- Email: [email protected]
Fluticasone Propionate Combination Therapy
Original Article
Comparative Analysis of Persistence to Treatment among Patients with Asthma or COPD Receiving AirFluSal Forspiro or Seretide Diskus Salmeterol/ Fluticasone Propionate Combination Therapy Bruce G. Bender, PhDa, Ramon A. Hernandez Vecino, PhDb, Kevin McGrath, MScc, and Spencer Jones, PhDb Colo; Holzkirchen, Germany; and Staffordshire, United Kingdom
Denver,
What is already known about this topic? Although effective treatments for both asthma and chronic obstructive pulmonary disease are available, low adherence and persistence result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality. What does this article add to our knowledge? Persistence to treatment for first-time users of salmeterol/fluticasone propionate combination therapy is suboptimal. Inhaler devices, which differ in the technique required for correct drug inhalation, may influence persistence behavior. How does this study impact current management guidelines? These new data provide a basis for further research to better understand persistence behavior and to develop strategies to address poor persistence. BACKGROUND: Low adherence and persistence to inhaled therapy result in poor outcomes in patients with asthma and chronic obstructive pulmonary disease (COPD). Although adherence has been widely studied, growing awareness of the large number of patients who abandon their asthma treatment suggests that persistence to treatment may be more relevant for longer term outcomes. OBJECTIVE: The objective of this study was to compare persistence to salmeterol/fluticasone propionate combination treatment as AirFluSal Forspiro with persistence to Seretide
a
Department of Pediatrics, National Jewish Health, Denver, Colo Global Medical Affairs, Sandoz International GmbH, Holzkirchen, Germany c Health Informatics, Healthcare at Home Ltd., Burton upon Trent, East Staffordshire, United Kingdom The study was funded by Sandoz International GmbH. Medical writing assistance was also funded by Sandoz International GmbH. Conflicts of interest: B.G. Bender has consultant arrangements with Teva and GlaxoSmithKline and has received payment for lectures from Merck and Sandoz. R.A. Hernandez Vecino is employed by Sandoz International. S. Jones is employed by Sandoz International and has stock options in Novartis. K. McGrath declares no relevant conflicts of interest. AirFluSal and Forspiro are registered trademarks of Novartis AG. Seretide, Viani, Atmadisc and Diskus are registered trademarks of Glaxo Group Limited. Seretide Diskus is marketed in Germany under the trademark Viani Diskus. Received for publication March 17, 2016; revised June 24, 2016; accepted for publication July 7, 2016. Corresponding author: Bruce G. Bender, PhD, Department of Pediatrics, National Jewish Health, 1400 Jackson Street, Denver, CO 80206. E-mail: BenderB@ NJHealth.org. 2213-2198 Ó 2016 The Authors. Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.jaip.2016.07.006 b
884
Diskus in patients with asthma or COPD aged 12 years and above. METHODS: This study analyzed dispensing data from a large German pharmacy database. Male and female patients who were prescribed AirFluSal Forspiro were randomly paired with those who were prescribed Seretide Diskus controlling for month of treatment initiation (to limit potential seasonality effects), age groups, and gender. Matched patient pair analysis was conducted on a total of 11,774 patients (45.1% male) to compare persistence between the 2 products. RESULTS: The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus. The Renyi family of tests demonstrated a statistically significant difference (P [ .01) in persistence to AirFluSal Forspiro compared with Seretide Diskus in the overall survival experience of the 2 populations. CONCLUSIONS: In this large retrospective pharmacy database analysis, patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). These new data provide a basis for further research to better understand persistence behavior and to develop strategies to address poor persistence. Ó 2016 The Authors. Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/). (J Allergy Clin Immunol Pract 2016;4:884-9) Key words: Asthma; Chronic obstructive pulmonary disease; Combination therapy; Fluticasone propionate; Persistence; Retrospective analysis; Salmeterol
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
Abbreviations used COPD- Chronic obstructive pulmonary disease ISPOR- The International Society for Pharmacoeconomics and Outcomes Research LABA- Long-acting b-adrenoceptor agonists MPP- Matched patient pair
Asthma and chronic obstructive pulmonary disease (COPD) are chronic respiratory disorders that have a high global prevalence and are associated with significant morbidity and mortality worldwide. Although effective treatments for both asthma and COPD are available, low adherence and persistence result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality, as well as a substantial health care burden and high economic costs.1-4 Adherence and persistence to treatment are interrelated but distinct concepts. The International Society for Pharmacoeconomics and Outcomes Research (ISPOR) defines adherence as “the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen,” and persistence as “the duration of time from initiation to discontinuation of therapy.”5 Persistence is a useful indication of patients’ behavior over the long term. Evidence shows that many patients prescribed drugs to treat chronic conditions discontinue therapy well before the potential health benefits can be realized.6 Lack of persistence to inhaled medication is a significant problem in people with asthma and COPD. One study reported controller medication discontinuation rates of 43% among patients with asthma.7 Strategies to address poor adherence may differ from those that address lack of persistence, so it is crucial to understand persistence behavior further. In one study of 5504 patients who filled an initial prescription for inhaled salmeterol/ fluticasone propionate therapy, 59% never filled the prescription again over a 12-month period; the number of patients continuing to refill their prescription decreased with time, with only 9% still persisting at 12 months.8 The reasons behind poor persistence in patients with asthma and COPD are complex, and may include lack of symptoms, confusion around medications, and difficulty in handling the inhaler device. A variety of inhaler devices are used to deliver treatment for asthma and COPD, and these differ in the technique required for correct drug inhalation. The design of inhaler devices, in terms of both initial ease-of-use and intuitive features that help to maintain a good inhaler technique over time, may potentially influence persistence. To date, there have been relatively few studies relating to persistence of widely prescribed treatments for asthma and COPD,7,8 so this is an area warranting further investigation. Information from drug prescribing or dispensing databases has been used to study adherence and persistence to treatment in large numbers of patients across a wide range of therapy areas, including respiratory disease. Analysis of such data can provide valuable insights into persistence to treatment over the long term. The objective of the study was to compare persistence of treatment to AirFluSal Forspiro with persistence to Seretide Diskus in patients aged 12 years and above in a real-world setting, using a matched patient pair (MPP) analysis. Anonymized data from a German pharmacy database were analyzed to compare persistence to treatment with AirFluSal Forspiro
885
salmeterol/fluticasone propionate combination dry powder inhaler (50 mg/500 mg) with persistence to treatment with Seretide Diskus salmeterol/fluticasone propionate combination dry powder inhaler (50 mg/500 mg). Seretide Diskus is also known as Viani and Atmadisc. The Diskus device is designed to be “easy to use.” The AirFluSal Forspiro device has been specifically designed to be intuitive so that users can operate the device correctly in the long term.9 Differences in device design could potentially impact patient behavior.
METHODS The research protocol was designed using the ISPOR checklist for medication adherence and persistence studies using retrospective databases.10
Data source The data were obtained from the German supplier “INSIGHT Health.” Established in 1999, INSIGHT Health is one of the leading data providers in the German health care market and provides data from 4000 pharmacies, representing approximately 20% of the total public pharmacies in Germany. Anonymized dispensing data were obtained for patients treated with AirFluSal Forspiro or Seretide Diskus. Both AirFluSal Forspiro and Seretide Diskus are licensed for treatment of asthma and COPD; however, the patients’ specific diagnosis was not recorded in the pharmacy database.
Inclusion criteria Male and female patients aged 12 years and above who were prescribed AirFluSal Forspiro or Seretide Diskus were included in the analysis. Analysis was conducted only for patients who were naïve to fluticasone propionate/salmeterol combination therapy (ie, those who had not received a prescription for a study drug within the 6 months before the study period).
Exclusion criteria Patients less than 12 years of age or those with missing demographic data, such as for age and gender, were excluded from the analysis.
Definition of persistence Persistence was defined as “the duration of time from initiation to discontinuation of therapy” as per the ISPOR guidelines.10 Treatment was classified as “continuing” if prescriptions were refilled within 30 days after the completion date of the previous prescription. Treatment was recorded as “discontinued” if the prescription was not filled within this time. If the additional 30 days exceeded the end of the study timeframe (February 28, 2015) or if the patient still had a medication supply, then the patients were censored and were assumed to be still persisting to treatment. Patients who switched products were defined as “discontinuing” their therapy.
Statistical methods Patients prescribed AirFluSal Forspiro were randomly paired with those prescribed Seretide Diskus. Anonymized dispensing data were obtained for patients treated with AirFluSal Forspiro between February 2014 and February 2015, which represented the first 13 months of product availability in Germany. For Seretide Diskus, data collection was between February 2013 and February 2014. The follow-up periods overlapped and the data sets did not differ by more than 1 year. A total of 7178 patients received AirFluSal Forspiro during the study period. Of these, 5887 patients were included in the MPP analysis; 1291 did not meet the eligibility criteria because
886
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT SEPTEMBER/OCTOBER 2016
TABLE I. Numbers of matched patients by age and gender
Age group 12-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91þ Gender Female Male
AirFluSal Forspiro (N [ 5887)
Seretide Diskus/Viani/ Atmadisc (N [ 5887)
Total (N [ 11,774)
155 478 574 911 1213 1072 1029 419 36
155 478 574 911 1213 1072 1029 419 36
310 956 1148 1822 2426 2144 2058 838 72
3230 2657
3230 2657
6460 (54.9%) 5314 (45.1%)
of missing gender and/or age data, and were excluded. An MPP analysis was conducted on a total of 11,774 patients to compare persistence between the 2 products, controlling for month of initiation (to limit the effect of seasonality on the findings), age groups, and gender. To test for differences in persistence of treatment, product and/or device, age group, and gender were compared and the following measures were reported: sample size, including percentage of censored subjects; mean persistence, including 95% confidence intervals; Kaplan-Meier survival curves (representing persistence over time); and P values derived from a nonparametric log-rank test (Mantel-Cox test). The log-rank tests the equality of the survival distributions for AirFluSal Forspiro and Seretide Diskus. All time points are treated equally. The null hypothesis was that there is no difference between survival curves (ie, persistence to treatment with both products is equal). The log-rank test has optimum power under the assumption of proportional hazard rates; however, this assumption is often violated when 2 survival curves cross each other.11 As the Kaplan-Meier survival curves crossed, the supremum (Renyi) family of tests, designed to detect differences in survival curves that cross, were subsequently performed to provide a more comprehensive analysis.
PATIENT DEMOGRAPHICS In all, 5887 patients using AirFluSal Forspiro were paired with 5887 using Seretide Diskus. Table I shows the patient numbers by age group and gender and Table II shows the patient numbers by month. RESULTS At 6 months, the survival curves for persistence of treatment for the 2 product groups started to separate (Figure 1). The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus (Figure 1). The Renyi family of tests demonstrated a statistically significant difference (P ¼ .01) in persistence to AirFluSal Forspiro compared with Seretide Diskus in the overall survival experience of the 2 populations (Table III). The mean persistence in days was longer for
TABLE II. Numbers of matched patients by month of treatment initiation
February March April May June July August September October November December January February
AirFluSal Forspiro (N [ 5887)
Seretide Diskus/Viani/ Atmadisc (N [ 5887)
Total (N [ 11,774)
135 188 200 170 112 411 341 461 673 755 707 852 882
135 188 200 170 112 411 341 461 673 755 707 852 882
270 376 400 340 224 822 682 922 1346 1510 1414 1704 1764
AirFluSal Forspiro, 149.926 versus 134.228 for AirFluSal Forspiro and Seretide Diskus, respectively (Table IV). A higher proportion of patients using Seretide Diskus were prescribed 60 days’ treatment than for AirFluSal Forspiro; the proportion of 30-day prescriptions was 62.12% for AirFluSal Forspiro compared with 50.23% for Seretide Diskus. This difference in dispensing patterns may explain the apparent difference in persistence seen between Seretide Diskus and AirFluSal Forspiro in the first 90 days. The aim of this study was to explore persistence to treatment, to provide an indicator of overall persistence of treatment over the long term.
DISCUSSION A variety of inhaler devices are used to deliver treatment for asthma and COPD. These differ in the technique required for correct drug inhalation, which may influence adherence and persistence to treatment.12,13 Although adherence relates to conformity to recommendations about day-to-day treatment, persistence refers to the act of continuing treatment for the recommended duration.5 Several studies have reported high levels of nonpersistence with inhaled medication for asthma and COPD.7,8 The AirFluSal Forspiro device has been designed to be easy to teach and to use and achieves preference ratings of more than 90% in patients with asthma and COPD.9 In addition, intuitive design cues have been incorporated to encourage correct use of the device over the long term.9 This retrospective analysis of patients (N ¼ 11,774) included in a large German dispensing database found that overall persistence to treatment for first-time users of fluticasone propionate/salmeterol combination therapy was suboptimal. Patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). As both asthma and COPD are chronic conditions, it was important for this analysis to investigate long-term persistence behavior in patients; therefore, a 12-month time point was chosen to be appropriate from a clinical perspective. The large sample size of 11,774 patients means that this is a substantial real-world data study that provides relevant information to prescribers.
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
887
FIGURE 1. Kaplan-Meier survival (persistence) curve. The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus.
TABLE III. Overall comparison of persistence: supremum (Renyi) family of tests Weight
Log-Rank (Mantel-Cox) Breslow (Generalized Wilcoxon) Tarone-Ware Fleming Harrington (p¼1, q¼1)
X2
Classical Log-Rank P value
Q - Renyi statistic
Renyi family: P value
11.36 62.4 42.025
0.001 <0.001 <0.001
10.2 10.38 10.67 6.37
0.01 0.01 0.01 0.01
NA, not applicable.
TABLE IV. Mean persistence in days Mean* (d) 95% confidence interval
Product
Estimate
Standard error
Lower bound
Upper bound
AirFluSal Forspiro Seretide Diskus Overall
149.926 134.228 142.111
3.024 2.867 2.137
143.999 128.608 137.923
155.853 139.848 146.300
*Estimation is limited to the largest survival time if it is censored.
Retrospective database analyses Information from drug prescribing or dispensing databases has been shown to provide valuable insights into persistence of a prescribed treatment.14 For example, in North America, claims data have been used since the early 1980s for research purposes.15 In Europe, medical records databases, such as the General Practice Research Database,16 and administrative data, such as the German Statutory Health Insurance claims data, are
useful sources for pharmacoepidemiology and health services research.15 Use of such databases can enable measurement of both the total duration of therapy and the intensity of medication taking, by measuring the interval between the date of the first prescription and the point at which the patient would have had an insufficient supply of available drug to cover the gap between refills.6 Previous retrospective database studies have also shown persistence rates among people with asthma and COPD to be suboptimal.14,17,18 In a population-based cohort study of patients with COPD starting a long-acting muscarinic antagonist, long-acting b-adrenoceptor agonists (LABA), or LABAinhaled corticosteroids fixed-dose combination, only 21%-27% of patients were persistent after 1 year.17 Similarly, low persistence rates (5%-20%) were seen among patients with COPD after 1 year in a Canadian prescription claims database analysis.19 The results reported here are consistent with these earlier findings, although direct comparison between studies with differing methodologies is problematic. Strengths of this study included the large patient numbers included in the analysis and also the MPP design, which
888
BENDER ET AL
minimized differences between the 2 groups, reducing variability in treatment comparison. Furthermore, patients were matched for month of treatment initiation, limiting the effect of seasonality on the findings. Inclusion of patients who were naïve to treatment for 6 months before the start of the study allowed the impact of specific treatments on persistence behavior to be assessed. Although database analyses are a well-accepted methodology, there are limitations. For example, prescription refill data do not provide information about actual daily medication use; they simply provide evidence of the extent to which patients obtain their medicines.8 In addition, the data source was such that the specific indication for which AirFluSal Forspiro or Seretide Diskus had been prescribed was not known; however, both products are licensed for asthma and COPD. It was also not possible to identify specific health centers at which patients received their care or to determine the social background of patients, and the database covered only approximately 20% of the total public pharmacies in Germany. As the anonymized pharmacy data could not be linked to patient records, it was not possible to examine outcomes in relation to treatment persistence. Several studies have previously shown the link between suboptimal adherence to poor health outcomes among people with asthma and COPD.20,21 At this stage, the reasons for longer persistence to AirFluSal Forspiro compared with Seretide Diskus are unclear and further research is required to establish the underlying reasons for this finding. The AirFluSal Forspiro (fluticasone propionate/salmeterol; Sandoz, International GmbH, Germany) device has been designed to be easy to teach and to use9 and achieves preference ratings of more than 90% in patients with asthma and COPD.9 In addition, intuitive design cues have been incorporated to encourage the correct use of the device over the long term,9 which may encourage persistence to therapy. Furthermore, although several studies have investigated strategies to improve adherence to inhaled medication,4 long-term persistence to treatment is a different concept and may require different interventions.
CONCLUSIONS Although effective treatments for asthma and COPD are available, lack of adherence and low persistence may result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality, as well as a burden on health care systems and high economic costs. An apparently easy-to-use device still requires the clinician to provide a clear, initial demonstration and the inhaler technique is known to deteriorate over time.22 In addition, intuitive device design may aid consistently a good inhaler technique over the long term, potentially promoting persistence. Analyses of large prescribing and dispensing databases can provide valuable insights into patient behavior. In this initial, retrospective pharmacy database analysis, patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). Accepting the limitations of this type of analysis, these new data provide a basis for further investigation and research, which could involve using other national databases
J ALLERGY CLIN IMMUNOL PRACT SEPTEMBER/OCTOBER 2016
with the capability to link drug utilization to respiratory disease outcomes. These data are supported by previous retrospective database analyses that have concluded that persistence to inhaled treatment among patients with asthma and COPD is poor, and further strategies to address poor persistence are required.
Acknowledgments Medical writing assistance was provided by Synergy. All authors contributed to preparation of the manuscript. Sciensus at Healthcare at Home Ltd designed the research and performed the data analysis. All authors contributed to the interpretation of results. REFERENCES 1. Bryant J, McDonald VM, Boyes A, Sanson-Fisher R, Paul C, Melville J. Improving medication adherence in chronic obstructive pulmonary disease: a systematic review. Respir Res 2013;14:109. 2. Toy EL, Beaulieu NU, McHale JM, Welland TR, Plauschinat CA, Swensen A, et al. Treatment of COPD: relationships between daily dosing frequency, adherence, resource use, and costs. Respir Med 2011;105:435-41. 3. Thomas M. Why aren’t we doing better in asthma: time for personalised medicine? NPJ Prim Care Respir Med 2015;25:15004. 4. Mäkelä MJ, Backer V, Hedegaard M, Larsson K. Adherence to inhaled therapies, health outcomes and costs in patients with asthma and COPD. Respir Med 2013;107:1481-90. 5. Cramer JA, Roy A, Burrell A, Fairchild CJ, Fuldeore MJ, Ollendorf DA, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44-7. 6. Caetano PA, Lam JM, Morgan SG. Toward a standard definition and measurement of persistence with drug therapy: examples from research on statin and antihypertensive utilization. Clin Ther 2006;28:1411-24. 7. Vanelli M, Pedan A, Liu N, Hoar J, Messier D, Kiarsis K. The role of patient inexperience in medication discontinuation: a retrospective analysis of medication nonpersistence in seven chronic illnesses. Clin Ther 2009;31:2628-52. 8. Bender BG, Pedan A, Varasteh LT. Adherence and persistence with fluticasone propionate/salmeterol combination therapy. J Allergy Clin Immunol 2006;118: 899-904. 9. Virchow JC, Weuthen T, Harmer QJ, Jones S. Identifying the features of an easy-to-use and intuitive dry powder inhaler for asthma and chronic obstructive pulmonary disease therapy: results from a 28-day device handling study, and an airflow resistance study. Expert Opin Drug Deliv 2014;11:1849-57. 10. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A checklist for medication compliance and persistence studies using retrospective databases. Value Health 2007;10:3-12. 11. Li H, Han D, Hou Y, Chen H, Chen Z. Statistical inference methods for two crossing survival curves: a comparison of methods. PLoS One 2016;10: e0116774. 12. Lavorini F, Braido F, Baiardini I, Blasi F, Canonica GW, SIAAC-SIMER. Asthma and COPD: interchangeable use of inhalers. A document of Italian Society of Allergy, Asthma and Clinical Immunology (SIAAIC) & Italian Society of Respiratory Medicine (SIMeR). Pulm Pharmacol Ther 2015;34: 25-30. 13. Darbà J, Ramírez G, Sicras A, Francoli P, Torvinen S, Sánchez-de la Rosa R. The importance of inhaler devices: the choice of inhaler device may lead to suboptimal adherence in COPD patients. Int J Chorn Obstruct Pulmon Dis 2015; 10:2335-45. 14. Hasford J, Uricher J, Tauscher M, Bramlage P, Virchow JC. Persistence with asthma treatment is low in Germany especially for controller medication—a population based study of 483 051 patients. Allergy 2010;65:347-54. 15. Kreis K, Neubauer S, Klora M, Lange A, Zeidler J. Status and perspectives of claims data analyses in Germany—a systematic review. Health Policy 2016; 120:213-26. 16. García Rodríguez LA, Pérez Gutthann S. Use of the UK General Practice Research Database for pharmacoepidemiology. Br J Clin Pharmacol 1998;45: 419-25. 17. Penning-van Beest F, Van Herk-Sukel M, Gale R, Lammers JW, Herings R. Three-year dispensing patterns with long-acting inhaled drugs in COPD: a database analysis. Respir Med 2011;105:259-65. 18. Barnes CB, Ulrik CS. Asthma and adherence to inhaled corticosteroids: current status and future perspectives. Respir Care 2015;60:455-68.
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
19. Cramer JA, Bradley-Kennedy C, Scalera A. Treatment persistence and compliance with medications for chronic obstructive pulmonary disease. Can Respir J 2007;14:25-9. 20. Bender BG. Advancing the science of adherence measurement: implications for the clinician. J Allergy Clin Immunol Pract 2013;1:92-3.
BENDER ET AL
889
21. Antoniu SA. Adherence to inhaled therapy in COPD: effects on survival and exacerbations. Expert Rev Pharmacoecon Outcomes Res 2010;10: 115-7. 22. Lavorini F. Inhaled drug delivery in the hands of the patient. J Aerosol Med Pulm Drug Deliv 2014;27:414-8.
Comparative Analysis of Persistence to Treatment among Patients with Asthma or COPD Receiving AirFluSal Forspiro or Seretide Diskus Salmeterol/ Fluticasone Propionate Combination Therapy Bruce G. Bender, PhDa, Ramon A. Hernandez Vecino, PhDb, Kevin McGrath, MScc, and Spencer Jones, PhDb Colo; Holzkirchen, Germany; and Staffordshire, United Kingdom
Denver,
What is already known about this topic? Although effective treatments for both asthma and chronic obstructive pulmonary disease are available, low adherence and persistence result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality. What does this article add to our knowledge? Persistence to treatment for first-time users of salmeterol/fluticasone propionate combination therapy is suboptimal. Inhaler devices, which differ in the technique required for correct drug inhalation, may influence persistence behavior. How does this study impact current management guidelines? These new data provide a basis for further research to better understand persistence behavior and to develop strategies to address poor persistence. BACKGROUND: Low adherence and persistence to inhaled therapy result in poor outcomes in patients with asthma and chronic obstructive pulmonary disease (COPD). Although adherence has been widely studied, growing awareness of the large number of patients who abandon their asthma treatment suggests that persistence to treatment may be more relevant for longer term outcomes. OBJECTIVE: The objective of this study was to compare persistence to salmeterol/fluticasone propionate combination treatment as AirFluSal Forspiro with persistence to Seretide
a
Department of Pediatrics, National Jewish Health, Denver, Colo Global Medical Affairs, Sandoz International GmbH, Holzkirchen, Germany c Health Informatics, Healthcare at Home Ltd., Burton upon Trent, East Staffordshire, United Kingdom The study was funded by Sandoz International GmbH. Medical writing assistance was also funded by Sandoz International GmbH. Conflicts of interest: B.G. Bender has consultant arrangements with Teva and GlaxoSmithKline and has received payment for lectures from Merck and Sandoz. R.A. Hernandez Vecino is employed by Sandoz International. S. Jones is employed by Sandoz International and has stock options in Novartis. K. McGrath declares no relevant conflicts of interest. AirFluSal and Forspiro are registered trademarks of Novartis AG. Seretide, Viani, Atmadisc and Diskus are registered trademarks of Glaxo Group Limited. Seretide Diskus is marketed in Germany under the trademark Viani Diskus. Received for publication March 17, 2016; revised June 24, 2016; accepted for publication July 7, 2016. Corresponding author: Bruce G. Bender, PhD, Department of Pediatrics, National Jewish Health, 1400 Jackson Street, Denver, CO 80206. E-mail: BenderB@ NJHealth.org. 2213-2198 Ó 2016 The Authors. Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). http://dx.doi.org/10.1016/j.jaip.2016.07.006 b
884
Diskus in patients with asthma or COPD aged 12 years and above. METHODS: This study analyzed dispensing data from a large German pharmacy database. Male and female patients who were prescribed AirFluSal Forspiro were randomly paired with those who were prescribed Seretide Diskus controlling for month of treatment initiation (to limit potential seasonality effects), age groups, and gender. Matched patient pair analysis was conducted on a total of 11,774 patients (45.1% male) to compare persistence between the 2 products. RESULTS: The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus. The Renyi family of tests demonstrated a statistically significant difference (P [ .01) in persistence to AirFluSal Forspiro compared with Seretide Diskus in the overall survival experience of the 2 populations. CONCLUSIONS: In this large retrospective pharmacy database analysis, patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). These new data provide a basis for further research to better understand persistence behavior and to develop strategies to address poor persistence. Ó 2016 The Authors. Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/ 4.0/). (J Allergy Clin Immunol Pract 2016;4:884-9) Key words: Asthma; Chronic obstructive pulmonary disease; Combination therapy; Fluticasone propionate; Persistence; Retrospective analysis; Salmeterol
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
Abbreviations used COPD- Chronic obstructive pulmonary disease ISPOR- The International Society for Pharmacoeconomics and Outcomes Research LABA- Long-acting b-adrenoceptor agonists MPP- Matched patient pair
Asthma and chronic obstructive pulmonary disease (COPD) are chronic respiratory disorders that have a high global prevalence and are associated with significant morbidity and mortality worldwide. Although effective treatments for both asthma and COPD are available, low adherence and persistence result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality, as well as a substantial health care burden and high economic costs.1-4 Adherence and persistence to treatment are interrelated but distinct concepts. The International Society for Pharmacoeconomics and Outcomes Research (ISPOR) defines adherence as “the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen,” and persistence as “the duration of time from initiation to discontinuation of therapy.”5 Persistence is a useful indication of patients’ behavior over the long term. Evidence shows that many patients prescribed drugs to treat chronic conditions discontinue therapy well before the potential health benefits can be realized.6 Lack of persistence to inhaled medication is a significant problem in people with asthma and COPD. One study reported controller medication discontinuation rates of 43% among patients with asthma.7 Strategies to address poor adherence may differ from those that address lack of persistence, so it is crucial to understand persistence behavior further. In one study of 5504 patients who filled an initial prescription for inhaled salmeterol/ fluticasone propionate therapy, 59% never filled the prescription again over a 12-month period; the number of patients continuing to refill their prescription decreased with time, with only 9% still persisting at 12 months.8 The reasons behind poor persistence in patients with asthma and COPD are complex, and may include lack of symptoms, confusion around medications, and difficulty in handling the inhaler device. A variety of inhaler devices are used to deliver treatment for asthma and COPD, and these differ in the technique required for correct drug inhalation. The design of inhaler devices, in terms of both initial ease-of-use and intuitive features that help to maintain a good inhaler technique over time, may potentially influence persistence. To date, there have been relatively few studies relating to persistence of widely prescribed treatments for asthma and COPD,7,8 so this is an area warranting further investigation. Information from drug prescribing or dispensing databases has been used to study adherence and persistence to treatment in large numbers of patients across a wide range of therapy areas, including respiratory disease. Analysis of such data can provide valuable insights into persistence to treatment over the long term. The objective of the study was to compare persistence of treatment to AirFluSal Forspiro with persistence to Seretide Diskus in patients aged 12 years and above in a real-world setting, using a matched patient pair (MPP) analysis. Anonymized data from a German pharmacy database were analyzed to compare persistence to treatment with AirFluSal Forspiro
885
salmeterol/fluticasone propionate combination dry powder inhaler (50 mg/500 mg) with persistence to treatment with Seretide Diskus salmeterol/fluticasone propionate combination dry powder inhaler (50 mg/500 mg). Seretide Diskus is also known as Viani and Atmadisc. The Diskus device is designed to be “easy to use.” The AirFluSal Forspiro device has been specifically designed to be intuitive so that users can operate the device correctly in the long term.9 Differences in device design could potentially impact patient behavior.
METHODS The research protocol was designed using the ISPOR checklist for medication adherence and persistence studies using retrospective databases.10
Data source The data were obtained from the German supplier “INSIGHT Health.” Established in 1999, INSIGHT Health is one of the leading data providers in the German health care market and provides data from 4000 pharmacies, representing approximately 20% of the total public pharmacies in Germany. Anonymized dispensing data were obtained for patients treated with AirFluSal Forspiro or Seretide Diskus. Both AirFluSal Forspiro and Seretide Diskus are licensed for treatment of asthma and COPD; however, the patients’ specific diagnosis was not recorded in the pharmacy database.
Inclusion criteria Male and female patients aged 12 years and above who were prescribed AirFluSal Forspiro or Seretide Diskus were included in the analysis. Analysis was conducted only for patients who were naïve to fluticasone propionate/salmeterol combination therapy (ie, those who had not received a prescription for a study drug within the 6 months before the study period).
Exclusion criteria Patients less than 12 years of age or those with missing demographic data, such as for age and gender, were excluded from the analysis.
Definition of persistence Persistence was defined as “the duration of time from initiation to discontinuation of therapy” as per the ISPOR guidelines.10 Treatment was classified as “continuing” if prescriptions were refilled within 30 days after the completion date of the previous prescription. Treatment was recorded as “discontinued” if the prescription was not filled within this time. If the additional 30 days exceeded the end of the study timeframe (February 28, 2015) or if the patient still had a medication supply, then the patients were censored and were assumed to be still persisting to treatment. Patients who switched products were defined as “discontinuing” their therapy.
Statistical methods Patients prescribed AirFluSal Forspiro were randomly paired with those prescribed Seretide Diskus. Anonymized dispensing data were obtained for patients treated with AirFluSal Forspiro between February 2014 and February 2015, which represented the first 13 months of product availability in Germany. For Seretide Diskus, data collection was between February 2013 and February 2014. The follow-up periods overlapped and the data sets did not differ by more than 1 year. A total of 7178 patients received AirFluSal Forspiro during the study period. Of these, 5887 patients were included in the MPP analysis; 1291 did not meet the eligibility criteria because
886
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT SEPTEMBER/OCTOBER 2016
TABLE I. Numbers of matched patients by age and gender
Age group 12-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91þ Gender Female Male
AirFluSal Forspiro (N [ 5887)
Seretide Diskus/Viani/ Atmadisc (N [ 5887)
Total (N [ 11,774)
155 478 574 911 1213 1072 1029 419 36
155 478 574 911 1213 1072 1029 419 36
310 956 1148 1822 2426 2144 2058 838 72
3230 2657
3230 2657
6460 (54.9%) 5314 (45.1%)
of missing gender and/or age data, and were excluded. An MPP analysis was conducted on a total of 11,774 patients to compare persistence between the 2 products, controlling for month of initiation (to limit the effect of seasonality on the findings), age groups, and gender. To test for differences in persistence of treatment, product and/or device, age group, and gender were compared and the following measures were reported: sample size, including percentage of censored subjects; mean persistence, including 95% confidence intervals; Kaplan-Meier survival curves (representing persistence over time); and P values derived from a nonparametric log-rank test (Mantel-Cox test). The log-rank tests the equality of the survival distributions for AirFluSal Forspiro and Seretide Diskus. All time points are treated equally. The null hypothesis was that there is no difference between survival curves (ie, persistence to treatment with both products is equal). The log-rank test has optimum power under the assumption of proportional hazard rates; however, this assumption is often violated when 2 survival curves cross each other.11 As the Kaplan-Meier survival curves crossed, the supremum (Renyi) family of tests, designed to detect differences in survival curves that cross, were subsequently performed to provide a more comprehensive analysis.
PATIENT DEMOGRAPHICS In all, 5887 patients using AirFluSal Forspiro were paired with 5887 using Seretide Diskus. Table I shows the patient numbers by age group and gender and Table II shows the patient numbers by month. RESULTS At 6 months, the survival curves for persistence of treatment for the 2 product groups started to separate (Figure 1). The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus (Figure 1). The Renyi family of tests demonstrated a statistically significant difference (P ¼ .01) in persistence to AirFluSal Forspiro compared with Seretide Diskus in the overall survival experience of the 2 populations (Table III). The mean persistence in days was longer for
TABLE II. Numbers of matched patients by month of treatment initiation
February March April May June July August September October November December January February
AirFluSal Forspiro (N [ 5887)
Seretide Diskus/Viani/ Atmadisc (N [ 5887)
Total (N [ 11,774)
135 188 200 170 112 411 341 461 673 755 707 852 882
135 188 200 170 112 411 341 461 673 755 707 852 882
270 376 400 340 224 822 682 922 1346 1510 1414 1704 1764
AirFluSal Forspiro, 149.926 versus 134.228 for AirFluSal Forspiro and Seretide Diskus, respectively (Table IV). A higher proportion of patients using Seretide Diskus were prescribed 60 days’ treatment than for AirFluSal Forspiro; the proportion of 30-day prescriptions was 62.12% for AirFluSal Forspiro compared with 50.23% for Seretide Diskus. This difference in dispensing patterns may explain the apparent difference in persistence seen between Seretide Diskus and AirFluSal Forspiro in the first 90 days. The aim of this study was to explore persistence to treatment, to provide an indicator of overall persistence of treatment over the long term.
DISCUSSION A variety of inhaler devices are used to deliver treatment for asthma and COPD. These differ in the technique required for correct drug inhalation, which may influence adherence and persistence to treatment.12,13 Although adherence relates to conformity to recommendations about day-to-day treatment, persistence refers to the act of continuing treatment for the recommended duration.5 Several studies have reported high levels of nonpersistence with inhaled medication for asthma and COPD.7,8 The AirFluSal Forspiro device has been designed to be easy to teach and to use and achieves preference ratings of more than 90% in patients with asthma and COPD.9 In addition, intuitive design cues have been incorporated to encourage correct use of the device over the long term.9 This retrospective analysis of patients (N ¼ 11,774) included in a large German dispensing database found that overall persistence to treatment for first-time users of fluticasone propionate/salmeterol combination therapy was suboptimal. Patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). As both asthma and COPD are chronic conditions, it was important for this analysis to investigate long-term persistence behavior in patients; therefore, a 12-month time point was chosen to be appropriate from a clinical perspective. The large sample size of 11,774 patients means that this is a substantial real-world data study that provides relevant information to prescribers.
BENDER ET AL
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
887
FIGURE 1. Kaplan-Meier survival (persistence) curve. The survival probability estimates at 12 months were 0.229 (0.02 standard error) for AirFluSal Forspiro versus 0.105 (0.025 standard error) for Seretide Diskus.
TABLE III. Overall comparison of persistence: supremum (Renyi) family of tests Weight
Log-Rank (Mantel-Cox) Breslow (Generalized Wilcoxon) Tarone-Ware Fleming Harrington (p¼1, q¼1)
X2
Classical Log-Rank P value
Q - Renyi statistic
Renyi family: P value
11.36 62.4 42.025
0.001 <0.001 <0.001
10.2 10.38 10.67 6.37
0.01 0.01 0.01 0.01
NA, not applicable.
TABLE IV. Mean persistence in days Mean* (d) 95% confidence interval
Product
Estimate
Standard error
Lower bound
Upper bound
AirFluSal Forspiro Seretide Diskus Overall
149.926 134.228 142.111
3.024 2.867 2.137
143.999 128.608 137.923
155.853 139.848 146.300
*Estimation is limited to the largest survival time if it is censored.
Retrospective database analyses Information from drug prescribing or dispensing databases has been shown to provide valuable insights into persistence of a prescribed treatment.14 For example, in North America, claims data have been used since the early 1980s for research purposes.15 In Europe, medical records databases, such as the General Practice Research Database,16 and administrative data, such as the German Statutory Health Insurance claims data, are
useful sources for pharmacoepidemiology and health services research.15 Use of such databases can enable measurement of both the total duration of therapy and the intensity of medication taking, by measuring the interval between the date of the first prescription and the point at which the patient would have had an insufficient supply of available drug to cover the gap between refills.6 Previous retrospective database studies have also shown persistence rates among people with asthma and COPD to be suboptimal.14,17,18 In a population-based cohort study of patients with COPD starting a long-acting muscarinic antagonist, long-acting b-adrenoceptor agonists (LABA), or LABAinhaled corticosteroids fixed-dose combination, only 21%-27% of patients were persistent after 1 year.17 Similarly, low persistence rates (5%-20%) were seen among patients with COPD after 1 year in a Canadian prescription claims database analysis.19 The results reported here are consistent with these earlier findings, although direct comparison between studies with differing methodologies is problematic. Strengths of this study included the large patient numbers included in the analysis and also the MPP design, which
888
BENDER ET AL
minimized differences between the 2 groups, reducing variability in treatment comparison. Furthermore, patients were matched for month of treatment initiation, limiting the effect of seasonality on the findings. Inclusion of patients who were naïve to treatment for 6 months before the start of the study allowed the impact of specific treatments on persistence behavior to be assessed. Although database analyses are a well-accepted methodology, there are limitations. For example, prescription refill data do not provide information about actual daily medication use; they simply provide evidence of the extent to which patients obtain their medicines.8 In addition, the data source was such that the specific indication for which AirFluSal Forspiro or Seretide Diskus had been prescribed was not known; however, both products are licensed for asthma and COPD. It was also not possible to identify specific health centers at which patients received their care or to determine the social background of patients, and the database covered only approximately 20% of the total public pharmacies in Germany. As the anonymized pharmacy data could not be linked to patient records, it was not possible to examine outcomes in relation to treatment persistence. Several studies have previously shown the link between suboptimal adherence to poor health outcomes among people with asthma and COPD.20,21 At this stage, the reasons for longer persistence to AirFluSal Forspiro compared with Seretide Diskus are unclear and further research is required to establish the underlying reasons for this finding. The AirFluSal Forspiro (fluticasone propionate/salmeterol; Sandoz, International GmbH, Germany) device has been designed to be easy to teach and to use9 and achieves preference ratings of more than 90% in patients with asthma and COPD.9 In addition, intuitive design cues have been incorporated to encourage the correct use of the device over the long term,9 which may encourage persistence to therapy. Furthermore, although several studies have investigated strategies to improve adherence to inhaled medication,4 long-term persistence to treatment is a different concept and may require different interventions.
CONCLUSIONS Although effective treatments for asthma and COPD are available, lack of adherence and low persistence may result in poor patient outcomes including symptoms, reduced quality of life, exacerbations, hospitalizations, and increased mortality, as well as a burden on health care systems and high economic costs. An apparently easy-to-use device still requires the clinician to provide a clear, initial demonstration and the inhaler technique is known to deteriorate over time.22 In addition, intuitive device design may aid consistently a good inhaler technique over the long term, potentially promoting persistence. Analyses of large prescribing and dispensing databases can provide valuable insights into patient behavior. In this initial, retrospective pharmacy database analysis, patients using AirFluSal Forspiro were more likely to persist with treatment compared with those using Seretide Diskus as demonstrated by the overall survival experience of the 2 populations (12-month study period). Accepting the limitations of this type of analysis, these new data provide a basis for further investigation and research, which could involve using other national databases
J ALLERGY CLIN IMMUNOL PRACT SEPTEMBER/OCTOBER 2016
with the capability to link drug utilization to respiratory disease outcomes. These data are supported by previous retrospective database analyses that have concluded that persistence to inhaled treatment among patients with asthma and COPD is poor, and further strategies to address poor persistence are required.
Acknowledgments Medical writing assistance was provided by Synergy. All authors contributed to preparation of the manuscript. Sciensus at Healthcare at Home Ltd designed the research and performed the data analysis. All authors contributed to the interpretation of results. REFERENCES 1. Bryant J, McDonald VM, Boyes A, Sanson-Fisher R, Paul C, Melville J. Improving medication adherence in chronic obstructive pulmonary disease: a systematic review. Respir Res 2013;14:109. 2. Toy EL, Beaulieu NU, McHale JM, Welland TR, Plauschinat CA, Swensen A, et al. Treatment of COPD: relationships between daily dosing frequency, adherence, resource use, and costs. Respir Med 2011;105:435-41. 3. Thomas M. Why aren’t we doing better in asthma: time for personalised medicine? NPJ Prim Care Respir Med 2015;25:15004. 4. Mäkelä MJ, Backer V, Hedegaard M, Larsson K. Adherence to inhaled therapies, health outcomes and costs in patients with asthma and COPD. Respir Med 2013;107:1481-90. 5. Cramer JA, Roy A, Burrell A, Fairchild CJ, Fuldeore MJ, Ollendorf DA, et al. Medication compliance and persistence: terminology and definitions. Value Health 2008;11:44-7. 6. Caetano PA, Lam JM, Morgan SG. Toward a standard definition and measurement of persistence with drug therapy: examples from research on statin and antihypertensive utilization. Clin Ther 2006;28:1411-24. 7. Vanelli M, Pedan A, Liu N, Hoar J, Messier D, Kiarsis K. The role of patient inexperience in medication discontinuation: a retrospective analysis of medication nonpersistence in seven chronic illnesses. Clin Ther 2009;31:2628-52. 8. Bender BG, Pedan A, Varasteh LT. Adherence and persistence with fluticasone propionate/salmeterol combination therapy. J Allergy Clin Immunol 2006;118: 899-904. 9. Virchow JC, Weuthen T, Harmer QJ, Jones S. Identifying the features of an easy-to-use and intuitive dry powder inhaler for asthma and chronic obstructive pulmonary disease therapy: results from a 28-day device handling study, and an airflow resistance study. Expert Opin Drug Deliv 2014;11:1849-57. 10. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A checklist for medication compliance and persistence studies using retrospective databases. Value Health 2007;10:3-12. 11. Li H, Han D, Hou Y, Chen H, Chen Z. Statistical inference methods for two crossing survival curves: a comparison of methods. PLoS One 2016;10: e0116774. 12. Lavorini F, Braido F, Baiardini I, Blasi F, Canonica GW, SIAAC-SIMER. Asthma and COPD: interchangeable use of inhalers. A document of Italian Society of Allergy, Asthma and Clinical Immunology (SIAAIC) & Italian Society of Respiratory Medicine (SIMeR). Pulm Pharmacol Ther 2015;34: 25-30. 13. Darbà J, Ramírez G, Sicras A, Francoli P, Torvinen S, Sánchez-de la Rosa R. The importance of inhaler devices: the choice of inhaler device may lead to suboptimal adherence in COPD patients. Int J Chorn Obstruct Pulmon Dis 2015; 10:2335-45. 14. Hasford J, Uricher J, Tauscher M, Bramlage P, Virchow JC. Persistence with asthma treatment is low in Germany especially for controller medication—a population based study of 483 051 patients. Allergy 2010;65:347-54. 15. Kreis K, Neubauer S, Klora M, Lange A, Zeidler J. Status and perspectives of claims data analyses in Germany—a systematic review. Health Policy 2016; 120:213-26. 16. García Rodríguez LA, Pérez Gutthann S. Use of the UK General Practice Research Database for pharmacoepidemiology. Br J Clin Pharmacol 1998;45: 419-25. 17. Penning-van Beest F, Van Herk-Sukel M, Gale R, Lammers JW, Herings R. Three-year dispensing patterns with long-acting inhaled drugs in COPD: a database analysis. Respir Med 2011;105:259-65. 18. Barnes CB, Ulrik CS. Asthma and adherence to inhaled corticosteroids: current status and future perspectives. Respir Care 2015;60:455-68.
J ALLERGY CLIN IMMUNOL PRACT VOLUME 4, NUMBER 5
19. Cramer JA, Bradley-Kennedy C, Scalera A. Treatment persistence and compliance with medications for chronic obstructive pulmonary disease. Can Respir J 2007;14:25-9. 20. Bender BG. Advancing the science of adherence measurement: implications for the clinician. J Allergy Clin Immunol Pract 2013;1:92-3.
BENDER ET AL
889
21. Antoniu SA. Adherence to inhaled therapy in COPD: effects on survival and exacerbations. Expert Rev Pharmacoecon Outcomes Res 2010;10: 115-7. 22. Lavorini F. Inhaled drug delivery in the hands of the patient. J Aerosol Med Pulm Drug Deliv 2014;27:414-8.