The Burden of Self-Reported Rhinitis and Associated Risk for Exacerbations with Moderate-Severe Asthma in Primary Care Patients

Introduction

Asthma and rhinitis are both heterogenous chronic respiratory conditions, with prevalence rates increasing worldwide.^1,2 Asthma affects ~30 million people in Europe¹ and over 5 million people in the United Kingdom,³ with allergic rhinitis (AR), the most common form of non-infectious rhinitis, affecting up to 32% of people in Europe and 29% of people in the United Kingdom.^2,4 If uncontrolled, AR has a substantial negative impact on peoples’ lives, imposing a high socioeconomic and health burden on individuals and society.^5–9 Asthma and AR frequently coexist,¹⁰ with up to 80% of asthma patients having AR and over 40% of AR patients having asthma.¹¹ The links between asthma and AR have been well characterised,¹² and can be regarded as manifestations of a single syndrome in two contiguous parts of the respiratory tract.¹⁰ Both diseases share similar triggers and pathophysiology.⁴ AR usually precedes asthma and is an independent risk factor for the subsequent development of asthma.^13–15

Beyond the symptomatic burden of uncontrolled rhinitis in asthma patients, significant uncontrolled rhinitis is a major trigger of asthma symptoms¹⁶ and predictor of poor asthma control,^17–21 which in turn is a risk factor for exacerbations,¹⁶ thereby further compounding its impact and socioeconomic burden.^18–24 In patients with asthma across various healthcare settings, the presence of uncontrolled rhinitis is associated with reduced asthma-related quality of life (QOL)^18–22 and productivity,²³ more physical limitations^21,22 asthma medication use,^20,23,24 increased rate of asthma-related general practitioner visits,^23–25 absences from work,²³ emergency department visits^24,25 and hospitalisations²⁴ compared to those without rhinitis. Furthermore, asthma patients with more severe rhinitis have worse asthma control than those with mild disease,¹⁷ and an elevated risk of hospital readmission due to asthma.²⁶

The Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines highlight the importance of the link between asthma and AR, and suggest that patients with asthma are assessed for AR, and vice versa, so that management of symptoms can be optimised.⁴ Despite a global strategy for asthma management¹⁶ and evidence-based guidelines for AR management,⁴ recommendations are not fully applied in clinical practice.^27–29 From the physician’s perspective, establishing a diagnosis of AR is becoming more demanding,^12, and AR management is often difficult due to polysensitisation,³⁰ the presence of both allergic and non-allergic disease elements (ie, mixed rhinitis)³¹ and entangling by phenotypes such as severe chronic upper airways disease.³² From the patient’s perspective, AR symptoms are often trivialised,^33,34 and management is less than ideal due to the high level of self-diagnosis, self-selection of medication, and subsequent suboptimal or inappropriate self-management of rhinitis.^34–38 This has far-reaching ramifications for patients with asthma, and adds to the need of ensuring that these individuals are identified, diagnosed and treated appropriately. While patients are known to trivialise AR symptoms, it appears that healthcare professionals (HCPs) might also trivialise and under-recognise this condition in patients with poorly controlled asthma, which may lead to untreated rhinitis and overtreatment of asthma;^8,37–40 thus, highlighting the importance of this study in primary care where most asthma in managed.

This study was developed in order to provide more data on the burden of rhinitis in asthma patients seen in general practice as there is a dearth of research regarding the prevalence and nature of patient-reported rhinitis and its relationship with the risk of exacerbations. We hypothesised that the majority of self-reported rhinitis among primary care patients with a doctor diagnosis of asthma is undiagnosed as previously described,^39,40 and that significant rhinitis is associated with risk of asthma exacerbations. This research aimed to use the iHARP (initiative Helping Asthma in Real-life Patients) database⁴¹ to achieve the study objectives which were to (i) determine the prevalence, severity and treatment of self-reported rhinitis symptoms among adults aged ≥18 years with asthma treated at Global Initiative for Asthma (GINA) Step 3 and above and (ii) compare the demographics, clinical characteristics, medication-taking behaviours and HCP review by between patients who report rhinitis symptoms and those who do not and (iii) determine whether patient-reported rhinitis is associated with risk of asthma exacerbations in the total sample.

Methods

Data Source

This research used the iHARP database, an international database containing anonymised patient data obtained from practices receiving the iHARP asthma review service.⁴¹ Data were collected prospectively between June 2011 until December 2014 from enrolled primary care practices in Australia and 7 European countries (United Kingdom (UK), Italy, Spain, France, the Netherlands, Norway and Sweden).

Ethics approval for this asthma review service was acquired from each participating site according to country-specific requirements. In the UK, data were collected as part of quality improvement and became part of the clinical record, whereas participants in all other countries provided signed informed consent. This cross-sectional observational study was registered with the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCEPP) (as ENCePP/SDPP/9651).

Inclusion and Exclusion Criteria

Patients had to meet the following inclusion criteria for the asthma review service: they had to be 16 years and older, have a current diagnosis of asthma and were prescribed at least 12 months of fixed-dose combination inhaled corticosteroid and long-acting beta₂ agonist therapy (ie, two prescriptions of GINA Step 3 or above therapy) in the year before the review. Patients were excluded if they had a diagnosis of any other chronic respiratory disease other than asthma, or they had received oral corticosteroids and/or antibiotics for a lower respiratory condition in the 2 weeks before the review, or they had received maintenance systemic therapy for asthma. These exclusions were specified to minimise confounding on disease stability and control by additional treatments. The study cohort from the iHARP database for the current analysis was limited to adult patients (≥18 years old) who were from practices in the above 7 European countries.

Study Data and Assessments

In the iHARP asthma review, comprehensive information about patients’ demographics, asthma symptoms, rhinitis symptoms, inhaler technique, medication use, side-effects, occurrence of exacerbations and HCP review were collected using practitioner-led and questionnaire-led assessments. The questionnaire that was used is available in publications listed on the iHARP website⁴¹ and included the following sections:

Rhinitis symptoms were assessed using a single question based on the ARIA and International Primary Care Respiratory Group definition of rhinitis;^4,42 ‘Do you have any of these symptoms: itchy, runny, blocked nose or sneezing when you don”t have a cold?’, with responses ranging from, 0 (“no”), 1 (“occasionally and little bother”), 2“occasionally but quite a bother”), 3 (“most days but little bother”), or 4 (“most days and a lot of bother”). Patients were classified as having mild rhinitis if their responses were ‘1ʹ or ‘3ʹ and as having moderate-severe rhinitis if their responses were ‘2’ or ‘4’. Patient-reported intranasal corticosteroid (INCS) and/or oral antihistamine use for the treatment of rhinitis in the past year were also recorded.

The GINA criteria at the time were used to assess the level of asthma control during the week before the review visit.⁴³ Patients were asked if they experienced: daytime symptoms (more than twice/week); any night wakening due to asthma; need for short-acting β-agonist (SABA) reliever (more than twice/week) and any limitation in daytime activity. The level of asthma control was defined as controlled (none of the above), as partially controlled (1 or 2 of the above) or as uncontrolled (3 or 4 of the above). Patients were also asked to rate their level of asthma control in response to the question; “In the past 4 weeks, did you believe that your asthma was ‘well controlled’?” with “yes” or “no” response options.

Adherence to asthma preventer therapy was assessed with a single question from the REALISE (Recognise Asthma and Link to Symptoms and Experience) study,^44,45 asking patients to choose a statement that best described how they take their preventer asthma medication. Patients were also asked to indicate the highest number of puffs of SABA reliever taken in one day during the previous 4 weeks with ‘0–4’ or “5–12 or more” response options.

Patients were asked whether their inhaler technique had been checked by a HCP in the past year and whether they had seen a respiratory specialist for asthma outside the practice with response options “in the previous year”, “more than a year ago” or “never”. Patient-reported oropharyngeal effects while inhaling preventer treatment (ie, feeling a sensation at the back of the throat, a need to cough and/or of medication being deposited at the back of the throat) and side-effects from preventer asthma medication (ie, continual sore mouth/throat, oral thrush, bruising, hoarse voice, cough and abnormal weight gain) were recorded during the review.

Patients reported whether they had experienced any asthma exacerbations in the previous 12 months before the review visit. Exacerbations were identified by one of the following events: (a) hospital admission with breathing or chest problems; (b) emergency department (ED) visit due to asthma; or (c) a short burst (5–10 days) of oral corticosteroids for worsening asthma. The total number of exacerbations in the previous year (ie, a+ b+ c) were calculated for each patient; a total score of zero = no risk of exacerbations, and a total score ≥1 = yes risk of exacerbations.

Data Analysis

Statistical analyses were performed using SPSS (IBM^® SPSS^® Statistics) Version 24. There were two parts to the statistical analysis plan:

Part 1: Descriptive statistics were used to summarise sample characteristics. Demographics, clinical characteristics, behaviours regarding medication use and review by HCPs of patients who reported rhinitis symptoms were compared to those that did not. Normally distributed continuous variables such as age were compared using Student’s t-test, and for categorical variables such as gender and smoking status the Pearson chi-square test was used.

Part 2: An exploratory analysis was performed to identify variables associated with risk of exacerbations among the total patient sample using univariable logistic regression models, where the risk of exacerbations (no/yes) was the dependent variable and patient demographics, clinical characteristics, behaviours regarding medication use and HCP review were the predictor variables. Tolerance values were used to test for intercorrelations among predictor variables, and those with high intercorrelations were not included for further analysis. Predictor variables that were significantly associated with risk of exacerbations were then entered into a multivariable logistic regression to predict the risk of exacerbations using a threshold for entry into the model of p < 0.05 and interactions between covariates were tested. The goodness of fit of the logistic regression model was confirmed by the Hosmer and Lemeshow test. A two-tailed significance level of 0.05 was used for all statistical procedures.

A sample size of at least 818 participants was needed for the multiple regression analysis based on a candidate set of 10 independent variables, a small effect size (0.02 ƒ²) with 80% power and a probability level of 0.05.

Results

Patient Cohort

There were 4274 adults aged ≥18 years with asthma in the iHARP database who had their asthma reviewed between June 2011 and December 2014 and all were included in the analysis. The study cohort included 2574 (60.2%) from the UK, 652 (15.3%) from the Netherlands, 527 (12.3%) from Spain, 403 (9.4%) from Italy, 62 (1.5%) from Sweden, 36 (0.8%) from Norway and 20 (0.5%) from France. Table 1 shows patient characteristics: mean (SD) age of patients 50.9 (14.3) years; 60.8% female; 33.1% obese (body mass index ≥30 kg/m²); 12.9% current smokers and 30.3% had GINA-defined controlled asthma.

Table 1 Patient Demographics and Characteristics Overall and by Patient-Reported Rhinitis

Prevalence and Severity of Coexisting Rhinitis Symptoms and Treatment

Overall, 2881/4274 (67.4%) patients reported rhinitis symptoms; however, of those reporting rhinitis symptoms only 34.3% (987/2881) had a doctor diagnosis of rhinitis (Table 1), ie, 65.7% (1894/2881) of patients reporting rhinitis had not received a doctor diagnosis (Figure 1). Of patients reporting rhinitis symptoms, 63.5% (1829/2881) were classified as having mild rhinitis and 36.5% (1052/2881) having moderate-severe rhinitis. In the past year, 12.4% (358/2881) of patients indicated using intranasal corticosteroid (INCS) therapy for rhinitis and 19.8% (569/2881) oral antihistamines.

Figure 1 Incidence of diagnosed and undiagnosed rhinitis among patients reporting rhinitis symptoms (N=2881).

Characteristics of Patients Reporting Coexisting Rhinitis Symptoms

Patients reporting rhinitis symptoms were significantly more likely to be younger, female, have a normal body mass index and have a positive smoking history than those who did not (Table 1). The proportion of patients with GINA-defined uncontrolled asthma was significantly higher among those who reported rhinitis symptoms than those who did not (26.8% vs 21.0%, p <0.001) (Table 1); and even higher among patients with moderate-severe rhinitis than those with mild rhinitis (31.4% vs 24.2%, p <0.001) (Figure 2).

Figure 2 Asthma symptom control status in patients with mild (N=1829) and moderate-severe rhinitis (N=1052).

In the week before the review visit, a significantly higher proportion of patients who reported rhinitis symptoms experienced asthma symptoms for ≥3 days, had asthma symptoms affecting their sleep or had symptoms that interfered with daily activities than those who did not (Table 2). While there was no significant difference in frequency of SABA reliever use between the two groups, patients who reported rhinitis symptoms were significantly more likely to have taken 5–12 puffs or more of SABA reliever on at least 1 day in the previous 4 weeks than those who did not (12.8% vs 9.5%, p=0.002) (Table 2). Furthermore, compared to patients who did not report rhinitis symptoms, those who did were more likely to have experienced exacerbations requiring short-term courses of oral steroids (38.8% vs 30.2%, p <0.001), have taken time off work/education (10% vs 16.6%, p<0.001) or visited the ED due to asthma (9.8% vs 7.6%, p <0.001) in the 12 months prior the review visit (Table 2). Patients with GINA-defined partially controlled/uncontrolled asthma who reported rhinitis were more likely to accurately perceived that their asthma was “not well controlled” than those with asthma alone (40.9% vs 32.7%, p<0.001).

Table 2 Indicators of Asthma Symptoms and Exacerbations Overall and by Patient-Reported Rhinitis

Patients who reported rhinitis were significantly less likely to take preventer inhaler every day and to have had their inhaler technique reviewed by a HCP in the past year than those who did not (Table 3); yet they were significantly more likely to experience oropharyngeal effects while inhaling preventer treatment and side-effects from preventer asthma therapy symptoms than those who did not (Table 3). In addition, patients who reported rhinitis symptoms were significantly less likely to have seen a respiratory specialist for asthma in the past year (vs more than a year ago/never) than those who did not (Table 3).

Table 3 Clinical and Behavioural Characteristics Overall and by Patient-Reported Rhinitis

Association of Rhinitis with Risk of Asthma Exacerbations

Results of exploratory analysis using univariable logistic regression identified moderate-severe rhinitis as one of the 10 variables significantly associated with the risk of exacerbations (Table 4). All 10 variables were included in a multivariable logistic regression as there were no correlations between the univariate predictors. The multivariable logistic regression model was statistically significant (χ² = 122.61, df = 12, p< 0.001), accounting for 16.9% (Nagelkerke R squared) of the overall variance. As reported in Table 5, moderate-severe rhinitis was significantly associated with risk of asthma exacerbations (OR=1.40, 95% CI: 1.02–1.90). Other significant factors associated with exacerbation risk included: GINA-defined uncontrolled asthma (OR=2.67, 95% CI: 1.76–4.04); not using preventer therapy every day (OR=1.96, 95% CI: 1.27–3.03); oropharyngeal effects while inhaling preventer treatment (OR=1.93, 95% CI: 1.32–2.83); side-effects from preventer therapy (OR=1.82, 95% CI: 1.17–2.83); high-dose of SABA use (OR=1.83, 95% CI: 1.18–2.82); diagnosis of GORD (OR=1.83, 95% CI: 1.20–2.80); specialist review for asthma more than a year ago/never (vs previous year) (OR=1.77, 95% CI: 1.30–2.41) and obesity (OR=1.57, 95% CI: 1.16–2.12). Smoking status was not associated with risk of exacerbations. There were two significant interactions: between GINA-defined asthma control and oropharyngeal effects; and between preventer adherence and side-effects (Table 5). The risk of exacerbations among patients with controlled/partially controlled asthma increased if they experienced oropharyngeal effects, and similarly among preventer medication adherers if they experienced side-effects.

Table 4 Univariable Associations Between Patient Characteristics and Risk of Asthma Exacerbations

Table 5 Logistic Regression Predicting Likelihood of Asthma Exacerbations

Discussion

This research using the data from a large cross-sectional observational study investigated patient-reported rhinitis and its association with risk of asthma exacerbations in a large cohort of adults with asthma treated at GINA Step 3 and above who are managed in primary care. Our study revealed: (1) the prevalence of patient-reported rhinitis was higher than that of doctor-diagnosed rhinitis, with undiagnosed and/or poorly managed rhinitis highly prevalent in this cohort; (2) asthma patients with moderate-severe rhinitis were more likely to have GINA-defined uncontrolled asthma than those with mild rhinitis or no rhinitis and (3) moderate-severe rhinitis is associated with 40% increased risk of asthma exacerbations.

This cross-sectional observational analysis detected a major discrepancy between patient-reported and doctor-diagnosed rhinitis using data collected from the iHARP asthma review, which would have been difficult to detect via other approaches such as general practice databases or medical records used in other studies.^18–26 Our results showed that concomitant rhinitis is highly prevalent in patients who are considered to have moderate-severe asthma; however, the fact that rhinitis is likely to be undiagnosed and/or poorly managed suggests that it may be more difficult for HCPs to diagnose rhinitis among patients with asthma, and/or is being overlooked by both patients and HCPs as a factor that can impair asthma control even with the most effective asthma therapy being prescribed.^{12,21,39,40,46} While asthma patients are known to underestimate the seriousness of their asthma,⁴⁴ our findings suggest that they also underestimate the severity of their coexisting rhinitis, with only 1 in 3 adults indicating symptoms of moderate-severe rhinitis, which is inconsistent with their asthma severity,^10, and rates reported in previous studies.^19,20,47 This research is novel as it for the first time, takes a real-world/pragmatic approach to understand the relationship between rhinitis and asthma. Research to date has almost exclusively evaluated the impact of having doctor diagnosed AR on asthma exacerbations.^24,25 The current research focused on patient-reported rhinitis. Therefore, our study enables us to identify potential confounding effects of undiagnosed or undocumented rhinitis in studies reporting on the impact of diagnosed rhinitis on asthma outcomes. This study uncovered the nature and extent of concomitant rhinitis in adults with asthma by using data collected via a structured asthma review approach, highlighting the need for such approaches to be an integral part of asthma consultations in primary care.

Our study indicated that rhinitis has a negative impact on asthma symptom control. This finding confirms previous reports showing a negative effect of rhinitis on the level of asthma control assessed by the Asthma Control Questionnaire among adult asthma patients with a diagnosis of rhinitis or documented allergen sensitisation,^18–20 and that patients with moderate-severe rhinitis had the worst asthma control.^17,48 In our study patients with rhinitis were less likely to use their preventer medication every day, yet were more likely to report side-effects from asthma therapy. While these findings seem contradictory, they may be explained by several factors: first, regular use of inhalers has shown to be associated with maintenance of correct inhaler technique (ie, practice makes perfect), thus irregular use of inhalers is associated with a deterioration in inhaler technique over time;⁴⁹ second, patients reporting rhinitis in our study were less likely to have had their inhaler technique reviewed by a HCP in the past year – inhaler technique deteriorates over time and correct technique needs to be reinforced regularly, thus these patients are more likely to have poor inhaler technique;⁵⁰ third, poor inhaler technique increases the risk of local side-effects such as dysphonia,⁵¹ and when side-effects occur, this reduces patient’s willingness to use their inhalers regularly, which in turn becomes a vicious cycle, and fourth, since the majority of patients with rhinitis were not using optimal rhinitis therapy, our findings suggest that rhinitis with an inflamed mucosa might be a risk factor for local side-effects from preventer inhalers. This study also brings to light an interesting concept concerning the patient’s ability to differentiate between coexisting upper and lower airway symptoms. Given that individually, worsening asthma symptoms and symptomatic episodes of rhinitis can cause night-time awakenings, breathlessness and cough,^12,33 and impact on the ability to undertake activities,^7,52 these findings question whether asthma patients are able to distinguish between asthma symptoms, asthma and rhinitis symptoms or just rhinitis symptoms alone when they respond to GINA-defined criteria/questions that are being asked for an objective evaluation of asthma control. These findings identify a gap in the management of rhinitis in general practice and the need for practitioners to identify, evaluate and optimally treat rhinitis in patients with asthma and better patient education concerning the link between rhinitis and asthma outcomes.

Interestingly, patients with suboptimal asthma control who reported rhinitis symptoms were more likely to accurately perceive that their asthma was “not well controlled” than patients with asthma alone. This finding suggests that these patients are more likely to voice their concerns and report troublesome asthma symptoms, search for asthma-related information, education or care; this flags that while they recognise their asthma control is suboptimal, they may fail to recognise the coexistence of both chronic conditions,¹² are unaware of the link between asthma and rhinitis,³⁷ underestimate the magnitude of the impact of rhinitis on their daily lives^33,37,38 and often accept living with nasal symptoms,^37,53 and hence, do not complain or volunteer to provide information about “trivial” nasal symptoms unless their physician asks.^37,38 Primary care providers need to “prompt” patients by questioning them about the frequency, duration and impact of nasal symptoms on their daily lives when they are seeking asthma care. Our findings suggest the need to utilise simple validated assessment tools suitable for primary care, that capture both asthma and rhinitis symptom control, such as the Control of Allergic Rhinitis and Asthma Test (CARAT), which concurrently assesses symptoms of the upper and lower airways, sleep disturbances, limitation in normal activities and the need to adjust medication over a four-week period,⁵⁴ or the MASK (Mobile Airways Sentinel Network)-air mobile phone application, which is specific for monitoring rhinitis in asthma and includes the CARAT.⁵⁵

Importantly, this study found that moderate-severe rhinitis is associated with 40% increased risk of exacerbations among patients with asthma, despite GINA Step 3 and above maintenance therapy being prescribed. This finding emphasises the need to ensure that rhinitis symptoms are optimally controlled in patients dually affected by asthma and rhinitis. Optimal treatment of the upper airway may improve lower airway-specific symptoms,^56,57 reduce the need for rescue medication,^23,56 improve rates of controlled asthma and reduce future risk of adverse events.⁵⁸ The finding that current smoking was not linked to increased risk of exacerbations in our study is inconsistent with previous reports among patients seen in general practice.¹⁷ A possible explanation for current smoking not to be statistically significant in the multivariable regression model may be due to a lack of power as only 12.9% of our patient cohort were current smokers. Two factors associated with exacerbation risk identified in our study, which have not been reported in the global strategy for asthma management,¹⁶ include side-effects of preventer therapy and oropharyngeal effects while inhaling preventer treatment. Risk of exacerbations doubled in patients experiencing side-effects and/or oropharyngeal effects from inhaled preventer medication; this risk increased even in patients with controlled/partially controlled asthma if they experienced oropharyngeal effects and in preventer medication adherers if they experienced side-effects. Oropharyngeal side-effects suggest poor inhaler technique which we know is related to poorer outcomes, such as poor asthma symptom control and increased risk of asthma exacerbations.^51,59,60 Identification of the above three risk factors also suggests that these are potentially issues that could also be addressed by community pharmacy. Community pharmacists when dispensing asthma medication to people with asthma should check inhaler technique, adherence and oropharyngeal effects/side-effects of asthma medication, and recommend optimal rhinitis therapy, which, in some countries is available over-the-counter without prescription.

The strengths of the study include its large-scale size, observational design, evidence from real-life clinical practice and utilisation of a structured face-to-face asthma review approach that included objective and patient-reported outcomes. Furthermore, this approach provided insight into patient behaviour such as inhaler technique that would have been difficult to capture via other approaches such as online surveys used in other studies. The study focused on those patients who had recent general practice contact who were prescribed combination preventer therapy (ie, GINA Step 3 and above); this being the most commonly used maintenance therapy for asthma at the time in primary care. There are several potential limitations to this study, which are related to the cross-sectional study design and a reliance on patient recall for self-reported exacerbations and medication taking behaviour, particularly with oral steroid use for asthma that may have been under- or over-stated by patients, or the confounding effect of oral steroid use for other conditions such as chronic rhinosinusitis. Other study limitations relate to patient recruitment over several countries, which may have led to greater variability in responses and the exclusion of patients with mild asthma (treated at GINA Step 1 or 2). While there were criteria to exclude patients with a diagnosis of chronic obstructive pulmonary disease (COPD) and given that most patients were 50 years of age or older with a past or current smoking history (and a greater risk of COPD), there is a possibility that patients with both asthma and COPD were included and could have biased the results. It is likely that most rhinitis identified in this study was indeed allergic in origin, but this cannot be confirmed due to the absence of data relating to the sensitisation status of the study population. Finally, while significant risk factors were identified, the overall variance accounted by the multivariable logistic regression model was low, suggesting other clinical, attitudinal or behavioural characteristics not regarded in the current analyses may relate to exacerbation risk and the need for further research in this area.

Conclusion

This study identified a major gap in the diagnosis and management of rhinitis in a cohort of people with asthma treated at GINA Step 3 and above who are managed in general practice. It highlights the need for practitioners to identify, evaluate and optimally treat rhinitis in adults with asthma, which is a significant factor associated with exacerbation risk.