- Email: [email protected]
Asthma control in patients treated with inhaled corticosteroids and long-acting beta agonists: A population-based analysis in Germany
Respiratory Medicine 118 (2016) 58e64
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Asthma control in patients treated with inhaled corticosteroids and long-acting beta agonists: A population-based analysis in Germany Anke Kondla b, Thomas Glaab b, Riccardo Pedersini c, d, Marek Lommatzsch a, * a
University of Rostock, Department of Respiratory Medicine, Rostock, Germany Boehringer Ingelheim Pharma GmbH & Co. KG, Medical Affairs Respiratory Medicine, Ingelheim, Germany c Kantar Health, Epsom, Surrey, UK d RTI Health Solutions, Barcelona, Spain b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2016 Received in revised form 12 July 2016 Accepted 14 July 2016 Available online 16 July 2016
Background: The prevalence and the characteristics of poor asthma control among adults treated with combinations of inhaled corticosteroids (ICS) and long-acting beta-agonists (LABA) are not completely understood. Methods: Data from adult patients in Germany with self-reported asthma treated with an ICS-LABA combination in the National Health and Wellness Survey (NHWS) were analysed. Patients with wellcontrolled and not well-controlled asthma according to the Asthma Control Test (ACT) score were compared, with respect to socio-demographic characteristics, attitudes, adherence and outcomes. Results: Among the German patients with self-reported asthma (5.2% of the respondents), 16.2% (382 patients) were treated with an ICS-LABA combination and did not report concomitant chronic obstructive pulmonary disease, chronic bronchitis or emphysema. In this subgroup, 55.8% had not wellcontrolled asthma (ACT < 20). ICS-LABA treated patients with not well-controlled asthma were more likely to report emergency visits (16.4% vs. 8.9%), missed more time from work (absenteeism: 12.9% vs. 4.3%), were more impaired while at work (presenteeism: 29.0% vs. 14.9%) and were more likely to be women (69.0% vs. 57.4%), compared with well-controlled patients. There were no significant differences in age, body mass index, smoking, income, education or self-reported adherence between the two groups, but different attitudes regarding the patient-physician relationship. Conclusions: A substantial proportion of patients treated with ICS and LABA had not well-controlled asthma. These patients did not differ from well-controlled patients in terms of education or selfreported adherence, but in terms of their attitudes regarding the patient-physician relationship. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Asthma control Quality of life Inhaled corticosteroids (ICS) Long-acting beta-agonitsts (LABA) Germany
1. Introduction Asthma ranks among the most common chronic diseases globally and affects an estimated 300 million people worldwide [1]. Previous classifications of asthma severity based on lung function impairment have been replaced by classifications based on asthma control. This concept has been adopted in the recommendations of the global initiative for asthma (GINA) (www.ginasthma.org) and in several national asthma guidelines. Accordingly, controlled asthma is defined by minimal or no daytime and night-time symptoms,
* Corresponding author. Abteilung für Pneumologie und Internistische Intensiv€t Rostock, Ernst-Heydemann-Str. 6, medizin, Zentrum für Innere Medizin, Universita 18057 Rostock, Germany. E-mail address: [email protected] (M. Lommatzsch). http://dx.doi.org/10.1016/j.rmed.2016.07.012 0954-6111/© 2016 Elsevier Ltd. All rights reserved.
maintenance of normal activity and little or no need for reliever medication. Asthma control can be measured using questionnaires such as the Asthma Control Questionnaire (ACQ) or the Asthma Control Test (ACT) [2]. Large surveys showed that approximately 50% of all patients with asthma are not well controlled [3e7]. Poor asthma control is associated with increased risk of exacerbations, impaired quality of life, absences from school or work and increased health-care resource utilisation [8,9]. A minority of patients with asthma remain symptomatic despite treatment with an inhaled corticosteroid (ICS) and a long-acting beta-agonist (LABA). Although this subpopulation accounts for the majority of medical resource use [10], there is little information on the prevalence and the characteristics of poor asthma control in patients treated with ICS-LABA combinations. It was the aim of this analysis, therefore, to investigate this issue using a population-based
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
Abbreviations ICS LABA ACQ ACT NHWS BMI MMAS-4 MMAS-8 GINA SF-12 SF-36 MCS PCS WPAI COPD EHWS INSPIRE
Inhaled corticosteroid Long-acting beta-agonist Asthma Control Questionnaire Asthma Control Test National Health and Wellness Survey Body mass index 4-item Morisky Medication Adherence Scale 8-item Morisky Medication Adherence Scale Global Initiative for Asthma 12-item Short Form Survey 36-item Short Form Survey Mental Component Summary Physical Component Summary Work Productivity and Activity Impairment Chronic Obstructive Pulmonary Disease European National Health and Wellness Survey Investigating New Standards for Prophylaxis in Reducing Exacerbations LIAISON InternationaL cross-sectIonAl and longitudinal assessment on aSthma cONtrol
approach in Germany. In contrast to previous analyses describing the whole population of patients with asthma [3e5], we focused our analysis on the sub-population of patients treated with ICS-LABA combinations to explore patient characteristics associated with poor asthma control in subjects treated with ICS-LABA combinations.
59
composition of the NHWS sample is representative of each country's adult population. The NHWS was reviewed and approved by Essex Institutional Review Board (Lebanon, NJ, USA). 2.2. Patient characteristics The following socio-demographic and health characteristic variables were included for description of the study population: self-reported physician-diagnosed asthma, age, gender, marital status, education, income, employment type, health insurance, body mass index (BMI), smoking behaviour and self-reported asthma characteristics such as duration of the disease, frequency of asthma symptoms and severity with and without prescribed asthma medication. According to the guidelines of the World Health Organization (WHO), the following BMI categories were used: underweight to normal weight (<18.5 to <25 kg/m2), overweight (25 to <30 kg/m2) and obesity (30 kg/m2). 2.3. Asthma control The Asthma Control Test (ACT) [11] is a patient self-administered tool for the assessment of asthma control, composed of 5 items, with a 4-week recall on symptoms and daily functioning. The scores range from 5 (poor control of asthma) to 25 (complete control of asthma). An ACT score 20 indicates well-controlled asthma, a score of less than 20 indicates not well-controlled asthma. 2.4. Healthcare resource use Healthcare resource use was assessed by the self-reported number of physician visits, emergency visits and hospitalisations during the past six months. Total numbers were considered (not limited to asthma symptoms).
2. Methods 2.5. Health-related quality of life 2.1. Data source Data were extracted from the 2010, 2011 and 2013 German National Health and Wellness Survey (NHWS; not conducted in Europe in 2012) and deduplicated, i.e., only answers from the respective first year were included for any respondent participating several times in the survey. The NHWS was a self-administered, Internet-based questionnaire from a nationwide sample of adults that was stratified by gender and age to represent the demographic composition of the German adult population. Potential respondents to the NHWS were recruited primarily through an existing web-based consumer panel (Lightspeed LLC®, Washington, USA). Online recruitment was supplemented by telephone recruitment to allow better representation of older adults (65 years of age or older), who were less likely to be a member of the consumer panel. Since the survey did not allow skipping of questions reported here, there were no missing data. In the NHWS, each respondent had to complete all relevant items in the questionnaire to be included in the final data, with some items such as height, weight, and income providing an option to decline to answer. Potential respondents to the NHWS were recruited through an existing, general-purpose (i.e. not health carespecific) web-based consumer panel. The consumer panel recruited its panel members through opt-in e-mails, co-registration with panel partners, e-newsletter campaigns, banner placements and affiliate networks. All panellists explicitly agreed to be a panel member, registered with the panel through a unique e-mail address and completed an in-depth demographic registration profile. A stratified random sampling procedure, with strata by age and gender, was implemented to ensure that the demographic
Short form health surveys were used to assess generic Healthrelated Quality of Life (HR-QoL). The 2011 respondents completed the revised SF-12, and 2013 respondents completed the longer revised SF-36 scale. From these data, Physical and Mental Component Summaries (PCS and MCS) and health utility scores were derived. Physical Component Summary (PCS), an index of overall physical health status (population standard mean ¼ 50), Mental Component Summary (MCS), an index of overall mental and emotional health status, and the Short Form 6D health state (SF-6D) health utility score, a preference-based index describing overall health on a scale from 0 (equivalent to death) to 1 (equivalent to perfect health) were used. 2.6. Adherence The 4-item Morisky Medication Adherence Scale (MMAS-4) is a generic self-reported behaviour scale on medication adherence. It consists of four items with dichotomised scoring (“Yes” ¼ 1 and “No” ¼ 0) [12e14]. The items are summed to give a range of 0e4, corresponding with high (0) to medium (1e2) to low adherence (3e4). The MMAS-8 is the more recent and expanded version of MMAS-4. It consists of eight items, with yes/no-responses for the first seven and a 5-point Likert response for the last item. The items are summed and scores range from 0 to 8, indicating high (0) to medium (1e2) to low adherence (3e8). The MMAS-4 was included in the 2011 and 2012 surveys, while the MMAS-8 was included in 2013. To allow comparison, the MMAS-8 in this study was scored to proxy the 4-item measure. This scoring provides values similar to those of the MMAS-4.
60
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
2.7. Attitudes Patients' attitudes towards medications and doctors were measured with 26 statements. These statements dealt with general health issues and were not asthma-specific. Each self-reported statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree). 2.8. Work productivity and activity impairment (WPAI-9) The WPAI [15] is a 6-item validated instrument that consists of four metrics: absenteeism, presenteeism, overall work impairment and activity impairment (see: http://www.reillyassociates.net/ WPAI_General.html for further coding and scoring details). Absenteeism is the percentage of work time missed because of one's health in the past seven days. Presenteeism is the percentage of impairment experienced while being at work in the past seven days because of one's health. Overall work impairment is a combination of absenteeism and presenteeism, and activity impairment is the percentage of impairment in daily activities due to one's health in the past seven days. Only full-time, part-time, or selfemployed respondents provided data for absenteeism, presenteeism, and overall work impairment. All respondents provided data for activity impairment. 2.9. Statistical analysis Data were analysed using STATA v13.1. First, not well-controlled and well-controlled patient groups were compared using Chisquare tests (for categorical variables) or two-tailed t-tests (for cardinal variables). Probability values of p < 0.05 (two-tailed) were regarded as statistically significant. The results of the bivariate comparisons above were then used to choose the variables to be included in a series of regression models describing the association between asthma control and health outcomes, while controlling for other covariates and accounting for the hierarchical structure of the data (age and gender strata). The association with HR-QoL was modelled with linear regression, the association with WPAI and healthcare resource use with negative binomial regression and adherence with binary logistic regression.
with ICS-LABA combinations. There were more females among patients with not well-controlled asthma (69.0% vs. 57.4%, p ¼ 0.019) (Table 1). 3.3. Symptoms Frequency of asthma-related symptoms differed significantly between well-controlled and not well-controlled patients treated with ICS-LABA combinations (Table 2). Pharmacological therapy was associated with a significant reduction of symptom severity in both groups (Fig. S1, online supplement). 3.4. Health status and healthcare resource use Results of the SF-12 and SF-36 scores showed significant impairment in health-related Quality of life (HR-QoL) in patients with not well-controlled asthma as compared with well-controlled asthma (for both the physical (PCS) and the mental (MCS) component summary scores)(Fig. 2). The results were confirmed by the linear regression models (Table S1). Likewise, health utility scores were greater among those patients with well-controlled disease. However, this difference was not maintained after controlling for covariates, suggesting that socioeconomic variables (specifically gender and household income) may have a stronger association with health utility than asthma control in our sample. In this context, we found that males and patients with higher-income tend to have significantly higher health utility scores. Doctor visits (GP and specialists) during the 6 months preceding the survey did not differ significantly between the two groups (Fig. 2). Likewise, there was no significant difference in patients' perceived need for regular physician consultations. In contrast, patients with not well-controlled asthma reported more emergency visits (16.4% vs. 8.9%, p < 0.05; 0.25 vs. 0.12, p < 0.10) and hospital visits (14.6% vs. 8.3%, p < 0.10; 0.20 vs. 0.08, p < 0.05) in the previous six months compared with well-controlled patients (Fig. 2 and S2). These significant differences were not confirmed by negative binomial modelling, suggesting that household income, the presence of comorbidities and adherence to medication have a stronger association with resource use. Patients with lower income showed a
3. Results 3.1. Population sample The combined 2010, 2011 and 2013 German NHWS comprised a total of 44,895 respondents (Fig. 1). Response rates were 27.7% in 2010, 18.2% in 2011 and 14.5% in 2013. Of all respondents, 5.2% (N ¼ 2350) reported to have physician-diagnosed asthma. In this group, 31.3% (N ¼ 735) were treated with ICS plus LABA (free or fixed-dose ICS-LABA combination). Of the ICS-LABA treated patients, 52.0% (N ¼ 382) reported no concomitant COPD, chronic bronchitis or emphysema and constituted the sample of the current study. This subgroup comprising 16.2% of all participants with selfreported, physician-diagnosed asthma (N ¼ 382) was analysed in the present study (Fig. 1). According to their ACT scores, 169 (44.2%) respondents had well-controlled asthma and 213 (55.8%) not wellcontrolled asthma (Table 1). 3.2. Socio-demographic characteristics There were no differences in terms of marital status, education, income, employment, regular smoking behaviour, BMI, or age between well-controlled and not well-controlled patients treated
Fig. 1. Flow diagram of data analysis. Among all German NHWS respondents, subjects with self-reported, physician-diagnosed asthma treated with an ICS-LABA combination and not reporting COPD, chronic bronchitis or emphysema were identified (data set for analysis). This group was subdivided into respondents with wellcontrolled and not well-controlled asthma.
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
61
Table 1 Subject characteristics. ICS-LABA treated respondents
Age (years) Gender Married/in a relationship Education Yearly income
Employed full/part time BMI (kg/m2)
Smoking regularly
Well-controlled (N ¼ 169)
Not well-controlled (N ¼ 213)
Mean (SD) Female Yes Up to high school College/University Less than V20 k V20 to <40 k V40 to <75 k V75 or more No answer Yes Under-Normal Overweight Obese No answer Yes
N
%
N
%
47.78 (14.82) 147 136 176 37 61 68 38 12 34 120 74 61 73 5 59
69.01% 63.85% 82.63% 17.37% 28.64% 31.92% 17.84% 5.64% 15.96% 56.34% 34.74% 28.64% 34.27% 2.35% 27.70%
45.60 (14.19) 97 110 125 44 36 52 38 18 25 109 58 53 54 4 36
57.40% 65.09% 73.96% 26.04% 21.30% 30.77% 22.49% 10.65% 14.79% 64.50% 34.32% 31.36% 31.95% 2.37% 21.30%
p-value
0.145 0.019* 0.802 0.085 0.288
0.106 0.942
0.151
Percentages refer to the total number of patients in each group. Significant differences between the groups (p < 0.05), measured with chi-square tests (categorical variables) or t tests (continuous variables), are marked with an asterisk.
significant underuse of healthcare resources, whereas patients with more comorbidities or stronger adherence to pharmacological treatment displayed an increased use of health care resources. 3.5. Work productivity and activity impairment Significantly higher work and activity impairment scores were reported for patients with not well-controlled versus wellcontrolled asthma: time missed from work (absenteeism; 12.9% vs. 4.3%), impairments while at work (presenteeism; 29.0% vs. 14.9%), overall work impairment (36.4% vs. 17.3%), and activity impairment (42.8% vs. 22.3%) were all statistically significantly increased among patients with not well-controlled asthma (all pvalues < 0.01) (Fig. 2). These results were confirmed by a series of negative binomial regressions, which also showed that males and patients with higher household incomes had a higher work productivity. 3.6. Adherence and patient attitudes Self-reported medication adherence did not differ significantly between groups: 55.4% of well-controlled patients and 60.4% of not well-controlled patients reported high levels of adherence as measured by MMAS-4 and -8. The result was confirmed by a logistic regression analysis showing that adherence tended to increase with age. By contrast, several patient attitudes differed significantly between the groups, including the willingness to continue with a prescribed medication (Table 3). Well-controlled patients reported
a significantly higher level of attention by their physician than not well-controlled patients (Table 3). Well-controlled patients were also more likely to inform their doctor of all the over-the-counter medications they took and less likely to rely on the opinion of friends and family on which medication to use (Table 4). 4. Discussion The factors associated with not well-controlled asthma despite ICS-LABA treatment are still not completely understood. Utilising a population based survey approach across three years in Germany, we found a high prevalence (55.8%) of not well-controlled asthma (ACT < 20) among the subpopulation of asthma patients treated with an ICS-LABA combination. We explored possible determinants (such as age, gender, body mass index, and education) and indicators (such as adherence, attitudes, quality of life, and work productivity) of not well-controlled asthma in this patient subpopulation. Not wellcontrolled asthma was associated with statistically significant impairments in health-related quality of life, work productivity and general activity. Notably, in patients treated with ICS-LABA, several previously reported predictors of poor asthma control such as body mass index, smoking or level of education did not differ consistently between well-controlled and not well-controlled patients. In contrast, there were different attitudes regarding the patientphysician relationship between these two groups. Asthma control has been assessed in various surveys in Europe and North America [16,17]. Over the last decade, the asthma control test (ACT) and the asthma control questionnaire (ACQ) have been
Table 2 Frequency of asthma symptoms. ICS-LABA
Frequency of asthma symptoms
Not well-controlled N ¼ 213
Daily 4-6/week 2-3/week 1/week 2-3/month 1-/month
Well-controlled N ¼ 169
p-value
N
%
N
(%)
81 36 58 17 13 8
38.0 16.9 27.2 8.0 6.1 3.8
30 15 30 17 23 54
17.8 8.9 17.8 10.1 13.6 32.0
<0.001*
The table shows the frequency of asthma symptoms in well-controlled and not well-controlled patients. There was a statistically significant difference between patients with well-controlled and not well-controlled asthma (chi-square test with p < 0.001 for the group comparison, marked with an asterisk).
62
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
Fig. 2. Quality of Life, healthcare resource use and work productivity loss. Health-related Quality of Life components scores from SF-12/SF-36 (A), health care resource use (B) and work productivity and activity impairment components from the WPAI-9 (C) in respondents with not well-controlled (black bars) and well-controlled asthma (grey bars). Statistically significant differences (p < 0.05) between the groups are marked with the specific p-values. Abbreviations denote: Short Form (SF), Mental component summary (MCS), Physical component summary (PCS).
validated as simple tools to measure asthma control [18e20]. Several large surveys have adopted these tests to measure asthma control in Europe. The INSPIRE and LIAISON studies recruited patients with physician-diagnosed asthma in outpatient hospitals or general practices and measured asthma control using the ACQ. The INSPIRE study, which analysed 3415 patients in 11 countries between 2004 and 2005, showed a prevalence of 72% of not wellcontrolled asthma (ACQ > 0.75) [8], while the more recent and larger LIAISON study (8119 patients in 12 countries) revealed a prevalence of 56.6% of not well-controlled asthma [21,22]. The population-based European National Health and Wellness Survey (EHWS) that examined over 37,000 adults in five European countries between 2006 and 2010 at three time points identified 2337 respondents with self-reported asthma (prevalence of asthma: 5.8%). In the total group of patients with self-reported asthma, the prevalence of not well-controlled asthma (ACT < 20) ranged between 53.5% and 56.6% [3e5]. The prevalence of 55.8% of not well-controlled asthma in our NHWS subgroup of German patients treated with an ICS-LABA combination was similar to the prevalence reported in the total populations of asthma patients of
the LIAISON and EHWS studies, suggesting that patients receiving regular maintenance treatment with ICS and LABA are still characterised by a high prevalence of not well-controlled asthma. There are many possible reasons for poor asthma control despite effective therapies [10]. Several surveys analysed predictors of poor asthma control in general populations of patients with asthma. These surveys have consistently shown that patient-related determinants such as obesity, older age, depression, smoking, level of education and poor adherence are predictors of poor asthma control in general asthma populations of European countries [3,22,23]. However, there is limited information on predictors of poor asthma control among patients treated with ICS-LABA combinations. In this sample of German patients treated with ICS-LABA combinations, factors such as age, body mass index, smoking, level of education, or self-reported adherence did not differ between well-controlled and not well-controlled patients. By contrast, we found significant differences in attitudes regarding the patient-physician relationship between well-controlled and not well-controlled ICS-LABA treated patients. These differences between well-controlled and not well-controlled patients were also found in the total European
Table 3 Attitudes of the patients towards medications and doctors.
Unless there is a good reason to change my medication, I think it is best to continue taking my medication as I currently do My doctor knows about all the over-the counter products that I use I like to see what my friends and family think of a medication before I try it I feel that my doctor is very attentive to my needs and concerns If I get sick, it is my own behaviour which determines how soon I get well again Having regular contact with my physician is the best way for me to avoid illness Whenever I don't feel well, I should consult a medically trained professional
Not well-controlled (Mean)
Well-controlled (Mean)
pvalue
3.77
4.01
0.009*
3.17 2.34 3.70 3.49 3.12 2.60
3.46 2.10 3.91 3.67 3.04 2.67
0.030* 0.032* 0.044* 0.058 0.531 0.543
Each statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree), shown are mean values for each statement. Significant differences between the groups (p < 0.05) are marked with an asterisk.
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
63
Table 4 Attitudes of the patients towards the patient-physician relationship.
Alternative healthcare providers (e.g., acupuncturist, herbalist) are as important to good health as traditional medical providers (e.g., doctor) I feel that my doctor is very attentive to my needs and concerns I prefer brand name medications to generic ones Whenever I don't feel well, I should consult a medically trained professional Having regular contact with my physician is the best way for me to avoid illness Unless there is a good reason to change my medication, I think it is best to continue taking my medication as I currently do My doctor knows about all the over-the counter products that I use I would prefer if my medications were combined into fewer Pills It is much more difficult to take medication on schedule if it has to be taken with food I like to see what my friends and family think of a medication before I try it I am willing to make any lifestyle changes necessary to avoid having to take a prescription medication I am more likely to remember to take my medications in the morning than at night I am afraid of needles I prefer to treat myself with an over-the-counter medication, than to depend on a doctor to give me a prescription medication Sometimes I take other people's medication even though it is not prescribed for me I would rather bear moderate to severe pain than treat my condition with a prescription medication The healthcare industry's profits are high yet they keep on raising their prices All prescription medications have side effects If I get sick, it is my own behaviour which determines how soon I get well again I stop taking medication when I feel better I try to take my medication at the same time every day I am not willing to tolerate side effects from my prescription medication(s) I try to take a multivitamin each day to improve or maintain my health As I age I am willing to consider medical intervention to improve my appearance As I age I am more concerned about my appearance I would take a prescription medication every day for the rest of my life to prevent a disease I may be at risk of having in the future
Not wellcontrolled
Wellcontrolled
pvalue
3.38
3.53
0.202
3.70 2.54 2.60 3.12 3.77 3.17 3.24 2.69 2.34 3.54 2.71 2.19 2.56 1.59 2.08 4.08 3.59 3.49 2.64 3.65 3.04 2.08 1.72 2.77 2.97
3.91 2.54 2.67 3.04 4.01 3.46 3.09 2.56 2.10 3.69 2.71 2.07 2.50 1.44 1.98 4.20 3.47 3.67 2.58 3.83 3.09 2.18 1.65 2.55 3.13
0.044* 0.998 0.543 0.531 0.009* 0.030* 0.179 0.262 0.032* 0.134 0.993 0.375 0.612 0.110 0.381 0.202 0.281 0.058 0.607 0.084 0.634 0.414 0.523 0.087 0.280
Each statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree), shown are mean values for each statement. Significant differences between the groups (p < 0.05) are marked with an asterisk.
EHWS asthma population [3e5], suggesting that these differences were not specific for patients treated with ICS-LABA combinations. However, it is noteworthy that patient attitudes toward their patient-physician relationship were the main differentiator between well-controlled and not well-controlled ICS-LABA treated patients (in contrast to the plethora of different patient characteristics between well-controlled and not well-controlled patients in the whole EHWS asthma population [3e5]). This finding suggests that the patient-physician relationship and the effectiveness of patient-physician communication might be of particular importance for achieving asthma control in patients with more severe disease. This analysis has several limitations. As with all survey-based studies, only patients interested in participating were included. Thus, the characteristics of the sample may differ from those who chose not to participate and may not be truly representative. Furthermore, the presence of asthma and the variables used for the classification in well-controlled and not well-controlled asthma according to the ACT were self-reported and might not reflect the real asthma prevalence and asthma control in the participants. Neither the level of asthma control nor adherence to medication were validated independently or objectively [24e26]. In addition, recall and selection biases cannot be excluded. Furthermore, the performance of multiple tests may increase the risk of false positive results. Finally, the cross-sectional nature of the study can guide hypotheses regarding the causal relationship of study variables; however, causal inferences cannot be drawn. Despite these limitations, the current analysis provides new insights into patientprovider relationships in patients with more severe asthma. However, further studies, including prospective clinical trials with wellcharacterised patients, are needed to better understand the specific determinants of asthma control in patients treated with ICS-LABA combinations. This information could guide the development of potential interventions to enhance the patient-provider
relationship and thus improve asthma control. In conclusion, this analysis demonstrates that a substantial proportion of ICS-LABA treated patients with asthma in Germany had not well-controlled disease. Not well-controlled asthma was associated with significant impairment across multiple health outcomes and different attitudes regarding the patient-physicianrelationship, emphasising the need for a better understanding of this patient subgroup and further research in this field. Contributors statement AK, TG and ML designed and supervised the analysis and wrote the manuscript draft; RP performed the statistical analysis of the data. Competing interests ML acted as a consultant and/or speaker for ALK Abello, Allergopharma, Astra Zeneca, Bencard, Berlin-Chemie, BoehringerIngelheim, Boston Scientific, Chiesi, GSK, Janssen-Cilag, MSD, Mundipharma, Novartis, Nycomed/Takeda, TEVA, UCB. TG and AK are currently employed by Boehringer Ingelheim (Ingelheim, Germany). RP was an employee of Kantar Health (Epsom, UK). Funding The data analysis was funded by Boehringer Ingelheim (Ingelheim, Germany). The National Health and Wellness Survey (NHWS) and its data are property of Kantar Health. Acknowledgement We thank Errol J. Philip, Gina Isherwood and Jeffrey Vietri (all employees of Kantar Health) for organizational support, comments
64
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
on the manuscript and writing assistance (methods section). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.rmed.2016.07.012. References [1] M.R. Sears, Trends in the prevalence of asthma, Chest 145 (2014) 219e225. [2] C.E. Jia, H.P. Zhang, Y. Lv, et al., The asthma control test and asthma control questionnaire for assessing asthma control: systematic review and metaanalysis, J. Allergy Clin. Immunol. 131 (2013) 695e703. [3] P. Demoly, P. Paggiaro, V. Plaza, et al., Prevalence of asthma control among adults in France, Germany, Italy, Spain and the UK, Eur. Respir. Rev. 18 (2009) 105e112. [4] P. Demoly, B. Gueron, K. Annunziata, L. Adamek, R.D. Walters, Update on asthma control in five European countries: results of a 2008 survey, Eur. Respir. Rev. 19 (2010) 150e157. [5] P. Demoly, K. Annunziata, E. Gubba, L. Adamek, Repeated cross-sectional survey of patient-reported asthma control in Europe in the past 5 years, Eur. Respir. Rev. 21 (2012) 66e74. [6] D. Price, M. Fletcher, T. van der Molen, Asthma control and management in 8,000 European patients: the REcognise asthma and LInk to symptoms and experience (REALISE) survey, Prim. Care Respir. Med. 24 (2014) 14009. [7] K.R. Chapman, L.P. Boulet, R.M. Rea, E. Franssen, Suboptimal asthma control: prevalence, detection and consequences in general practice, Eur. Respir. J. 31 (2008) 320e325. [8] M.R. Partridge, T. van der Molen, S.E. Myrseth, W.W. Busse, Attitudes and actions of asthma patients on regular maintenance therapy: the INSPIRE study, BMC Pulm. Med. 6 (2006) 13. [9] M.E. Wechsler, Getting control of uncontrolled asthma, Am. J. Med. 127 (2014) 1049e1059. [10] M. Lommatzsch, C.J. Virchow, Severe asthma: definition, diagnosis and treatment, Dtsch. Arztebl Int. 111 (2014) 847e855. [11] R.A. Nathan, C.A. Sorkness, M. Kosinski, et al., Development of the asthma control test: a survey for assessing asthma control, J. Allergy Clin. Immunol. 113 (2004) 59e65. [12] D.E. Morisky, A. Ang, M. Krousel-Wood, H.J. Ward, Predictive validity of a medication adherence measure in an outpatient setting, J. Clin. Hypertens. (Greenwich, Conn) 10 (2008) 348e354.
[13] D.E. Morisky, L.W. Green, D.M. Levine, Concurrent and predictive validity of a self-reported measure of medication adherence, Med. Care 24 (1986) 67e74. [14] R. Pedersini, G. Isherwood, J. Vietri, Harmonizing measurement of adherence across the 4-Item and 8-Item Morisky medication adherence scale using cross-sectional data from patients treated for irritable bowel syndrome, Value Health 16 (2013). A604. [15] M.C. Reilly, A.S. Zbrozek, E.M. Dukes, The validity and reproducibility of a work productivity and activity impairment instrument, PharmacoEconomics 4 (1993) 353e365. [16] S.T. Holgate, R. Polosa, The mechanisms, diagnosis, and management of severe asthma in adults, Lancet 368 (2006) 780e793. [17] K.F. Rabe, M. Adachi, C.K. Lai, et al., Worldwide severity and control of asthma in children and adults: the global asthma insights and reality surveys, J. Allergy Clin. Immunol. 114 (2004) 40e47. [18] S. Korn, J. Both, M. Jung, M. Hubner, C. Taube, R. Buhl, Prospective evaluation of current asthma control using ACQ and ACT compared with GINA criteria, Ann. Allergy Asthma Immunol. 107 (2011) 474e479. [19] H.K. Reddel, Characterizing asthma phenotypes: predictors and outcomes at the extremes of asthma severity, Respirology 14 (2009) 778e780. [20] M. Thomas, S. Kay, J. Pike, et al., The Asthma Control Test (ACT) as a predictor of GINA guideline-defined asthma control: analysis of a multinational crosssectional survey, Prim. Care Respir. J. 18 (2009) 41e49. [21] F. Braido, G. Brusselle, E. Ingrassia, et al., InternationaL cross-sectIonAl and longItudinal assessment on aSthma cONtrol in European adult patientsethe LIAISON study protocol, BMC Pulm. Med. 13 (2013) 18. [22] F. Braido, G. Brusselle, D. Guastalla, et al., Asthma control index in European countries: the LIAISON (InternationaL cross-sectIonAl and longitudinal assessment on aSthma cONtrol) study cross-sectional results, Eur. Respir. J. (2014) 44e50. [23] L. Cazzoletti, A. Marcon, C. Janson, et al., Asthma control in Europe: a realworld evaluation based on an international population-based study, J. Allergy Clin. Immunol. 120 (2007) 1360e1367. [24] D. Bellamy, T. Harris, Poor perceptions and expectations of asthma control: results of the International Control of Asthma Symptoms (ICAS) survey of patients and general practitioners, Prim. Care Respir. J. 14 (2005) 252e258. [25] J. Hasford, J. Uricher, M. Tauscher, P. Bramlage, J.C. Virchow, Persistence with asthma treatment is low in Germany especially for controller medication - a population based study of 483,051 patients, Allergy 65 (2010) 347e354. [26] H. Friedel, J.C. Virchow, [Health care of asthma patients in combination therapy with ICS and LABA - an analysis of statutory health insurance data], Pneumologie 68 (2014) 336e343.
Contents lists available at ScienceDirect
Respiratory Medicine journal homepage: www.elsevier.com/locate/rmed
Asthma control in patients treated with inhaled corticosteroids and long-acting beta agonists: A population-based analysis in Germany Anke Kondla b, Thomas Glaab b, Riccardo Pedersini c, d, Marek Lommatzsch a, * a
University of Rostock, Department of Respiratory Medicine, Rostock, Germany Boehringer Ingelheim Pharma GmbH & Co. KG, Medical Affairs Respiratory Medicine, Ingelheim, Germany c Kantar Health, Epsom, Surrey, UK d RTI Health Solutions, Barcelona, Spain b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 1 May 2016 Received in revised form 12 July 2016 Accepted 14 July 2016 Available online 16 July 2016
Background: The prevalence and the characteristics of poor asthma control among adults treated with combinations of inhaled corticosteroids (ICS) and long-acting beta-agonists (LABA) are not completely understood. Methods: Data from adult patients in Germany with self-reported asthma treated with an ICS-LABA combination in the National Health and Wellness Survey (NHWS) were analysed. Patients with wellcontrolled and not well-controlled asthma according to the Asthma Control Test (ACT) score were compared, with respect to socio-demographic characteristics, attitudes, adherence and outcomes. Results: Among the German patients with self-reported asthma (5.2% of the respondents), 16.2% (382 patients) were treated with an ICS-LABA combination and did not report concomitant chronic obstructive pulmonary disease, chronic bronchitis or emphysema. In this subgroup, 55.8% had not wellcontrolled asthma (ACT < 20). ICS-LABA treated patients with not well-controlled asthma were more likely to report emergency visits (16.4% vs. 8.9%), missed more time from work (absenteeism: 12.9% vs. 4.3%), were more impaired while at work (presenteeism: 29.0% vs. 14.9%) and were more likely to be women (69.0% vs. 57.4%), compared with well-controlled patients. There were no significant differences in age, body mass index, smoking, income, education or self-reported adherence between the two groups, but different attitudes regarding the patient-physician relationship. Conclusions: A substantial proportion of patients treated with ICS and LABA had not well-controlled asthma. These patients did not differ from well-controlled patients in terms of education or selfreported adherence, but in terms of their attitudes regarding the patient-physician relationship. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Asthma control Quality of life Inhaled corticosteroids (ICS) Long-acting beta-agonitsts (LABA) Germany
1. Introduction Asthma ranks among the most common chronic diseases globally and affects an estimated 300 million people worldwide [1]. Previous classifications of asthma severity based on lung function impairment have been replaced by classifications based on asthma control. This concept has been adopted in the recommendations of the global initiative for asthma (GINA) (www.ginasthma.org) and in several national asthma guidelines. Accordingly, controlled asthma is defined by minimal or no daytime and night-time symptoms,
* Corresponding author. Abteilung für Pneumologie und Internistische Intensiv€t Rostock, Ernst-Heydemann-Str. 6, medizin, Zentrum für Innere Medizin, Universita 18057 Rostock, Germany. E-mail address: [email protected] (M. Lommatzsch). http://dx.doi.org/10.1016/j.rmed.2016.07.012 0954-6111/© 2016 Elsevier Ltd. All rights reserved.
maintenance of normal activity and little or no need for reliever medication. Asthma control can be measured using questionnaires such as the Asthma Control Questionnaire (ACQ) or the Asthma Control Test (ACT) [2]. Large surveys showed that approximately 50% of all patients with asthma are not well controlled [3e7]. Poor asthma control is associated with increased risk of exacerbations, impaired quality of life, absences from school or work and increased health-care resource utilisation [8,9]. A minority of patients with asthma remain symptomatic despite treatment with an inhaled corticosteroid (ICS) and a long-acting beta-agonist (LABA). Although this subpopulation accounts for the majority of medical resource use [10], there is little information on the prevalence and the characteristics of poor asthma control in patients treated with ICS-LABA combinations. It was the aim of this analysis, therefore, to investigate this issue using a population-based
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
Abbreviations ICS LABA ACQ ACT NHWS BMI MMAS-4 MMAS-8 GINA SF-12 SF-36 MCS PCS WPAI COPD EHWS INSPIRE
Inhaled corticosteroid Long-acting beta-agonist Asthma Control Questionnaire Asthma Control Test National Health and Wellness Survey Body mass index 4-item Morisky Medication Adherence Scale 8-item Morisky Medication Adherence Scale Global Initiative for Asthma 12-item Short Form Survey 36-item Short Form Survey Mental Component Summary Physical Component Summary Work Productivity and Activity Impairment Chronic Obstructive Pulmonary Disease European National Health and Wellness Survey Investigating New Standards for Prophylaxis in Reducing Exacerbations LIAISON InternationaL cross-sectIonAl and longitudinal assessment on aSthma cONtrol
approach in Germany. In contrast to previous analyses describing the whole population of patients with asthma [3e5], we focused our analysis on the sub-population of patients treated with ICS-LABA combinations to explore patient characteristics associated with poor asthma control in subjects treated with ICS-LABA combinations.
59
composition of the NHWS sample is representative of each country's adult population. The NHWS was reviewed and approved by Essex Institutional Review Board (Lebanon, NJ, USA). 2.2. Patient characteristics The following socio-demographic and health characteristic variables were included for description of the study population: self-reported physician-diagnosed asthma, age, gender, marital status, education, income, employment type, health insurance, body mass index (BMI), smoking behaviour and self-reported asthma characteristics such as duration of the disease, frequency of asthma symptoms and severity with and without prescribed asthma medication. According to the guidelines of the World Health Organization (WHO), the following BMI categories were used: underweight to normal weight (<18.5 to <25 kg/m2), overweight (25 to <30 kg/m2) and obesity (30 kg/m2). 2.3. Asthma control The Asthma Control Test (ACT) [11] is a patient self-administered tool for the assessment of asthma control, composed of 5 items, with a 4-week recall on symptoms and daily functioning. The scores range from 5 (poor control of asthma) to 25 (complete control of asthma). An ACT score 20 indicates well-controlled asthma, a score of less than 20 indicates not well-controlled asthma. 2.4. Healthcare resource use Healthcare resource use was assessed by the self-reported number of physician visits, emergency visits and hospitalisations during the past six months. Total numbers were considered (not limited to asthma symptoms).
2. Methods 2.5. Health-related quality of life 2.1. Data source Data were extracted from the 2010, 2011 and 2013 German National Health and Wellness Survey (NHWS; not conducted in Europe in 2012) and deduplicated, i.e., only answers from the respective first year were included for any respondent participating several times in the survey. The NHWS was a self-administered, Internet-based questionnaire from a nationwide sample of adults that was stratified by gender and age to represent the demographic composition of the German adult population. Potential respondents to the NHWS were recruited primarily through an existing web-based consumer panel (Lightspeed LLC®, Washington, USA). Online recruitment was supplemented by telephone recruitment to allow better representation of older adults (65 years of age or older), who were less likely to be a member of the consumer panel. Since the survey did not allow skipping of questions reported here, there were no missing data. In the NHWS, each respondent had to complete all relevant items in the questionnaire to be included in the final data, with some items such as height, weight, and income providing an option to decline to answer. Potential respondents to the NHWS were recruited through an existing, general-purpose (i.e. not health carespecific) web-based consumer panel. The consumer panel recruited its panel members through opt-in e-mails, co-registration with panel partners, e-newsletter campaigns, banner placements and affiliate networks. All panellists explicitly agreed to be a panel member, registered with the panel through a unique e-mail address and completed an in-depth demographic registration profile. A stratified random sampling procedure, with strata by age and gender, was implemented to ensure that the demographic
Short form health surveys were used to assess generic Healthrelated Quality of Life (HR-QoL). The 2011 respondents completed the revised SF-12, and 2013 respondents completed the longer revised SF-36 scale. From these data, Physical and Mental Component Summaries (PCS and MCS) and health utility scores were derived. Physical Component Summary (PCS), an index of overall physical health status (population standard mean ¼ 50), Mental Component Summary (MCS), an index of overall mental and emotional health status, and the Short Form 6D health state (SF-6D) health utility score, a preference-based index describing overall health on a scale from 0 (equivalent to death) to 1 (equivalent to perfect health) were used. 2.6. Adherence The 4-item Morisky Medication Adherence Scale (MMAS-4) is a generic self-reported behaviour scale on medication adherence. It consists of four items with dichotomised scoring (“Yes” ¼ 1 and “No” ¼ 0) [12e14]. The items are summed to give a range of 0e4, corresponding with high (0) to medium (1e2) to low adherence (3e4). The MMAS-8 is the more recent and expanded version of MMAS-4. It consists of eight items, with yes/no-responses for the first seven and a 5-point Likert response for the last item. The items are summed and scores range from 0 to 8, indicating high (0) to medium (1e2) to low adherence (3e8). The MMAS-4 was included in the 2011 and 2012 surveys, while the MMAS-8 was included in 2013. To allow comparison, the MMAS-8 in this study was scored to proxy the 4-item measure. This scoring provides values similar to those of the MMAS-4.
60
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
2.7. Attitudes Patients' attitudes towards medications and doctors were measured with 26 statements. These statements dealt with general health issues and were not asthma-specific. Each self-reported statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree). 2.8. Work productivity and activity impairment (WPAI-9) The WPAI [15] is a 6-item validated instrument that consists of four metrics: absenteeism, presenteeism, overall work impairment and activity impairment (see: http://www.reillyassociates.net/ WPAI_General.html for further coding and scoring details). Absenteeism is the percentage of work time missed because of one's health in the past seven days. Presenteeism is the percentage of impairment experienced while being at work in the past seven days because of one's health. Overall work impairment is a combination of absenteeism and presenteeism, and activity impairment is the percentage of impairment in daily activities due to one's health in the past seven days. Only full-time, part-time, or selfemployed respondents provided data for absenteeism, presenteeism, and overall work impairment. All respondents provided data for activity impairment. 2.9. Statistical analysis Data were analysed using STATA v13.1. First, not well-controlled and well-controlled patient groups were compared using Chisquare tests (for categorical variables) or two-tailed t-tests (for cardinal variables). Probability values of p < 0.05 (two-tailed) were regarded as statistically significant. The results of the bivariate comparisons above were then used to choose the variables to be included in a series of regression models describing the association between asthma control and health outcomes, while controlling for other covariates and accounting for the hierarchical structure of the data (age and gender strata). The association with HR-QoL was modelled with linear regression, the association with WPAI and healthcare resource use with negative binomial regression and adherence with binary logistic regression.
with ICS-LABA combinations. There were more females among patients with not well-controlled asthma (69.0% vs. 57.4%, p ¼ 0.019) (Table 1). 3.3. Symptoms Frequency of asthma-related symptoms differed significantly between well-controlled and not well-controlled patients treated with ICS-LABA combinations (Table 2). Pharmacological therapy was associated with a significant reduction of symptom severity in both groups (Fig. S1, online supplement). 3.4. Health status and healthcare resource use Results of the SF-12 and SF-36 scores showed significant impairment in health-related Quality of life (HR-QoL) in patients with not well-controlled asthma as compared with well-controlled asthma (for both the physical (PCS) and the mental (MCS) component summary scores)(Fig. 2). The results were confirmed by the linear regression models (Table S1). Likewise, health utility scores were greater among those patients with well-controlled disease. However, this difference was not maintained after controlling for covariates, suggesting that socioeconomic variables (specifically gender and household income) may have a stronger association with health utility than asthma control in our sample. In this context, we found that males and patients with higher-income tend to have significantly higher health utility scores. Doctor visits (GP and specialists) during the 6 months preceding the survey did not differ significantly between the two groups (Fig. 2). Likewise, there was no significant difference in patients' perceived need for regular physician consultations. In contrast, patients with not well-controlled asthma reported more emergency visits (16.4% vs. 8.9%, p < 0.05; 0.25 vs. 0.12, p < 0.10) and hospital visits (14.6% vs. 8.3%, p < 0.10; 0.20 vs. 0.08, p < 0.05) in the previous six months compared with well-controlled patients (Fig. 2 and S2). These significant differences were not confirmed by negative binomial modelling, suggesting that household income, the presence of comorbidities and adherence to medication have a stronger association with resource use. Patients with lower income showed a
3. Results 3.1. Population sample The combined 2010, 2011 and 2013 German NHWS comprised a total of 44,895 respondents (Fig. 1). Response rates were 27.7% in 2010, 18.2% in 2011 and 14.5% in 2013. Of all respondents, 5.2% (N ¼ 2350) reported to have physician-diagnosed asthma. In this group, 31.3% (N ¼ 735) were treated with ICS plus LABA (free or fixed-dose ICS-LABA combination). Of the ICS-LABA treated patients, 52.0% (N ¼ 382) reported no concomitant COPD, chronic bronchitis or emphysema and constituted the sample of the current study. This subgroup comprising 16.2% of all participants with selfreported, physician-diagnosed asthma (N ¼ 382) was analysed in the present study (Fig. 1). According to their ACT scores, 169 (44.2%) respondents had well-controlled asthma and 213 (55.8%) not wellcontrolled asthma (Table 1). 3.2. Socio-demographic characteristics There were no differences in terms of marital status, education, income, employment, regular smoking behaviour, BMI, or age between well-controlled and not well-controlled patients treated
Fig. 1. Flow diagram of data analysis. Among all German NHWS respondents, subjects with self-reported, physician-diagnosed asthma treated with an ICS-LABA combination and not reporting COPD, chronic bronchitis or emphysema were identified (data set for analysis). This group was subdivided into respondents with wellcontrolled and not well-controlled asthma.
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
61
Table 1 Subject characteristics. ICS-LABA treated respondents
Age (years) Gender Married/in a relationship Education Yearly income
Employed full/part time BMI (kg/m2)
Smoking regularly
Well-controlled (N ¼ 169)
Not well-controlled (N ¼ 213)
Mean (SD) Female Yes Up to high school College/University Less than V20 k V20 to <40 k V40 to <75 k V75 or more No answer Yes Under-Normal Overweight Obese No answer Yes
N
%
N
%
47.78 (14.82) 147 136 176 37 61 68 38 12 34 120 74 61 73 5 59
69.01% 63.85% 82.63% 17.37% 28.64% 31.92% 17.84% 5.64% 15.96% 56.34% 34.74% 28.64% 34.27% 2.35% 27.70%
45.60 (14.19) 97 110 125 44 36 52 38 18 25 109 58 53 54 4 36
57.40% 65.09% 73.96% 26.04% 21.30% 30.77% 22.49% 10.65% 14.79% 64.50% 34.32% 31.36% 31.95% 2.37% 21.30%
p-value
0.145 0.019* 0.802 0.085 0.288
0.106 0.942
0.151
Percentages refer to the total number of patients in each group. Significant differences between the groups (p < 0.05), measured with chi-square tests (categorical variables) or t tests (continuous variables), are marked with an asterisk.
significant underuse of healthcare resources, whereas patients with more comorbidities or stronger adherence to pharmacological treatment displayed an increased use of health care resources. 3.5. Work productivity and activity impairment Significantly higher work and activity impairment scores were reported for patients with not well-controlled versus wellcontrolled asthma: time missed from work (absenteeism; 12.9% vs. 4.3%), impairments while at work (presenteeism; 29.0% vs. 14.9%), overall work impairment (36.4% vs. 17.3%), and activity impairment (42.8% vs. 22.3%) were all statistically significantly increased among patients with not well-controlled asthma (all pvalues < 0.01) (Fig. 2). These results were confirmed by a series of negative binomial regressions, which also showed that males and patients with higher household incomes had a higher work productivity. 3.6. Adherence and patient attitudes Self-reported medication adherence did not differ significantly between groups: 55.4% of well-controlled patients and 60.4% of not well-controlled patients reported high levels of adherence as measured by MMAS-4 and -8. The result was confirmed by a logistic regression analysis showing that adherence tended to increase with age. By contrast, several patient attitudes differed significantly between the groups, including the willingness to continue with a prescribed medication (Table 3). Well-controlled patients reported
a significantly higher level of attention by their physician than not well-controlled patients (Table 3). Well-controlled patients were also more likely to inform their doctor of all the over-the-counter medications they took and less likely to rely on the opinion of friends and family on which medication to use (Table 4). 4. Discussion The factors associated with not well-controlled asthma despite ICS-LABA treatment are still not completely understood. Utilising a population based survey approach across three years in Germany, we found a high prevalence (55.8%) of not well-controlled asthma (ACT < 20) among the subpopulation of asthma patients treated with an ICS-LABA combination. We explored possible determinants (such as age, gender, body mass index, and education) and indicators (such as adherence, attitudes, quality of life, and work productivity) of not well-controlled asthma in this patient subpopulation. Not wellcontrolled asthma was associated with statistically significant impairments in health-related quality of life, work productivity and general activity. Notably, in patients treated with ICS-LABA, several previously reported predictors of poor asthma control such as body mass index, smoking or level of education did not differ consistently between well-controlled and not well-controlled patients. In contrast, there were different attitudes regarding the patientphysician relationship between these two groups. Asthma control has been assessed in various surveys in Europe and North America [16,17]. Over the last decade, the asthma control test (ACT) and the asthma control questionnaire (ACQ) have been
Table 2 Frequency of asthma symptoms. ICS-LABA
Frequency of asthma symptoms
Not well-controlled N ¼ 213
Daily 4-6/week 2-3/week 1/week 2-3/month 1-/month
Well-controlled N ¼ 169
p-value
N
%
N
(%)
81 36 58 17 13 8
38.0 16.9 27.2 8.0 6.1 3.8
30 15 30 17 23 54
17.8 8.9 17.8 10.1 13.6 32.0
<0.001*
The table shows the frequency of asthma symptoms in well-controlled and not well-controlled patients. There was a statistically significant difference between patients with well-controlled and not well-controlled asthma (chi-square test with p < 0.001 for the group comparison, marked with an asterisk).
62
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
Fig. 2. Quality of Life, healthcare resource use and work productivity loss. Health-related Quality of Life components scores from SF-12/SF-36 (A), health care resource use (B) and work productivity and activity impairment components from the WPAI-9 (C) in respondents with not well-controlled (black bars) and well-controlled asthma (grey bars). Statistically significant differences (p < 0.05) between the groups are marked with the specific p-values. Abbreviations denote: Short Form (SF), Mental component summary (MCS), Physical component summary (PCS).
validated as simple tools to measure asthma control [18e20]. Several large surveys have adopted these tests to measure asthma control in Europe. The INSPIRE and LIAISON studies recruited patients with physician-diagnosed asthma in outpatient hospitals or general practices and measured asthma control using the ACQ. The INSPIRE study, which analysed 3415 patients in 11 countries between 2004 and 2005, showed a prevalence of 72% of not wellcontrolled asthma (ACQ > 0.75) [8], while the more recent and larger LIAISON study (8119 patients in 12 countries) revealed a prevalence of 56.6% of not well-controlled asthma [21,22]. The population-based European National Health and Wellness Survey (EHWS) that examined over 37,000 adults in five European countries between 2006 and 2010 at three time points identified 2337 respondents with self-reported asthma (prevalence of asthma: 5.8%). In the total group of patients with self-reported asthma, the prevalence of not well-controlled asthma (ACT < 20) ranged between 53.5% and 56.6% [3e5]. The prevalence of 55.8% of not well-controlled asthma in our NHWS subgroup of German patients treated with an ICS-LABA combination was similar to the prevalence reported in the total populations of asthma patients of
the LIAISON and EHWS studies, suggesting that patients receiving regular maintenance treatment with ICS and LABA are still characterised by a high prevalence of not well-controlled asthma. There are many possible reasons for poor asthma control despite effective therapies [10]. Several surveys analysed predictors of poor asthma control in general populations of patients with asthma. These surveys have consistently shown that patient-related determinants such as obesity, older age, depression, smoking, level of education and poor adherence are predictors of poor asthma control in general asthma populations of European countries [3,22,23]. However, there is limited information on predictors of poor asthma control among patients treated with ICS-LABA combinations. In this sample of German patients treated with ICS-LABA combinations, factors such as age, body mass index, smoking, level of education, or self-reported adherence did not differ between well-controlled and not well-controlled patients. By contrast, we found significant differences in attitudes regarding the patient-physician relationship between well-controlled and not well-controlled ICS-LABA treated patients. These differences between well-controlled and not well-controlled patients were also found in the total European
Table 3 Attitudes of the patients towards medications and doctors.
Unless there is a good reason to change my medication, I think it is best to continue taking my medication as I currently do My doctor knows about all the over-the counter products that I use I like to see what my friends and family think of a medication before I try it I feel that my doctor is very attentive to my needs and concerns If I get sick, it is my own behaviour which determines how soon I get well again Having regular contact with my physician is the best way for me to avoid illness Whenever I don't feel well, I should consult a medically trained professional
Not well-controlled (Mean)
Well-controlled (Mean)
pvalue
3.77
4.01
0.009*
3.17 2.34 3.70 3.49 3.12 2.60
3.46 2.10 3.91 3.67 3.04 2.67
0.030* 0.032* 0.044* 0.058 0.531 0.543
Each statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree), shown are mean values for each statement. Significant differences between the groups (p < 0.05) are marked with an asterisk.
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
63
Table 4 Attitudes of the patients towards the patient-physician relationship.
Alternative healthcare providers (e.g., acupuncturist, herbalist) are as important to good health as traditional medical providers (e.g., doctor) I feel that my doctor is very attentive to my needs and concerns I prefer brand name medications to generic ones Whenever I don't feel well, I should consult a medically trained professional Having regular contact with my physician is the best way for me to avoid illness Unless there is a good reason to change my medication, I think it is best to continue taking my medication as I currently do My doctor knows about all the over-the counter products that I use I would prefer if my medications were combined into fewer Pills It is much more difficult to take medication on schedule if it has to be taken with food I like to see what my friends and family think of a medication before I try it I am willing to make any lifestyle changes necessary to avoid having to take a prescription medication I am more likely to remember to take my medications in the morning than at night I am afraid of needles I prefer to treat myself with an over-the-counter medication, than to depend on a doctor to give me a prescription medication Sometimes I take other people's medication even though it is not prescribed for me I would rather bear moderate to severe pain than treat my condition with a prescription medication The healthcare industry's profits are high yet they keep on raising their prices All prescription medications have side effects If I get sick, it is my own behaviour which determines how soon I get well again I stop taking medication when I feel better I try to take my medication at the same time every day I am not willing to tolerate side effects from my prescription medication(s) I try to take a multivitamin each day to improve or maintain my health As I age I am willing to consider medical intervention to improve my appearance As I age I am more concerned about my appearance I would take a prescription medication every day for the rest of my life to prevent a disease I may be at risk of having in the future
Not wellcontrolled
Wellcontrolled
pvalue
3.38
3.53
0.202
3.70 2.54 2.60 3.12 3.77 3.17 3.24 2.69 2.34 3.54 2.71 2.19 2.56 1.59 2.08 4.08 3.59 3.49 2.64 3.65 3.04 2.08 1.72 2.77 2.97
3.91 2.54 2.67 3.04 4.01 3.46 3.09 2.56 2.10 3.69 2.71 2.07 2.50 1.44 1.98 4.20 3.47 3.67 2.58 3.83 3.09 2.18 1.65 2.55 3.13
0.044* 0.998 0.543 0.531 0.009* 0.030* 0.179 0.262 0.032* 0.134 0.993 0.375 0.612 0.110 0.381 0.202 0.281 0.058 0.607 0.084 0.634 0.414 0.523 0.087 0.280
Each statement was rated on a 5-point Likert scale (1 ¼ strongly disagree; 2 ¼ disagree; 3 ¼ neither disagree nor agree; 4 ¼ agree; 5 ¼ strongly agree), shown are mean values for each statement. Significant differences between the groups (p < 0.05) are marked with an asterisk.
EHWS asthma population [3e5], suggesting that these differences were not specific for patients treated with ICS-LABA combinations. However, it is noteworthy that patient attitudes toward their patient-physician relationship were the main differentiator between well-controlled and not well-controlled ICS-LABA treated patients (in contrast to the plethora of different patient characteristics between well-controlled and not well-controlled patients in the whole EHWS asthma population [3e5]). This finding suggests that the patient-physician relationship and the effectiveness of patient-physician communication might be of particular importance for achieving asthma control in patients with more severe disease. This analysis has several limitations. As with all survey-based studies, only patients interested in participating were included. Thus, the characteristics of the sample may differ from those who chose not to participate and may not be truly representative. Furthermore, the presence of asthma and the variables used for the classification in well-controlled and not well-controlled asthma according to the ACT were self-reported and might not reflect the real asthma prevalence and asthma control in the participants. Neither the level of asthma control nor adherence to medication were validated independently or objectively [24e26]. In addition, recall and selection biases cannot be excluded. Furthermore, the performance of multiple tests may increase the risk of false positive results. Finally, the cross-sectional nature of the study can guide hypotheses regarding the causal relationship of study variables; however, causal inferences cannot be drawn. Despite these limitations, the current analysis provides new insights into patientprovider relationships in patients with more severe asthma. However, further studies, including prospective clinical trials with wellcharacterised patients, are needed to better understand the specific determinants of asthma control in patients treated with ICS-LABA combinations. This information could guide the development of potential interventions to enhance the patient-provider
relationship and thus improve asthma control. In conclusion, this analysis demonstrates that a substantial proportion of ICS-LABA treated patients with asthma in Germany had not well-controlled disease. Not well-controlled asthma was associated with significant impairment across multiple health outcomes and different attitudes regarding the patient-physicianrelationship, emphasising the need for a better understanding of this patient subgroup and further research in this field. Contributors statement AK, TG and ML designed and supervised the analysis and wrote the manuscript draft; RP performed the statistical analysis of the data. Competing interests ML acted as a consultant and/or speaker for ALK Abello, Allergopharma, Astra Zeneca, Bencard, Berlin-Chemie, BoehringerIngelheim, Boston Scientific, Chiesi, GSK, Janssen-Cilag, MSD, Mundipharma, Novartis, Nycomed/Takeda, TEVA, UCB. TG and AK are currently employed by Boehringer Ingelheim (Ingelheim, Germany). RP was an employee of Kantar Health (Epsom, UK). Funding The data analysis was funded by Boehringer Ingelheim (Ingelheim, Germany). The National Health and Wellness Survey (NHWS) and its data are property of Kantar Health. Acknowledgement We thank Errol J. Philip, Gina Isherwood and Jeffrey Vietri (all employees of Kantar Health) for organizational support, comments
64
A. Kondla et al. / Respiratory Medicine 118 (2016) 58e64
on the manuscript and writing assistance (methods section). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.rmed.2016.07.012. References [1] M.R. Sears, Trends in the prevalence of asthma, Chest 145 (2014) 219e225. [2] C.E. Jia, H.P. Zhang, Y. Lv, et al., The asthma control test and asthma control questionnaire for assessing asthma control: systematic review and metaanalysis, J. Allergy Clin. Immunol. 131 (2013) 695e703. [3] P. Demoly, P. Paggiaro, V. Plaza, et al., Prevalence of asthma control among adults in France, Germany, Italy, Spain and the UK, Eur. Respir. Rev. 18 (2009) 105e112. [4] P. Demoly, B. Gueron, K. Annunziata, L. Adamek, R.D. Walters, Update on asthma control in five European countries: results of a 2008 survey, Eur. Respir. Rev. 19 (2010) 150e157. [5] P. Demoly, K. Annunziata, E. Gubba, L. Adamek, Repeated cross-sectional survey of patient-reported asthma control in Europe in the past 5 years, Eur. Respir. Rev. 21 (2012) 66e74. [6] D. Price, M. Fletcher, T. van der Molen, Asthma control and management in 8,000 European patients: the REcognise asthma and LInk to symptoms and experience (REALISE) survey, Prim. Care Respir. Med. 24 (2014) 14009. [7] K.R. Chapman, L.P. Boulet, R.M. Rea, E. Franssen, Suboptimal asthma control: prevalence, detection and consequences in general practice, Eur. Respir. J. 31 (2008) 320e325. [8] M.R. Partridge, T. van der Molen, S.E. Myrseth, W.W. Busse, Attitudes and actions of asthma patients on regular maintenance therapy: the INSPIRE study, BMC Pulm. Med. 6 (2006) 13. [9] M.E. Wechsler, Getting control of uncontrolled asthma, Am. J. Med. 127 (2014) 1049e1059. [10] M. Lommatzsch, C.J. Virchow, Severe asthma: definition, diagnosis and treatment, Dtsch. Arztebl Int. 111 (2014) 847e855. [11] R.A. Nathan, C.A. Sorkness, M. Kosinski, et al., Development of the asthma control test: a survey for assessing asthma control, J. Allergy Clin. Immunol. 113 (2004) 59e65. [12] D.E. Morisky, A. Ang, M. Krousel-Wood, H.J. Ward, Predictive validity of a medication adherence measure in an outpatient setting, J. Clin. Hypertens. (Greenwich, Conn) 10 (2008) 348e354.
[13] D.E. Morisky, L.W. Green, D.M. Levine, Concurrent and predictive validity of a self-reported measure of medication adherence, Med. Care 24 (1986) 67e74. [14] R. Pedersini, G. Isherwood, J. Vietri, Harmonizing measurement of adherence across the 4-Item and 8-Item Morisky medication adherence scale using cross-sectional data from patients treated for irritable bowel syndrome, Value Health 16 (2013). A604. [15] M.C. Reilly, A.S. Zbrozek, E.M. Dukes, The validity and reproducibility of a work productivity and activity impairment instrument, PharmacoEconomics 4 (1993) 353e365. [16] S.T. Holgate, R. Polosa, The mechanisms, diagnosis, and management of severe asthma in adults, Lancet 368 (2006) 780e793. [17] K.F. Rabe, M. Adachi, C.K. Lai, et al., Worldwide severity and control of asthma in children and adults: the global asthma insights and reality surveys, J. Allergy Clin. Immunol. 114 (2004) 40e47. [18] S. Korn, J. Both, M. Jung, M. Hubner, C. Taube, R. Buhl, Prospective evaluation of current asthma control using ACQ and ACT compared with GINA criteria, Ann. Allergy Asthma Immunol. 107 (2011) 474e479. [19] H.K. Reddel, Characterizing asthma phenotypes: predictors and outcomes at the extremes of asthma severity, Respirology 14 (2009) 778e780. [20] M. Thomas, S. Kay, J. Pike, et al., The Asthma Control Test (ACT) as a predictor of GINA guideline-defined asthma control: analysis of a multinational crosssectional survey, Prim. Care Respir. J. 18 (2009) 41e49. [21] F. Braido, G. Brusselle, E. Ingrassia, et al., InternationaL cross-sectIonAl and longItudinal assessment on aSthma cONtrol in European adult patientsethe LIAISON study protocol, BMC Pulm. Med. 13 (2013) 18. [22] F. Braido, G. Brusselle, D. Guastalla, et al., Asthma control index in European countries: the LIAISON (InternationaL cross-sectIonAl and longitudinal assessment on aSthma cONtrol) study cross-sectional results, Eur. Respir. J. (2014) 44e50. [23] L. Cazzoletti, A. Marcon, C. Janson, et al., Asthma control in Europe: a realworld evaluation based on an international population-based study, J. Allergy Clin. Immunol. 120 (2007) 1360e1367. [24] D. Bellamy, T. Harris, Poor perceptions and expectations of asthma control: results of the International Control of Asthma Symptoms (ICAS) survey of patients and general practitioners, Prim. Care Respir. J. 14 (2005) 252e258. [25] J. Hasford, J. Uricher, M. Tauscher, P. Bramlage, J.C. Virchow, Persistence with asthma treatment is low in Germany especially for controller medication - a population based study of 483,051 patients, Allergy 65 (2010) 347e354. [26] H. Friedel, J.C. Virchow, [Health care of asthma patients in combination therapy with ICS and LABA - an analysis of statutory health insurance data], Pneumologie 68 (2014) 336e343.