Article Text
PDF
Abstract
Objective To evaluate the burden of asthma in five European countries (5EU; France, Germany, Italy, Spain and United Kingdom [UK]).
Design A retrospective cross-sectional study was conducted based on the data from the 2018 National Health and Wellness Survey. Health-related quality of life (HRQoL), work productivity and activity impairment, and healthcare resource utilisation (HCRU) were compared between different groups: asthma versus non-asthma, mild/moderate/severe asthma versus non-asthma and moderate/severe asthma versus mild asthma.
Settings Internet-based survey across Western Europe.
Participants Adult patients (aged ≥18 years) with self-reported physician diagnosis of asthma and experienced asthma symptoms in the past 12 months.
Outcome measures Socio-demographic characteristics, asthma-related outcomes, HRQoL and productivity, HCRU and prevalence of asthma.
Results The prevalence of asthma in the 5EU was 6.7% (95% CI: 6.5% to 6.9%), with the UK reporting the highest rates (10.4%; 95% CI: 9.9% to 10.9%). About 52.0% of the respondents had mild asthma, 27.9% had moderate and 20.1% had severe asthma. The asthma group reported significantly poorer HRQoL, higher rates of overall work productivity impairment and activity impairment, and a greater number of visits to emergency room, healthcare provider and hospitalisations versus the non-asthma group (all p<0.001). Similar trend was observed for all outcomes among respondents with moderate or severe versus mild asthma.
Conclusion Asthma prevalence and burden are still high in Western Europe, indicating the need for effective interventions that could lead to improved outcomes.
- Quality of Life
- EPIDEMIOLOGY
- Patient Reported Outcome Measures
- Hospitalization
- Asthma
Data availability statement
Data are available upon reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
This National Health and Wellness Survey study data can be generalised to broader asthma patients in the five European countries.
Asthma was self-reported by respondents with no data on confirmed asthma diagnosis; self-reporting can introduce recall bias, misclassification and incorrect reporting of asthma diagnosis.
Study used validated self-reported scales that were designed to minimise the self-reported biases and memory errors.
Patients with severe disease may not have responded.
Introduction
Asthma is a chronic respiratory disease that affects 339.4 million people worldwide.1 2 A cross-sectional World Health Survey of adults from 70 countries reported that the prevalence of asthma in adults ranged from 0.2% in China to 21.0% in Australia.3 The 2008 National Health and Wellness Survey (NHWS) reported the overall prevalence of asthma in adults across the five European countries (5EU; United Kingdom [UK], Germany, Spain, France and Italy) as 6.1%.4 Asthma imposes a substantial burden on patients, impairs health-related quality of life (HRQoL),5–7 results in decreased physical activity,8 sleep disturbances,9–11 limited socialisation,12 decreases in work productivity, limitations in daily activities and increased healthcare resource utilisation (HCRU) and associated costs,5–7 particularly among those with severe asthma.13–16 Furthermore, use of oral or systemic corticosteroids is associated with side effects.17 18 In the UK, a 33% rise in asthma deaths (>12,700) was observed during the period from 2008 to 2018, due to suboptimal management of asthma.19
Despite well-established standards of care, >50% of the treated asthma patients report suboptimal or uncontrolled disease that is associated with the increased risk of exacerbations and poorer patient-reported outcomes.20 21 Previous studies have evaluated the prevalence of not well-controlled asthma and its effects on HRQoL, work productivity and activity impairment (WPAI) and HCRU in Europe.4 22 23 However, epidemiology studies on the prevalence and burden of asthma based on its severity in a large national representative sample from Europe are limited. Therefore, the present study aimed to estimate the prevalence of asthma and self-reported severity distribution in the 5EU as well as to evaluate the burden of illness by measuring HRQoL, WPAI and HCRU in respondents with asthma compared with those without asthma. We also studied the comparative burden between the severity groups.
Methods
Study design and data source
This retrospective cross-sectional study included data (between April 2018 and July 2018) from the 2018 NHWS in the 5EU. The NHWS is a self-administered, internet-based survey of adults (aged ≥18 years) that represents the health status of the general population of a country. The response rate was 8% for the survey. The protocol and questionnaire for the NHWS were reviewed and granted exemption by the Pearl Institutional Review Board (IRB; Indianapolis, IN; IRB Protocol Number: 18-KANT-160). Informed consent was obtained from all the respondents.
Study participants
Survey participants were recruited from a pre-existing general-purpose consumer panel24 using opt-in e-mails, coregistration with panel partners, e-newsletter campaigns, website banner advertisements and affiliate networks. Stratified random sampling was implemented to ensure representativeness across countries with respect to age and gender demographics.24
Asthma group
Respondents aged ≥18 years, with self-reported physician diagnosis of asthma who experienced symptoms in the past 12 months were included. Respondents with asthma using prescription medication were defined as the asthma group. They were further categorised into asthma severity groups based on self-reported disease severity: (1) mild (intermittent, persistent): symptoms occurring ≥2 days/week but not daily, (2) moderate (persistent): symptoms occurring daily and (3) severe (persistent): symptoms occurring throughout the day.
Non-asthma group
There were no inclusion criteria for the general population cohort. The general population group consisted of all adults aged ≥18 years, who reported never experiencing asthma and were included regardless of other health status or comorbidities. The exclusion criteria applied to respondents who met at least one of the following: they have ever been diagnosed with asthma by a physician or have experienced asthma symptoms in the past 12 months. This criterion was implemented to exclude past asthma patients who may be controlled or not currently suffering from asthma or undiagnosed patients with asthma.
Study measures
Socio-demographic characteristics
Age, sex, marital status, education, employment status and annual household income data were collected. Participants reported their body mass index (BMI), smoking status, alcohol use and frequency and comorbidities. The Charlson Comorbidity Index (CCI) was used for estimating the burden of comorbidities in the past 12 months.25
Asthma-related outcomes
Data on the timing of asthma exacerbations, frequency, main causes and episodes when asthma symptoms became worse (need for oral steroids or antibiotics, requiring unexpected hospitalisation or symptoms lasting for days or weeks) were collected. The asthma control test (ACT), a 5-item self-administered questionnaire with a 4-week recall period, was used to evaluate the effect of asthma on daily activities, frequency of asthma symptoms, shortness of breath, rescue medications use and overall asthma control. ACT scores were categorised as ‘not well-controlled (≤19)’, ‘well-controlled (20-24)’ and ‘total control’.26 Respondents reported current use of short-acting muscarinic antagonists and short-acting β−2-receptor agonists for asthma.
Health-related quality of life and productivity
HRQoL was assessed using the Medical Outcomes Study 12-Item Short Form Survey Instrument Version 2 (SF-12v2) questionnaire, consisting of 12 items that evaluate eight health domains and two component summary scores (physical component summary [PCS] and mental component summary [MCS]) to assess generic health outcomes.27 28 Health status was assessed using the EuroQol-5 Dimension – 5 Level (EQ-5D-5L) instrument, a self-reported measure of health for clinical and economic appraisal, that includes EQ-5D index score and the EQ visual analogue scale (EQ-VAS).29 30 The six-dimensional health state short form SF-6D,31 derived from the SF-12v2, was used to assess the health utility scores.32 Impairment of work and non-work activities was assessed using the WPAI-general health (WPAI-GH) questionnaire, a six-item validated instrument.33 Only respondents who were employed reported four metrics on absenteeism (the percentage of work time missed because of one’s health in the past 7 days), presenteeism (the percentage of impairment experienced while at work in the past 7 days because of one’s health), total work productivity impairment (an overall impairment estimate, which is a combination of absenteeism and presenteeism), total activity impairment (the percentage of impairment in daily activities because of one’s health in the past 7 days) was reported by all the respondents.
Healthcare resource utilisation
HCRU outcomes included number of healthcare provider (HCP) visits, number of emergency room (ER) visits and number of hospitalisations in the past 6 months.
Prevalence estimates
The prevalence of asthma, weighted by age and sex, was calculated for the 5EU and for individual countries. For further details, please see online supplemental section S1.
Supplemental material
Statistical methods
All statistical analyses were performed using SPSS Version 23, SAS Version 9.4 or R Version 3.6.1 or higher. Data were analysed using descriptive statistics, unmatched bivariate comparisons and multiple regression models. Descriptive statistics for categorical variables included frequencies and percentages, and for continuous variables included means, standard deviation (SD), medians, interquartile ranges and minimum and maximum values. Bivariate analyses were performed using the χ2 test and one-way ANOVA to examine the associations between potential covariates and dependent study variables. Multivariable analyses were conducted to identify predictors of outcome measures in the asthma group versus the non-asthma group, as well as between asthma severity groups.
Generalised linear mixed models (GLMMs) were used to characterise health outcomes in the asthma group versus non-asthma group and between groups defined by level of asthma severity. We included a random intercept for country to account for the fact that there is clustering within countries. Outcomes included HRQoL (SF-12v2, EQ-5D-5L and SF-6D scores), WPAI-GH, HCRU (past 6 months) and asthma-related outcome measures (asthma exacerbations, ACT and rescue medications). GLMMs specifying normal distribution were used for assessing HRQoL, and those specifying a negative binomial distribution were used for assessing WPAI-GH and HCRU. The GLMMs were controlled for confounders identified a priori based on the literature or those that were significant in the bivariate results. These included age, sex, marital status, education, income, BMI, smoking status, alcohol use and CCI for all comparisons. Time, frequency and cause of asthma exacerbations were considered for comparisons of moderate or severe asthma with mild asthma.
Adjusted marginal means, regression coefficients, p values and confidence intervals (CIs) were calculated. Cohen's d was used as a measure of effect size (ES) and adjusted for use of multilevel modelling. ES allows for an assessment of differences between groups independent of the sample size. ES were classified as small (d=0.2), medium (d=0.5) and large (d≥0.8).34 Negative Cohen’s d scores indicate the direction of the effect and suggest a deterioration in HRQoL outcomes. ES were calculated to present standard differences between marginal means reported in independent asthma severity groups for all comparisons, thus indicating the strength of the effect for each HRQoL domain scores. The non-centrality parameter method was used to estimate the CIs. Statistical significance tests for multiple comparisons, testing outcomes’ associations between severity groups (other than the reference), were adjusted for multiplicity using a Tukey adjustment. After the most parsimonious model was determined, post hoc tests were performed to understand the differences between each level of subgroup. NHWS data does not contain ‘true’ missing values, as values are only missing due to survey skip logic. For instance, respondents who do not use prescription medication for asthma do not answer the related questions. The number of missing values for each variable, if any, was provided as descriptive results. Moreover, the analyses in this study were exploratory in nature and did not require adjustments for multiplicity.
Patient and public involvement statement
Patients were not involved in the design, conduct, reporting or dissemination plans of the presented research.
Results
Asthma prevalence
The total NHWS respondents included 62,000 adults (figure 1). The prevalence of diagnosed asthma in the 5EU was 6.7% (95% CI: 6.5% to 6.9%); the highest prevalence being reported in the UK (10.4%; 95% CI: 9.9% to 10.9%) (online supplemental figure 1a). The majority of the respondents diagnosed with asthma received prescribed treatment (74.4%); the highest proportion were from the UK (83.3%) and lowest in Italy (66.3%) (online supplemental figure 1b). More than half of the respondents (52.0%) using prescription medication for asthma rated their severity as mild, whereas moderate and severe asthma was reported by 27.9% and 20.1% of respondents, respectively (figure 2).
Study sample flow chart. Estimates in the figure are weighted by age and sex of country population; values in parentheses are actual sample sizes from the NHWS. *The asthma group was analysed for the study. 5EU, five European countries (France, Germany, Italy, Spain and United Kingdom); NHWS, National Health and Wellness Survey.
Self-reported asthma severity in the asthma group. Estimates in the figure are weighted by age and sex of country population. 5EU, five European countries (France, Germany, Italy, Spain and United Kingdom).
Bivariate analyses
Socio-demographic characteristics
Respondents with asthma were significantly younger (mean age: 47.4±16.3 vs 49.5±16.6 years, p<0.001) with the majority being female (62.7% vs 52.1%, p<0.001) (table 1; figure 3A) and had a greater comorbidity burden including insomnia (31.0% vs 18.9%) and depression (30.8% vs 15.0%) (table 2; figure 3A). Respondents with severe asthma were older (mean age: 48.8±16.2 vs 46.6±16.5, p=0.007) (table 1; figure 3B) and reported greater burden of comorbidities than those in the mild asthma group; similar results were observed for moderate versus mild asthma groups (table 2; figure 3B).
(A) Demographic and comorbidity differences between asthma and non-asthma respondents; (B) Demographic and comorbidity burden across asthma severity levels.
Socio-demographic characteristics of respondents with asthma versus non-asthma and across asthma severity
General health characteristics of respondents with asthma versus non-asthma and across asthma severity
Asthma-related outcomes
About 20.0% of the respondents with asthma reported experiencing an exacerbation daily and worsening asthma symptoms lasting for days or weeks (19.2%). Furthermore, 32.9%, 19.8% and 9.9% of respondents reported requiring oral steroids, antibiotics and unexpected hospitalisation, respectively, for their asthma exacerbations. On the basis of the ACT scores, 47.5% of respondents with asthma did not have well-controlled asthma, 42.0% had well-controlled asthma and only 10.5% had total control of their asthma (online supplemental table 1).
A higher proportion of respondents in the severe asthma group experienced exacerbations daily versus the mild asthma group (38.2% vs 10.3%, p<0.001). The proportion of respondents with severe asthma requiring oral steroids, antibiotics or unexpected hospitalisation (p<0.001 for all) was higher than those in the mild asthma group (online supplemental table 1). A similar trend was observed for the moderate versus mild asthma group (online supplemental table 1). A higher proportion of respondents with severe (70.1%) and moderate asthma (60.8%) did not have well-controlled asthma (p<0.001 for both severity comparisons) than in the mild asthma group (31.6%). Only about 5% of the severe and moderate group had total control versus 15.3% of the mild asthma group (p<0.001 for both severity comparisons; online supplemental table 1).
HRQoL, productivity and HCRU
Bivariate results are detailed in online supplemental tables 1 and 2. Respondents with asthma had lower SF-12 mental and physical health components and SF-12v2 domain scores as well as lower EQ-VAS and utility scores than those in the non-asthma group (p<0.001 for all; online supplemental table 2). Similarly, lower scores in these outcomes were observed among all three asthma severity groups versus the non-asthma group (p<0.001 for all; online supplemental table 3), which worsened with increasing asthma severity (p<0.001 for all; online supplemental table 2). Furthermore, respondents in the asthma group reported higher mean absenteeism, presenteeism, total work impairment, activity impairment and higher HCRU (ER visits, hospitalisation and HCP visits; p<0.001 for all) versus the respondents in the non-asthma group (online supplemental table 2). An increase in the total work productivity impairment, activity impairment and mean number of HCP visits was observed in all the asthma severity groups versus the non-asthma group (online supplemental table 3). The productivity impairment and HCRU (hospitalisation and HCP visits) increased with increasing asthma severity (online supplemental table 2).
Adjusted analyses
Asthma-related outcomes
On the basis of the ACT scores, the proportion of respondents with not well-controlled asthma was higher in the severe (70.0% vs 43.0%, p<0.001) and moderate asthma groups (63.0% vs 43.0%, p<0.001) vs the mild asthma group (online supplemental table 4). After controlling for covariates, respondents in the asthma group had lower marginal mean scores for HRQoL outcomes than respondents in the non-asthma group: MCS score, PCS score, EQ-VAS score, utility measures: EQ-5D index score and SF-6D utility (p<0.001 for all; online supplemental table 5).
Respondents with severe asthma reported lower marginal mean (SE) scores than the non-asthma group for MCS (42.1 [0.8] vs 45.4 [0.7]), PCS (41.0 [0.5] vs 46.5 [0.3]), EQ-VAS scores (55.3 [1.2] vs 66.0 [0.9]), utility measures such as EQ-5D index score (0.7 [0.01] vs 0.8 [0.01]) and SF-6D (0.6 [0.01] vs 0.7 [0.01]) (p<0.001 for all). The adjusted means for these outcomes were lower in the moderate and mild asthma groups versus the non-asthma group (p<0.001 for all). SF-12v2 domain scores were also lower for all the three severity groups versus the non-asthma group (table 3; figure 4). The ESs for HRQoL mean score comparisons were small (consistently <0.2), and the difference between the severe asthma group versus the non-asthma ranged from −0.07 to −0.14. The ESs for physical health in the asthma group were slightly greater than those observed in other HRQoL domains, indicating a more pronounced impact on physical health compared with the non-asthma group (table 3; figure 4). Furthermore, a decrease in all the HRQoL scores (except the SF-12 Mental Health and Vitality domain scores) was observed in the severe versus the mild asthma group (p<0.001 for all; for MCS: p=0.03; online supplemental table 4). All the HRQoL measures (with the exception of MCS, SF-12 Mental Health and Vitality domain scores) were lower among the respondents with moderate asthma than those with mild asthma (p<0.001 for all). The EQ-5D index scores (mean [SD]) for France, Germany, Italy, Spain and UK were as follows: France, 0.8 (0.1); Germany, 0.8 (0.2); Italy, 0.9 (0.1); Spain, 0.9 (0.1) and UK, 0.8 (0.2). The five mean scores were significantly different across 5EU countries (ANOVA, p<0.001).
Health-related quality of life (SF-12v2 domain) scores by asthma severity. MCS: mental component summary; PCS: physical component summary; SF-12v2: Medical Outcomes Study 12-Item Short Form Survey Instrument Version 2.
Multivariable results for severity groups versus non-asthma group: HRQoL
Work productivity and activity impairment-general health
The marginal means (%) for absenteeism, presenteeism, overall work impairment and activity impairment were higher among the asthma group than the non-asthma group (all p<0.001; online supplemental figure 2a). The severe and moderate asthma groups had higher rates of presenteeism (p=0.001 and p=0.01, respectively), overall work impairment (p=0.002 and p=0.01, respectively) and activity impairment (p<0.001 for both) than the non-asthma group; additionally, the severe asthma group had higher adjusted rates of absenteeism (p=0.02) than the non-asthma group (figure 5). Respondents in the severe and moderate asthma group had higher marginal rates of presenteeism (p<0.001 and p=0.03, respectively), overall work impairment (p<0.001 and p=0.03, respectively) and activity impairment (p<0.001 and p<0.001, respectively) than those in the mild asthma group (online supplemental figure 2b). Respondents with mild asthma had higher adjusted rates of absenteeism (16.4% vs 11.3%, p=0.03) and activity impairment (40.8% vs 34.4%, p<0.001) than the non-asthma group (figure 5).
Comparison of WPAI across severity groups versus non-asthma group. *Calculated for employed individuals. WPAI: work productivity and activity impairment.
Healthcare resource utilisation
The asthma group had a higher mean number of HCP visits, ER visits and hospitalisations than the non-asthma group in the past 6 months (p<0.001 for all; online supplemental figure 3a). The marginal means for HCRU including the number of HCP, general practitioner (GP) and ER visits were higher for respondents in the severe, moderate and mild asthma groups than for the non-asthma group (p<0.001 for all; online supplemental figure 3b). Furthermore, the marginal means for HCP visits were higher in the severe (10.0 vs 9.1, p=0.04) and moderate asthma groups (10.4 vs 9.1, p=0.001) versus the mild asthma group. A similar trend was observed for the number of GP and ER visits with higher marginal means in the severe versus mild (p<0.001 for all) and moderate versus mild (p≤0.01 for all) comparisons (online supplemental figure 3c).
Discussion
In the current study, analyses were based on the asthma severity classified based on the respondents’ self-perception of disease, unlike other studies that were based on asthma control.4 21 22
The NHWS includes a variety of health outcomes measures and validated scales. Results from NHWS studies that include these measures align with non-NHWS studies that have administered these same measures or similar measures. A 5EU NHWS study on caregivers found that their results on the SF-12v2 were consistent with previous non-NHWS studies.35 Another study on epilepsy revealed that direct costs incurred by participants in the 5EU NHWS aligned with estimates from earlier non-NHWS epilepsy studies, including multicentre and observational studies conducted in Italy, Germany and France.36 Additionally, a UK study on asthma patients using data from the 5EU NHWS assessed asthma control via the ACT, finding similar rates of moderate or poor control compared with other non-NHWS UK studies, including a random-digit-dialling survey across multiple European countries.22
GLMMs were selected for this study due to their ability to handle non-normally distributed response variables37 and incorporate both fixed and random effects,38 which is essential for accounting for the complex data structure and variability within the dataset. In our study, GLMMs account for the hierarchical structure of the data, which includes participants nested within different countries. GLMMs allow for the analysis of both fixed effects, such as demographic variables and asthma severity, and random effects, which can capture country-level variations in asthma prevalence and management. By using GLMMs, we aimed to provide more robust estimates of the effects of asthma on HRQoL and healthcare utilisation, while accounting for potential confounders and hierarchical data structure.
To better identify the unmet needs and impact of asthma at a population level, it is important to understand patients’ self-perception of their illness that may help uncover valuable information for physicians, policymakers and healthcare regulators. We found that among the respondents with mild asthma, 31.6% had uncontrolled asthma, showing a potential unawareness of the severity of their disease, which may remain suboptimally treated and would likely be the reason for exacerbations in mild asthma,39 highlighting the importance of asthma education. Similarly, patients with moderate-to-severe osteoarthritis pain who are treated with prescription medications also experience poorer health status and HRQoL, along with increased HCRU, similar to those not on medication.40
We estimated the prevalence of asthma in the 5EU to be 6.7%, which is consistent with the 2008 NHWS data, indicating that the prevalence of asthma remained stable over the last decade. Previous studies from Europe reported varied prevalence ranging from 4.3% to 8.2%3 41; the differences could be attributed to the variation in the study design and definition of asthma. In line with prior research, a higher proportion of respondents with asthma were female,42 obese,43 and suffered from comorbidities such as allergies, hay fever, anxiety, insomnia, depression and atopic dermatitis.4 44 Although the proportion of respondents (47.5%) reporting not well-controlled asthma is lower than that previously reported using 2010 NHWS data (ranging from 55% to 59%),4 the number remains high, indicating the need for effective asthma management in clinical practice. In our study, few respondents with asthma (19.8%) reported the use of antibiotics; the proportion of respondents using antibiotics increased with increasing severity of the disease. A recent observational study reported that the use of antibiotics may be beneficial in a significant proportion of patients with difficult to treat (severely uncontrolled, but not necessarily refractory) asthma.45
In this study, respondents with asthma had significantly impaired mental and physical health including the eight domains of SF-12v2 compared with the non-asthma group, thus showing an impact on every aspect of HRQoL. It should be noted that the non-asthma group was not a healthy control group and they also had lower HRQoL than normative scores of 50. The respondents with asthma reported a major impact on their HRQoL with a difference from normative scores ranging from three points for vitality to nine points for role emotional domains. These results are consistent with published studies that reported lower scores on PCS and MCS.7 46 The current study also demonstrated a significant decrease in HRQoL when each asthma severity group was compared with the non-asthma group. This shows that the impact on patients’ HRQoL is irrespective of their asthma severity, and even patients with mild asthma should be appropriately managed by HCP. These findings are comparablewith other chronic condition, rheumatoid arthritis, where the HRQoL was worst compared with healthy population irrespective of severity.47–49 Whilst comparing severe or moderate asthma with mild asthma, the impairment was significantly worse for all SF-12v2 domains except for Mental Health and Vitality, which signals that these aspects of health were impacted equally across asthma severity and likely to be affected first. The ES comparisons showed that physical health was impaired more than mental health versus the mild asthma patients, possibly demonstrating the long-term impact of asthma on physical health.
The general health status (EQ-VAS) was significantly worse among respondents with asthma compared with the non-asthma group, which is consistent with the previous research.50 Similar results were observed while comparing the non-asthma group and the mild, moderate or severe asthma groups as well as between the severe or moderate asthma group and the mild asthma group. The scores were lower than population norms (range: 70.4–83.3) especially for the severe asthma group (55.3). Furthermore, the health utility scores (EQ-5D index score and SF-6D score) were also significantly lower in the asthma group compared with the non-asthma group, which corroborates previously published studies.7 50 In our study, overall, Spain had the highest health utility score (EQ-5D index) followed by Italy, France and Germany, and the UK had the lowest health utility score among the 5EU countries.
In the current study, respondents with asthma reported 6- to 12-fold greater impact on work and activity impairment compared with the non-asthma group. Our results are in agreement with previous multinational surveys7 51 among adult patients with asthma compared with the non-asthma group.7 It has been reported that work productivity impairment increases with asthma severity and affects day-to-day activities.52 53 Similarly, the findings from the European NHWS 2018 indicate that the impact of hypoglycaemia in individuals with type 2 diabetes varies by severity.54 In the present study, all WPAI-GH measures were higher among respondents with mild, moderate or severe asthma compared with those in the non-asthma group. Likewise, a previously published NHWS survey reported significantly higher levels of absenteeism and presenteeism, with a nearly 50% greater level of overall work impairment as well as activity impairment among patients with mild, moderate or severe asthma compared with matched non-asthma controls.55 Furthermore, respondents diagnosed with asthma reported higher ER visits, HCP visits and hospitalisations than those in the non-asthma group. These findings corroborate the previously published claim data analysis from Germany56 and NHWS from Brazil7 that reported higher HCRU (ER/emergency department visits and hospitalisation) among patients with asthma compared with non-asthma patients. HCRU was significantly higher among respondents with severe, moderate and mild asthma compared with the non-asthma group, which is consistent with a study by Ding et al that reported greater HCRU across the asthma severity groups compared with matched non-asthma controls.55 Consistently, we observed that the HCRU increased with increasing severity of the disease. Severe asthma resulting in increased disease-related HCRU was further emphasised by a claims data analysis from Colombia57 and data analysis of two large database studies from the US and UK.16 It should be noted that the lower mean pulmonologist visit frequency in our study among the asthma groups, stratified by severity though not significant, underlines the direct need for referral of patients with uncontrolled asthma to a specialist.
Limitations
The respondents have self-reported their asthma diagnosis, and hence, it is assumed that there was an accurate diagnosis and an accurate recall of it. Underdiagnosis or misdiagnosis of asthma and recall bias may have affected our estimates of disease prevalence. However, self-reported biases and memory errors were kept to a minimum in the NHWS by using validated self-report scales designed to eliminate these issues. Moreover, the results for the prevalence of asthma align with a previously published study.4
Self-reported asthma diagnosis and severity can introduce recall bias and misclassification, which could have resulted in biased estimates of the differences in outcomes across the groups. Despite efforts to minimise these biases using validated self-report scales, some degree of response bias and memory errors are inevitable in self-reported data.58 Patients' perceived severity may significantly influence the patient-reported outcomes and is likely to be influenced by their individual perceptions of the disease.
Our analysis incorporated socio-economic status measures by adjusting for them in multivariate models to control for potential confounding factors. However, it is important to acknowledge that unmeasured confounding factors, such as exercise habits, levels of family or social support, environmental exposures, life stress, or sleep, may still influence the associations observed in this study. These factors, which may affect asthma outcomes or quality of life, were not included in the data collection, so their potential impact on the study’s findings cannot be fully ruled out.59–62
A further limitation of this survey is the low response rate. This low rate is likely attributed to the online methodology, as some respondents may have limited access to a computer or limited experience with such technology. Additionally, participant demographics, such as older age and lower education levels, may also contribute to a decreased response rate. Despite these limitations, the study employed random and stratified sampling methods to ensure the sample was representative across the 5EU, capturing diverse demographics (5EU data was only obtained from the UK, France, Italy, Germany, and Spain). Inclusion and exclusion criteria for the NHWS were structured to render general population samples that are representative of the various countries in which the survey is fielded. This involves stratified random sampling by age, gender and ethnicity, among other things. Generalisability of NHWS data has been validated and weighted against reliable sources, including government agencies' health statistics and unaffiliated third parties.63
Findings may have limited applicability outside the 5EU, including other European countries, due to cultural and healthcare system differences. Generalising results is globally constrained by these factors, particularly in non-Western settings where asthma management practices may differ. However, due to the representative nature of NHWS data, results from the present study can likely be generalised to the broader asthma patient population in the 5EU. Finally, as this was a cross-sectional study, a causal relationship between variables of interest cannot be assessed.
Conclusions
Asthma is a prevalent condition in Western Europe, and a high proportion of patients have uncontrolled asthma, including those who self-perceive the disease severity to be mild, which may be impacting asthma behaviours as well as correct treatment usage. Asthma continues to have a wide-ranging impact on patients’ daily lives, and their ability to work as well, leads to high utilisation of healthcare resources. This was observed across the mild, moderate and severe asthma groups. These results highlight the burden of asthma and the need for further research to understand it from a patient’s perspective and develop improved interventions, including asthma education and treatment, which consequently reduce the large humanistic and economic burden. Future research focusing on long-term longitudinal studies to track patients over time is needed to enhance our understanding of asthma management and its long-term impacts.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants but the protocol and questionnaire for the NHWS were reviewed and granted exemption by the Pearl Institutional Review Board (IRB; Indianapolis, IN; IRB Protocol Number: 18 KANT-160). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
Editorial support and medical writing support were provided by Sumedha Kulkarni, M. Pharm, and Ashwini Atre, PhD from Indegene Pvt Ltd., India, and funded by Sanofi, France and Regeneron Pharmaceuticals, USA.
References
Footnotes
Contributors AHK, IG and JW conceived and designed the analyses. All the authors contributed to the identification of the relevant analytic approaches including the review of the statistical analyses, interpretation of results and defined the outline of the manuscript. DJ, dMM and BLB conducted the statistical analyses. All the authors critically reviewed and finalised the manuscript. AHK is the guarantor of this work.
Funding This study was funded by Sanofi, France, and Regeneron Pharmaceuticals, USA.
Competing interests AHK is an employee of Sanofi and may hold stock and/or stock options. IG is an employee of Sanofi. JJ-N is a former employee of Sanofi and may hold stock and/or stock options. SK is an employee of Regeneron Pharmaceuticals and holds stock or stock options. The funder didn’t influence the results/outcomes of the study despite author affiliations with the funder. DJ reports funding support from Sanofi for study design, analysis and interpretation. dMM reports funding support from Sanofi for conducting the study. BLB reports funding support from Sanofi for conducting the study. JW reports receiving honorarium from PPD, Banook, Atea and Propero; research grants from Sanofi, Regeneron, Arnold; consultants support for attending meetings and/or travel from ATS; and stock or stock options from Merus Pharmaceutical.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.