risk of asthma hospitalisation. Interestingly, these patterns were more pronounced in younger and middle-aged adults. These results highlight the importance of managing comorbid conditions to potentially reduce hospital admissions among people with asthma.
Several recent studies have explored the relationship between treatable traits and several asthma outcomes, including asthma control, risk for exacerbations, quality of life and risk of hospitalisations. For instance, a study by Agusti et al. [5] emphasised the importance of identifying treatable traits to improve asthma management and outcomes. Their findings suggest that addressing traits such as obesity, smoking and sleep apnoea can lead to better asthma control and reduced hospitalisations. Obesity and asthma frequently coexist, with evidence pointing to a bidirectional relationship. Asthma-related complications, including hospitalisations, are often associated with a diminished response to medications in individuals with obesity. Obesity itself is a major disease modifier in asthma, complicating its management and leading to higher hospitalisation rates [17–19]. Physical inactivity is a well-known treatable trait. A systematic review highlighted that sedentary behaviour is linked to more severe asthma symptoms and a higher likelihood of exacerbations, reinforcing the need for regular physical activity as part of asthma management [20]. Inactivity is particularly common in people with severe asthma, possibly due to concerns about triggering exercise-induced bronchoconstriction [21]. This aligns with our findings, which show a link between obesity and hospital admissions. Also, physical inactivity was associated with increased hospital admissions, although this association became non-significant after adjusting for sex and age.
Another study by McDonald et al. [7] also supports the notion that multimorbidity, particularly chronic rhinosinusitis and psychological factors such as insomnia, exacerbate asthma severity and lead to increased healthcare utilisation. With an overall prevalence of up to 65% [22], comorbid chronic rhinosinusitis with or without nasal polyps has been associated with poorer outcomes in patients with asthma, including increased airflow limitation, worsened asthma control, being more prone to exacerbations and reduced health-related quality of life [22, 23]. Insomnia is known to worsen asthma control, increase symptom frequency and contribute to exacerbations, as confirmed by studies highlighting the bidirectional relationship between sleep disturbances and asthma severity [24, 25]. In our study, chronic rhinosinusitis and insomnia were significantly associated with higher hospitalisation rates, supporting the robustness and importance of these traits, although the statistical significance for insomnia was lost after adjusting for confounding factors.
Other traits of interest include environmental factors. For example, a 2020 study highlighted the burden of environmental factors such as damp/mould exposure in homes and traffic-related air pollution on asthma control [26]. However, we found a significant negative association with damp/mould exposure in homes, and traffic-related air pollution was not significantly associated with hospitalisations. These findings suggest that although such exposures may worsen asthma symptoms or control, they may not be severe or consistent enough to increase the risk of hospitalisation. Alternatively, these unexpected results may reflect differences in exposure severity, population characteristics or underlying confounding factors across studies.
It is important to note that traits are not mutually exclusive. The main results of the NOVELTY study showed that the presence of treatable traits evaluated in that study varied widely between individuals, with participants with asthma having a mean of 4.6 treatable traits (with significant variability of ±2.6), and particularly high numbers observed in patients with more severe disease [27]. Notably, most treatable traits were of extrapulmonary origin, a finding supported by a separate study [7]. Our findings build on this knowledge by demonstrating a significant correlation between the number of treatable traits and the likelihood of hospitalisation among people with asthma. Specifically, as the number of treatable traits increases, both the overall and asthma-specific hospitalisation rates rise. This cumulative effect of comorbidities has also been shown for other asthma outcomes, such as increased healthcare utilisation [19, 28].
Multiple treatable traits often coexist in individuals with asthma, underscoring the importance of a comprehensive management approach. Indeed, although our results did not reveal strong independent associations for a number of traits, there was still a clear trend towards increased odds of hospitalisation with a greater number of treatable traits, particularly in younger individuals. We believe this indicates that although individual traits alone may not increase hospitalisations, the presence of multiple traits can sufficiently increase disease burden and, consequently, hospitalisations. This finding contrasts with a previous study in which an increasing number of traits within the behavioural and risk factors domain, such as poor inhaler technique or non-adherence, did not lead to a higher exacerbation risk in severe asthma [7]. However, the results of that study should be interpreted with caution because traits known to be linked to poorer outcomes may not have been fully captured or assessed. While pulmonary traits are central to managing asthma, extrapulmonary traits and psychosocial and behavioural traits can independently worsen asthma control and increase the risk of exacerbations or hospitalisations. Therefore, treating only the pulmonary aspects without addressing other contributing factors may lead to suboptimal outcomes. Addressing all relevant traits through a personalised, holistic approach has been shown to improve patient well-being and reduce healthcare utilisation, emphasising the multifaceted nature of asthma care.
Targeted pharmacological and nonpharmacological therapies for treatable traits are crucial to reducing asthma exacerbations, improving quality of life and preserving lung function. Trials have shown improvements in outcomes such as symptom control and quality of life, with some suggesting reduced hospitalisations [19, 29, 30]. Certain traits, such as being prone to exacerbations, depression and obstructive sleep apnoea, have strong links to exacerbations and hospitalisations [7]. Although some studies propose a hierarchy of traits, there is no evidence-based consensus on the ideal approach [7, 31].
Our findings highlight that while many traits influence hospitalisations, only some show strong associations and should therefore be prioritised. Personalised medicine should focus on traits with significant impacts, using multidimensional assessments to guide individualised treatment [5, 19, 31, 32]. Future research should evaluate targeted interventions to reduce hospitalisations and improve outcomes.
Strengths and limitations
This study has several strengths, including a large general population cohort, detailed questionnaires and a country-wide registration of hospitalisations for >10 years. It offers a comprehensive assessment of treatable traits associated with asthma hospitalisations, highlighting the significant correlations with chronic rhinosinusitis, obesity and insomnia. By focusing on younger and middle-aged adults, the research emphasises the importance of managing comorbidities in these populations to reduce hospitalisation rates. Using multivariate logistic regression strengthens the validity of the findings, contributing valuable insights to the existing literature on asthma management. In addition, this study supports the shift towards personalised medicine, advocating for tailored interventions that address specific traits to improve patient outcomes.
A limitation of our study is the inclusion of insomnia, damp/mould exposure and traffic-related air pollution as treatable traits, which are not traditionally classified as such in current clinical guidelines. However, their established associations with poor asthma control and their potential for modification through behavioural and environmental interventions support their consideration within an expanded treatable traits framework.
Unfortunately, a limitation of this study is that all treatable traits were based on self-report, without objective validation (e.g. body mass index measured clinically or chronic rhinosinusitis confirmed via endoscopy). Although self-reported data have been used extensively in population studies, we acknowledge that this may introduce misclassification bias. Future studies with clinical validation are warranted. Another limitation of this study is that the dataset lacked detailed information on asthma pharmacotherapy, including controller medication use, biologic therapy or adherence data. This precluded classification of treatment intensity (e.g. Global Initiative for Asthma steps) or severity. We acknowledge this as a limitation, as these factors may confound the relationship between traits and hospitalisation risk.
Further to the reliance on self-reported data for defining treatable traits, another limitation of this study is the lack of objective clinical measurements. For example, we did not include pulmonary biomarkers such as blood eosinophil counts or spirometry-based airflow obstruction measures, as these were not available in the RHINE or GA2LEN questionnaire datasets. The observational cohort design restricts the ability to establish causal relationships between traits and hospitalisations. In addition, the study's population may lack diversity in terms of ethnicity and socioeconomic status, affecting generalisability. The definitions of treatable traits could overlap, complicating interpretations, and the study does not account for psychosocial factors (e.g. anxiety or depression) or a comprehensive range of environmental influences (e.g. air quality or allergen exposure). We adjusted for age and sex to avoid adjusting away the mechanisms underlying the associations of the treatable traits.
Overall, this study emphasises the need for a holistic approach to asthma care that considers both pulmonary and extrapulmonary factors.
Conclusion
Chronic rhinosinusitis and obesity significantly increase the risk of asthma hospitalisations, especially in younger and middle-aged adults. In addition, having multiple treatable traits significantly increases hospitalisation risk in people with asthma, particularly among those <60 years. Addressing these modifiable factors through targeted interventions can reduce hospital admissions and improve quality of life, emphasising the need for comprehensive asthma care.
Footnotes
Provenance: Submitted article, peer reviewed.
This article has an editorial commentary: https://doi.org/10.1183/23120541.01312-2025
Conflict of interest: M. Labor has received consulting fees from Pierre Fabre and MSD; honoraria for speaking engagements from MSD, AstraZeneca, Chiesi, Menarini, GSK, Novartis, Boehringer Ingelheim and Sanofi; and reimbursement for attending the ELCC 2025 symposium from MSD; all unrelated to the topic of this manuscript. Outside this work, A. Palm has received personal fees for lectures and educational activities from ResMed, unrelated to the topic of this manuscript. Ö.I. Emilsson has received honoraria for advisory boards and participating in educational activities for AstraZeneca, unrelated to the topic of this manuscript. C. Janson, A. Malinovschi, M. Holm, L. Ekerljung, S-E. Dahlén and L. Modig have nothing to declare.
References
- 1.Global Initiative for Asthma (GINA) . Global strategy for asthma management and prevention. Date last accessed: 29 September 2024. Date last updated: 22 May 2024. https://ginasthma.org/2024-report/
- 2.World Health Organization (WHO) . Asthma. Date last accessed: 29 September 2024. Date last updated: 6 May 2024. www.who.int/news-room/fact-sheets/detail/asthma
- 3.Nurmagambetov T, Kuwahara R, Garbe P. The economic burden of asthma in the United States, 2008–2013. Ann Am Thorac Soc 2018; 15: 348–356. doi: 10.1513/AnnalsATS.201703-259OC [DOI] [PubMed] [Google Scholar]
- 4.Pavord ID, Beasley R, Agusti A, et al. After asthma: redefining airways diseases. Lancet 2018; 391: 350–400. doi: 10.1016/S0140-6736(17)30879-6 [DOI] [PubMed] [Google Scholar]
- 5.Agusti A, Bel E, Thomas M, et al. Treatable traits: toward precision medicine of chronic airway diseases. Eur Respir J 2016; 47: 410–419. doi: 10.1183/13993003.01359-2015 [DOI] [PubMed] [Google Scholar]
- 6.Bafadhel M, McKenna S, Terry S, et al. Acute exacerbations of chronic obstructive pulmonary disease: identification of biologic clusters and their biomarkers. Am J Respir Crit Care Med 2011; 184: 662–671. doi: 10.1164/rccm.201104-0597OC [DOI] [PubMed] [Google Scholar]
- 7.McDonald VM, Hiles SA, Godbout K, et al. Treatable traits can be identified in a severe asthma registry and predict future exacerbations. Respirology 2019; 24: 37–47. doi: 10.1111/resp.13389 [DOI] [PubMed] [Google Scholar]
- 8.Pavord ID, Hanania NA, Corren J. Controversies in allergy: choosing a biologic for patients with severe asthma. J Allergy Clin Immunol Pract 2022; 10: 410–419. doi: 10.1016/j.jaip.2021.12.014 [DOI] [PubMed] [Google Scholar]
- 9.Sommar JN, Ek A, Middelveld R, et al. Quality of life in relation to the traffic pollution indicators NO2 and NOx: results from the Swedish GA(2)LEN survey. BMJ Open Respir Res 2014; 1: e000039. doi: 10.1136/bmjresp-2014-000039 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Johannessen A, Verlato G, Benediktsdottir B, et al. Longterm follow-up in European respiratory health studies – patterns and implications. BMC Pulm Med 2014; 14: 63. doi: 10.1186/1471-2466-14-63 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Burney P, Chinn S, Jarvis D, et al. Variations in the prevalence of respiratory symptoms, self-reported asthma attacks, and use of asthma medication in the European Community Respiratory Health Survey (ECRHS). Eur Respir J 1996; 9: 687–695. doi: 10.1183/09031936.96.09040687 [DOI] [PubMed] [Google Scholar]
- 12.Bengtsson C, Jonsson L, Holmström M, et al. Incident chronic rhinosinusitis is associated with impaired sleep quality: results of the RHINE study. J Clin Sleep Med 2019; 15: 899–905. doi: 10.5664/jcsm.7846 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sauvere M, Lejeune S, Chagnon F, et al. Home exposure to moisture and mold is associated with poorer asthma control in children: CHAMPIASTHMA study. J Allergy Clin Immunol Glob 2025. Jan 17; 4: 100415. doi: 10.1016/j.jacig.2025.100415 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tiotiu AI, Novakova P, Nedeva D, et al. Impact of air pollution on asthma outcomes. Int J Environ Res Public Health 2020; 17: 6212. doi: 10.3390/ijerph17176212 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Miettinen OS. Proportion of disease caused or prevented by a given exposure, trait or intervention. Am J Epidemiol 1974; 99: 325–332. doi: 10.1093/oxfordjournals.aje.a121617 [DOI] [PubMed] [Google Scholar]
- 16.Rockhill B, Newman B, Weinberg C. Use and misuse of population attributable fractions. Am J Public Health 1998; 88: 15–19. doi: 10.2105/ajph.88.1.15 Erratum in: Am J Public Health 2008; 98: 2119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mosen DM, Schatz M, Magid DJ, et al. The relationship between obesity and asthma severity and control in adults. J Allergy Clin Immunol 2008; 122: 507–511.e6. doi: 10.1016/j.jaci.2008.06.024 [DOI] [PubMed] [Google Scholar]
- 18.McLoughlin RF, McDonald VM. The management of extrapulmonary comorbidities and treatable traits; obesity, physical inactivity, anxiety, and depression, in adults with asthma. Front Allergy 2021; 2: 735030. doi: 10.3389/falgy.2021.735030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.McDonald VM, Clark VL, Cordova-Rivera L, et al. Targeting treatable traits in severe asthma: a randomised controlled trial. Eur Respir J 2020; 55: 1901509. doi: 10.1183/13993003.01509-2019 [DOI] [PubMed] [Google Scholar]
- 20.McLoughlin RF, Clark VL, Urroz PD, et al. Increasing physical activity in severe asthma: a systematic review and meta-analysis. Eur Respir J 2022; 60: 2200546. doi: 10.1183/13993003.00546-2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cordova-Rivera L, Gibson PG, Gardiner PA, et al. Physical activity and exercise capacity in severe asthma: key clinical associations. J Allergy Clin Immunol 2018; 6: 814–822. doi: 10.1016/j.jaip.2017.09.022 [DOI] [PubMed] [Google Scholar]
- 22.Bachert C, Bhattacharyya N, Desrosiers M, et al. Burden of disease in chronic rhinosinusitis with nasal polyps. J Asthma Allergy 2021; 14: 127–134. doi: 10.2147/JAA.S290424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Laidlaw TM, Mullol J, Woessner KM, et al. Chronic rhinosinusitis with nasal polyps and asthma. J Allergy Clin Immunol Pract 2021; 9: 1133–1141. doi: 10.1016/j.jaip.2020.09.063 [DOI] [PubMed] [Google Scholar]
- 24.Sundbom F, Malinovschi A, Lindberg E, et al. Insomnia symptoms and asthma control–interrelations and importance of comorbidities. Clin Exp Allergy 2020; 50: 170–177. doi: 10.1111/cea.13517 [DOI] [PubMed] [Google Scholar]
- 25.Rhoads SL, Edinger J, Khatiwada A, et al. The impact of insomnia and depression on asthma control. J Asthma 2024; 61: 1248–1251. doi: 10.1080/02770903.2024.2335367 [DOI] [PubMed] [Google Scholar]
- 26.Hassan M, Davies SE, Trethewey SP, et al. Prevalence and predictors of adherence to controller therapy in adult patients with severe/difficult-to-treat asthma: a systematic review and meta-analysis. J Asthma 2020; 57: 1379–1388. doi: 10.1080/02770903.2019.1645169 [DOI] [PubMed] [Google Scholar]
- 27.Agustí A, Rapsomaniki E, Beasley R, et al. Treatable traits in the NOVELTY study. Respirology 2022; 27: 929–940. doi: 10.1111/resp.14325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.McDonald VM, Simpson JL, Higgins I, et al. Multidimensional assessment of older people with asthma and COPD: clinical management and health status. Age Ageing 2011; 40: 42–49. doi: 10.1093/ageing/afq134 [DOI] [PubMed] [Google Scholar]
- 29.Sarwar MR, McDonald VM, Abramson MJ, et al. Effectiveness of interventions targeting treatable traits for the management of obstructive airway diseases: a systematic review and meta-analysis. J Allergy Clin Immunol Pract 2022; 10: 2333–2345.e21. doi: 10.1016/j.jaip.2022.05.012 [DOI] [PubMed] [Google Scholar]
- 30.Petsky HL, Cates CJ, Kew KM, et al. Tailoring asthma treatment on eosinophilic markers (exhaled nitric oxide or sputum eosinophils): a systematic review and meta-analysis. Thorax 2018; 73: 1110–1119. doi: 10.1136/thoraxjnl-2018-211540 [DOI] [PubMed] [Google Scholar]
- 31.Hiles SA, Gibson PG, Agusti A, et al. Treatable traits that predict health status and treatment response in airway disease. J Allergy Clin Immunol Pract 2021; 9: 1255–1264.e2. doi: 10.1016/j.jaip.2020.09.046 [DOI] [PubMed] [Google Scholar]
- 32.McDonald VM, Higgins I, Wood LG, et al. Multidimensional assessment and tailored interventions for COPD: respiratory utopia or common sense? Thorax 2013; 68: 691–694. doi: 10.1136/thoraxjnl-2012-202646 [DOI] [PMC free article] [PubMed] [Google Scholar]