Abstract
Background:
Long-acting bronchodilator inhalers are widely used with or without inhaled corticosteroids (ICs) by patients with lung diseases. In Israel alone, there are 21 inhalers containing long-acting β2 agonists (LABAs) and/or long-acting muscarinic antagonists (LAMAs). Some patients are treated incorrectly with several inhalers of the same pharmacologic group.
Methods:
Electronic data of LABA and/or LAMA inhalers purchased during a period of 1 year were extracted in one district of Clalit Health Services in Israel. Patients who were treated with two or more inhalers from the same pharmacologic group were compared with patients without duplicate treatment. Inhaler purchases during the 12 months before and after the first duplicate purchase were compared with the purchases by patients without duplication of treatment. New diagnoses were compared to identify possible side effects.
Results:
Of the 13,528 patients who were treated with LABA and/or LAMA inhalers, 244 (1.8%) purchased at least two different inhalers from the same pharmacologic group. Inhaler purchases were 3.8 times higher in the duplication group during the 12 months before the first duplication. Inhaler purchase increased by 28% in the duplication group compared with a 4.5% increase in the nonduplication group (p < 0.001) during the following year. The risk for duplicated consumption was significantly higher in patients with a chronic obstructive pulmonary disease (COPD) diagnosis, males, and persons aged between 61 and 80 years.
Conclusions:
Nearly 2% of the patients treated with long-acting bronchodilators consumed different medications of the same pharmacologic group even when adherence was satisfactory. COPD patients are at higher risk for inhaler duplication. Clinical Trial Registration Number: 0151-20-COM1.
Introduction
Obstructive lung diseases, mainly asthma and chronic obstructive pulmonary disease (COPD), are common chronic diseases that are prone to exacerbation, causing school and work absence and hospitalization. They are responsible for the deaths of around 4 million people annually and affect the lives of hundreds of millions of others.(1,2) Goals of treatment are to lower symptoms, frequency, and severity of exacerbation and to improve endurance for physical activity and enhance quality of life. The main pharmaceutical treatments in these diseases are inhalers containing bronchodilators and/or inhaled corticosteroids (ICs).
There are two main families of bronchodilators: the short-acting and long-acting β2 agonists (SABAs and LABAs, respectively (and the anticholinergics: short-acting and long-acting muscarinic antagonists (SAMAs and LAMAs, respectively). The standard treatment for dyspnea in COPD patients includes long-acting bronchodilator inhalers as a single medication, or a combination of LABAs and LAMAs, while ICs are usually reserved for patients with frequent or severe exacerbation.(3)
In Israel, there are currently 30 inhalers that differ from each other mainly by the shape of the inhaler, its technical operation, and the acting substances. Twenty-one inhalers contain LABAs and/or LAMAs with or without ICs. While a suitable treatment is effective in lowering exacerbation, the benefits of inhaler usage are not fully accomplished, especially due to inappropriate adherence to the prescribed inhalers(4,5) and multiple errors in the technique of inhaler usage.(3,6–8)
Different studies have shown that adherence to inhaler treatment among COPD and asthma patients is the lowest,(9,10) compared with the compliance of patients with other oral treatments.(11,12) In addition, as adherence increases, exacerbation decreases.(12,13) Treatment with multiple inhalers and/or several doses a day are major reasons for low adherence(12) and therefore efforts should be made to lower the number of inhalers.
Nowadays, there is an increased use of computer programs to track pitfalls such as interactions between medicines, contraindications, inappropriate dosage, and also duplication of prescriptions.(14,15) Electronic pharmacy records are commonly utilized for measuring adherence to treatments.(10,16) It has been found that 30% to 60% of patients do not consistently purchase their chronic medications.(17) As a consequence, a reverse correlation between subjective compliance and the number of chronic inhalers purchased has been found.(10)
It is reasonable to assume that in cases of unresolved symptoms or unstable disease, physicians are more likely to increase the dosage or number of drugs regardless of patient adherence. The existence of many inhalers of the same pharmacologic type requires substantial knowledge on the part of the primary physician to avoid drug duplication.
Using electronic pharmacy data, one study showed that 67% of the alerts were due to duplication of medication.(15) An intervention study at an outpatient practice of a tertiary care medical center demonstrated a reduction of duplicated prescriptions of albuterol (SABA) from 4% to 2.6%.(18)
In another intervention study, in one pharmacy in Germany, duplicate prescriptions of inhalers among asthma and COPD patients during 10 months were evaluated.(19) Duplicated prescriptions were 1% of all prescriptions, while a higher rate was found involving combination drug inhalers. A higher rate of prescribing duplications was found among primary care physicians compared with specialists of lung diseases (0.86% vs. 0.57% of prescriptions, respectively).
The authors of this study suggested that possible reasons for duplication, other than errors and distractions, were multiple visits to different physicians and, more frequently, nonresolving symptoms and the fact that patients can purchase medications in several pharmacies. It was noted in this study that a survey conducted in Germany in 2015 found that 69.7% of lung disease specialists reported that they have encountered prescription duplication in referral patients (unpublished data).
Duplication can lead to overdose and an excess of morbidity.(20) A high dose of bronchodilators was correlated with side effects such as a dry mouth, a metal taste, cardiovascular events, and arrhythmias, including sinus tachycardia, tremors, and minor saturation declines.(3,21) High doses of ICs were associated with multiple fractures, cataract, high consumption of antacids, muscle weakness, back pain, skin ulcers, oral candida, and dental problems.(22) Despite these side effects, there were still cases where more than one inhaler of the same pharmacologic family was prescribed.
In this study, we explored the duplication rates of LABA and/or LAMA inhaler purchases among patients of the largest public health provider in one district in the center of Israel. Quantities of inhalers purchased during periods of 1 year before and 1 year after the first duplication were compared with patients without inhaler duplication to compare adherence between the two groups. We evaluated what types of patients were prone to inhaler duplication and whether new diagnoses were related to these duplications.
Our hypothesis was that fewer purchases of inhalers during the year before the first duplication could be a risk factor for duplication of prescriptions. We believe that more patients with COPD than asthma will have inhaler duplications since symptoms in COPD patients are more continuous and progressive than in asthma patients.
Methods
The study was conducted in community clinics in the Dan-Petah-Tiqwa district of Clalit Health Services, the largest publicly funded Health Maintenance Organization (HMO) in Israel. All patients in Israel have free access to their primary care physician, and all drug purchases are computerized. The study was approved by the institutional review board (IRB) of Meir Medical Center, Kfar-Saba, Israel.
Inclusion criteria were males and females aged 18 years and above who purchased inhalers containing long-acting bronchodilators (LABAs and/or LAMAs) during 2018. Inhalers that contain different generic drugs of the same pharmacologic family were considered duplicated.
The patients were divided into two groups: a duplication group consisting of those who had duplicated purchases in the same month, for at least 2 months during the year, not necessarily consecutive (since drugs could be purchased for 3 months at a time and therefore different duplications can be created in nonsequential months). The first month of a duplicated purchase was defined as the month of duplication for each patient (Fig. 1).

An example of duplication for a patient who purchased two different long-acting beta inhalers (salmeterol and formoterol). Data show 3 months with duplicate purchases (April, June, and October). The inclusion criterion for the duplication group is duplicated purchases in the same month, for at least 2 months, of different long-acting bronchodilator inhalers of the same pharmacologic group. April is considered the first month of duplication. We examined the number of inhalers purchased between the 12 months since the first duplication (April to March next year) and the 12 months before the first duplication (April of the prior year until March this year). ICs, inhaled corticosteroids.
The control group (nonduplication group) included the rest of the patients who purchased these inhalers at least for 2 months during 2018. Since the first month of duplication in the first group is not the same for all patients, to compare the amount of purchases before and after the first duplication between the two groups, we determined July 1, 2018, to be the reference date for the control group for which purchases would be evaluated a year before and after.
The next step was to compare inhaler purchases in both groups with respect to age, gender, body–mass index (BMI), smoking status, and the diagnosis of lung diseases for which inhalers were prescribed.
Statistical analysis
Comparison between the groups was made using t-tests for continuous variables and chi-squared tests or Fisher's exact test for categorical variables. The statistical analysis was conducted using the R Project for Statistical Computing—version 3.4.1 (https://www.R-project.org/).
Results
During 2018, LABA and/or LAMA inhalers were purchased by 13,528 patients. Of these, 244 (1.8%) patients were classified as the duplication group according to the above criteria. Six patients purchased three different LABA medications in their first month of duplication.
The demographic differences between the two groups are described in Table 1. There were statistically significant differences in the age and gender between the two groups. In the duplication group, 53% were males compared with 43% in the nonduplication group (p = 0.002). Moreover, the average age in the duplication group was 66.6 years (standard deviation [SD] 13.8) compared with 57.3 years (SD 18.7) in the nonduplication group (p < 0.001).
Demographic Data of the Two Groups
BMI, body–mass index; SD, standard deviation.
The most frequent age group for inhaler therapy was 61–70 years in both groups, although it was more prominent in the duplication group (31.6% vs. 22%). Most patients (60.3%) in the duplication group were between the ages of 61 and 80 years, and only 26.6% of the patients were 60 years old or younger in this group compared with 49.7% in the nonduplication group.
There was no statistically significant difference in the BMI between the two groups. Although active smoking was more common in the nonduplication group (23.1% vs. 11.1%, p < 0.001), this analysis is not representative since in most cases (73.8%), we could not differentiate between never smokers and those with lack of information about smoking status in our data.
The diagnoses for which the long-acting bronchodilator inhalers were prescribed are detailed in Table 2. While a respiratory diagnosis that could justify the prescription of inhalers was found in 94.3% of the duplication group, only 66.6% of patients in the nonduplication group had such a diagnosis (p < 0.001). Significant differences were found, as expected, with the diagnosis of COPD—70.1% in the duplication group versus 26.4% in the control group, (p < 0.001)—while smaller differences were found for asthma diagnosis, 59% versus 52.5% in the duplication and nonduplication groups, respectively (p = 0.052).
Diagnosis According to the Long-Acting Bronchodilators Prescribed in Both Groups
COPD, chronic obstructive pulmonary disease.
COPD as a single obstructive disease was found in 35.2% of patients in the duplication group, while only 13.6% of patients in the nonduplication group were pure COPD patients. Asthma as a single obstructive diagnosis was found in 39.8% in the nonduplication group compared with 24.2% in the duplication group (p < 0.001).
A statistically significant difference was found in the diagnosis of bronchiectasis, which was more prevalent in the duplication group than in the nonduplication group (2.5% vs. 0.8%, respectively; p = 0.015); however, only a small fraction of the population (110) was diagnosed with bronchiectasis. Upper respiratory tract infection, acute bronchitis, or dyspnea was diagnosed in a minority of patients with nonsignificant statistical differences between the groups.
Average inhaler purchase with long-acting bronchodilators in the duplication group during the 12 months before the first duplication was 16.9 inhalers and only 4.5 inhalers (3.8 times less) in the nonduplication group during the same period. Furthermore, the average inhaler purchase during the 12 months after the first duplication was 21.7 inhalers, a statistically significant increase of 28% in the duplication group compared with a 4.5% increase (4.7 inhalers) in the nonduplication group (p < 0.001) in the parallel year.
To assess the unwanted effects of long-acting bronchodilator duplication or overdose, we analyzed new diagnosis records with suspected and relevant side effects for major systems (cardiovascular, hypertension, glaucoma, neurological, and psychiatry). In the 12 months from the first duplicated purchase in the duplication group, compared with the parallel period in the nonduplication group, there were no statistically significant differences in any system, excluding a few psychiatric diagnoses (depression, anxiety, post-traumatic stress disorder, and narcissistic personality disorder, which were more common in the duplication group—3.7% compared with 1.7%, respectively; p = 0.04) and glaucoma cases (1.6% compared with 0.5%, respectively; p = 0.033).
Discussion
In this study, we found that 1.8% (244) of patients who were treated with LABA and/or LAMA inhalers purchased additional inhalers of the same pharmacologic family at the same time during a period of 1 year. Most concerning is that six patients purchased three different LABA medications in their first month of duplication. Our findings are similar to a previous study, which found a 1% inhaler duplication purchase rate during a period of 10 months in asthma and COPD patients at one pharmacy in Germany.(19)
Purchasing a drug is not synonymous with using it, however, since repeated purchasing (paying and collecting the drugs) is the most likely indicator that the drugs are being used rather than stockpiled. In addition, we considered those who had duplicated purchases in the same month, for at least 2 months during the year, which raises the possibility for actual usage of the drug. Nevertheless, we believe that prescribing the duplication is the main obstacle and physicians should be aware of the duplication.
These findings raise the need for electronic programs that alert the physician and pharmacists when duplicate drugs (or overlapping medications) are being prescribed or purchased. In addition, treatment algorithms according to the diagnosis should be linked to each patient. This is even more essential in current times where there are many similar drug options, which require strong pharmacological skills from the primary care physician.
The risk of receiving duplicated prescriptions was significantly higher in males and between the ages of 61 and 80 years, which better correlates with a COPD diagnosis than an asthma diagnosis. Asthma as a single diagnosis was more common in the nonduplication group, as expected, since symptoms in this disease fluctuate more and are treated mostly by a combination of ICs and LABAs in the same inhaler.
Nevertheless, it is concerning that duplication of LABAs may indicate that one of the inhalers could have been an LABA inhaler without ICs (in our experience, mostly formoterol). This is contraindicated in asthma patients who could omit the ICs with an LABA inhaler (e.g., fear of corticosteroids) and use only the LABA inhaler, which is mainly a symptom reliever.
Our study calls for awareness of the misuse of LABA monotherapy in asthma patients. A lack of a pulmonary diagnosis that justifies any inhaler prescription was more prevalent in the nonduplication group and could be explained by acute or subacute illnesses or due to underdiagnosed mild asthma patients. In these cases, it is likely that single inhalers comprising a combination of ICs and LABAs were prescribed by the primary care physicians. Pulmonary diseases other than COPD or asthma were found in lower frequencies, which preclude the possibility of reaching solid conclusions.
Active smoking was twice as high in the nonduplication group; however, these differences represent a minority of the study population since in most cases, we could not differentiate nonsmokers from those with missing information regarding smoking status.
As opposed to our hypothesis that duplications of long-acting bronchodilator inhalers are prescribed due to low adherence to the former treatment, our findings revealed a significantly higher purchase of these inhalers during the year before the duplicate purchase (3.8 times more than in the nonduplication group).
Furthermore, a 28% increase in inhaler purchases was found in this group during the year after the first duplication, as opposed to a nonsignificant increase in the nonduplication group. This can indicate that patients with more prominent and progressive disease such as COPD are undertreated and get “more of the same” treatment from the primary care physicians due to unresolved symptoms instead of combination therapy by a pulmonary specialist and assessment of inhaler technique usage.
The combination of several medications in one inhaler, preferably taken once a day, optimizes the treatment and increases adherence.(23) Therefore, several inhalers have been developed that contain two medications (LABA and LAMA or ICs and LABA), and in recent years, inhalers that contain three pharmacological groups (LABA+LAMA+ICs) have been developed. Current recommendations for patients with obstructive lung disease who are symptomatic using monotherapy are to combine several medications such as in the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines.(3)
Even more rigorous are the updated guidelines of the American Thoracic Society (ATS) of 2020, which strongly recommend an LABA+LAMA combined inhaler as a first-line treatment for COPD patients with dyspnea and effort intolerance.(24) Treatment with different types of LABAs for the same patient is not recommended, such as in asthma. According to the new guidelines of the Global Initiative for Asthma (GINA), the fast-acting LABA, formoterol, can be used in the same inhaler with ICs as a maintenance and rescue medication, but not in patients who are treated with a maintenance inhaler, which contains another type of LABA.(25)
Multiple usage of β2 agonists can increase morbidity and tolerance and could mask life-threatening symptoms such as in asthma.(26) High doses of LABAs could result in significant side effects. For example, in the instructions provided with Relvar Ellipta (gsk), a combined IC and LABA inhaler, there is a warning not to use the medication in combination with other medications containing LABAs to avoid side effects.(27)
In our study, we did not find significant differences between the two groups in new diagnoses during the year after the first duplication, excluding the number of psychiatric diagnoses (depression, anxiety, PTDS, and narcissistic personality disorder, which were higher (3.7%) in the duplication group compared with the nonduplication group (1.7%); p = 0.04) and glaucoma cases (1.6% compared with 0.5%, respectively; p = 0.033). However, our data are retrospective and limited and represent a small population of patients with duplication of treatment.
In addition, depression and anxiety are more common among asthma and COPD patients, especially in patients with nonresolving symptoms or more severe diseases.(3,25,28–30) In a prospective study, the risk of depression and anxiety among newly diagnosed COPD patients did not increase during the first 12 weeks of treatment with inhalers containing LABAs, LAMAs, or ICs.(31) Larger and more prolonged prospective studies are needed to evaluate the risk of psychiatric side effects in this population.
There are several limitations to this retrospective study. First, this study is based on incomplete electronic records that possibly lack the full clinical picture, including symptoms and the reason for which a particular inhaler was prescribed. Therefore, there are many patients, especially in the nonduplication group, without a pulmonary diagnosis for which the inhaler was prescribed.
Second, purchase of an inhaler does not indicate its actual usage; however, it is reasonable to believe that the more inhalers purchased, the more symptomatic the patient, and hence there is a higher motivation for usage of the purchased medication. In our experience, many patients who purchased multiple inhalers used them simultaneously on the same day. Another limitation is the different ways different physicians document the actual diagnosis at the time of drug prescription and later when documenting side effects.
Furthermore, there are several options to record the diagnosis such as different built-in diagnoses for the same clinical state or free text. Finally, often the diagnosis of asthma or COPD by the primary care physician does not meet the formal criteria, especially in cases of asthma-COPD overlap. To overcome these limitations, further studies using structural computer-based systems are needed.
Conclusions
Nearly 2% of the patients treated with long-acting bronchodilators consume different medications of the same pharmacologic group even when adherence is satisfactory. COPD patients are at higher risk for inhaler duplication and should be better evaluated, preferably by a pulmonary specialist when symptomatic, using long-acting bronchodilator monotherapy.
In an era where many inhalers are available under prescription, it would be most effective if electronic alerts of drug duplication could be embedded in the software for the physician writing the prescription and at the pharmacy.
Footnotes
Authors' Contributions
D.S. was involved in conceptualization, formal analysis, investigation, methodology, project administration, supervision, validation, writing—original draft, and writing—review and editing. J.C. was involved in investigation, supervision, and validation. A.A. was involved in data curation. B.O. was involved in formal analysis and writing—review and editing. I.K. was involved in formal analysis, investigation, and writing—original draft.
Author Disclosure Statement
Dr. D.S. has received lecture fees and/or consultancy fees from TEVA, GSK, Rafa Laboratories, AstraZeneca, Boehringer Ingelheim, and Kamada. All other authors have nothing to declare.
Funding Information
This research received no specific grant from any funding agency.
Reviewed by:
Omar Usmani
Mark Utell
