Effect of Combination Fluticasone Propionate and Salmeterol or Inhaled Corticosteroids on Asthma-Related Outcomes in a Medicare-Eligible Population

R.H. Stanford et al.

The American Journal of Geriatric Pharmacotherapy

Effect of Combination Fluticasone Propionate and Salmeterol or Inhaled Corticosteroids on Asthma-Related Outcomes in a Medicare-Eligible Population Richard H. Stanford, PharmD, MS1; Christopher M. Blanchette, PhD2; Melissa H. Roberts, MS3; Hans Petersen, MS3; and Anne L. Fuhlbrigge, MD, MS4 1

GlaxoSmithKline, Research Triangle Park, North Carolina; 2Health Economics and Outcomes Research, IMS Health, Alexandria, Virginia; 3Lovelace Clinic Foundation, Albuquerque, New Mexico; and 4Brigham and Women’s Hospital, Boston, Massachusetts

ABSTRACT Background: National asthma treatment guidelines recommend either the use of inhaled corticosteroids (ICS) or ICS in combination with a long-acting bronchodilator for the treatment of moderate to severe asthma. Even though asthma is common among older adults, few studies have assessed the differences in effectiveness between these two recommended therapies in patients over 65 years of age. Objective: The aim of this study was to assess the association of the fluticasone-salmeterol combination (FSC) or ICS initiation on asthma-related events in Medicare-eligible asthma patients. Methods: This was a retrospective observational study using a large health claims database (July 1, 2001 to June 30, 2008). Subjects 65 to 79 years of age with 12-month preindex and 3- to 12-month postindex eligibility, an asthma diagnosis (ICD-493.xx), and with 1 or more FSC or ICS claims at index were included. Subjects with an FSC or ICS claim in the preindex and any claim for chronic obstructive pulmonary disease were excluded. Subjects were observed until they had an event (emergency department [ED] inpatient hospitalization [IP], combined IP/ED or oral corticosteroid [OCS] use) or were no longer eligible in the database, whichever came first. Cox proportional hazards regression was used to assess risk of an asthma-related event (IP, ED, or IP/ED). Baseline characteristics (age, sex, region, index season, comorbidities, preindex use of short-acting ␤-agonists, OCS, other asthma controllers, and asthma-related ED/IP visits) were independent covariates in the model. Results: A total of 10,837 met the criteria (4843 ICS and 5994 FSC). Age (70.4 and 70.5 years, respectively) and the percentage of female subjects (65.5% and 64.8%, respectively) were similar. Asthma-related events were also similar at baseline. Postindex unadjusted rates occurring after ⬎30 days were ED (1.8% vs 1.5%, P ⫽ 0.18), IP (2.7% vs 1.7%, P ⬍ 0.001), and ED/IP (4.1% vs 2.8%, P ⬍ 0.001) for ICS and FSC, respectively. Subjects who received FSC were associated with a 32% (adjusted HR ⫽ 0.68; 95% CI, 0.51– 0.91) lower risk of experiencing an IP visit and a 22% (HR ⫽ 0.78; 95% CI, 0.62– 0.98) lower risk of experiencing an ED/IP visit. No differences were observed for ED visits (HR ⫽ 0.94; 95% CI, 0.68 –1.29). Conclusions: In Medicare-eligible asthma patients, FSC use was associated with lower rates of asthma-related serious exacerbations compared with ICS. (Am J Geriatr Pharmacother. 2012;10:343–351) © 2012 Elsevier HS Journals, Inc. All rights reserved. Key words: asthma, emergency department, fluticasone propionate/salmeterol, hospitalizations, Medicare eligible.

INTRODUCTION Asthma is a highly prevalent chronic lung disease with ⬎28 million Americans who report having had an asthma diagnosis at some point in their lives.1 Of those with asthma, a subset of symptomatic asthmatics experience exacerbations with varying frequency and sever-

© 2012 Elsevier HS Journals, Inc. All rights reserved.

ity. Exacerbations represent one of the most important outcomes in clinical and observational asthma studies because they have the greatest impact on quality of life. Furthermore, asthma exacerbations impose severe emotional and financial stress, reduce quality of life, and impair the ability to work and lead to emergency http://dx.doi.org/10.1016/j.amjopharm.2012.09.005 1543-5946/$ - see front matter

Volume 10 ● Number 6

December 2012

343

The American Journal of Geriatric Pharmacotherapy

R.H. Stanford et al.

department (ED) visits, hospitalization, and even death.2 Current treatment regimens for asthma emphasize efforts to control exacerbations by reducing exposure to triggers and promoting adherence to pharmaceutical regimens. Because exacerbations are linked to both increased health-care utilization as well as substantial impairment in functioning, the economic burden of asthma is high. The indirect cost of asthma exacerbations leads to work and school days missed as well as life-years lost. Estimates of those affected in the United States vary between 20% and 40%.3,4 Total direct costs of asthma in the United States have been estimated at $15.6 billion in direct health-care expenditures.4 Hospitalizations, ED visits, unscheduled outpatient visits, and rescue medication needs encompass the majority of exacerbation-related treatment costs. Patients who experience frequent exacerbations (⬃approximately 20% of those with asthma) are estimated to incur 80% of the total direct costs of asthma.3 The goal of long-term management of asthma is to reduce both the impairment (assessment of lung function and of the intensity and frequency of asthma symptoms) as well as the risk (frequency of exacerbations and the potential for progressive loss of lung function) associated with asthma. National Asthma Education and Prevention Program (NAEPP) guidelines recommend a stepwise approach to pharmacologic therapy that consists of short-term rescue medications to relieve acute symptoms and long-term daily use of controller medications to control symptoms among individuals with persistent asthma. Long-term asthma controller medications include corticosteroids, immunomodulators, methylxanthines (MTYs), and long-acting ␤2-agonists (LABAs). Inhaled corticosteroids (ICSs) are considered the preferred long-term controller medication for persistent asthma control.5–9 Current guidelines suggest that patients initiate ICS as first-line maintenance therapy.10 Although LABAs are not recommended as monotherapy, combination ICS and LABA therapies have demonstrated improvement in lung function and symptoms, as well as lower rates of exacerbation and use of short-acting ␤2-agonist (SABA) rescue medications.5,11–17 Approximately 15% of adults with a history of asthma are 65 years of age or older.1 The US population is rapidly aging, and by 2020, an estimated 21% of the population will be 65 years of age or older.18 Older asthmatics (⬎65 years of age) have been shown to have a lower quality of life, more severe and persistent symptoms, higher pharmacologic use, and a re-

344

duced chance of remission.7,19 –22 Even though asthma is common among older adults, few studies focus exclusively on measuring the effectiveness of treatments for older adults with asthma. Research has demonstrated that asthma patients 60 years of age or older more frequently use corticosteroids (both oral and inhaled) than patients younger than 60 years of age.23–25 Although ICS is recommended as first-line therapy, recent research has shown a significant benefit of the ICS/LABA combination treatment, fluticasone/salmeterol (FSC), over ICS in lowering the rate of exacerbation-related health-care utilization.26 Although there are potential risks involved with combination therapy, FSC has demonstrated a reduction in exacerbation-related health-care events and increased symptom control in children and adults.27 The objective of this study was to assess the incremental value of FCS compared with ICS on asthma-related health services use (ie, hospital and ED use) among Medicare-eligible (aged 65 years or older) asthma patients.

METHODS The design of this study was a retrospective intent-totreat longitudinal study of older adults with asthma from the PharMetrics Integrated Database. The database contains deidentified, Health Insurance Portability and Accountability Act– compliant medical and pharmacy claims from ⬎55 million enrollees from ⬎90 US health plans. It also includes dates of service and International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis codes, participating plan payments, and billed charges information. Although most participating plans provide information on billed charges, all plans provide information on payments. Payments are imputed by PharMetrics for claims with missing payment information due to capitation arrangements.28 Therefore, costs are reimbursed and do not need to be transformed with cost-to-charge ratios.

Sample This study was limited to those individuals with medical claims data for services provided between July 1, 2001 and June 30, 2008. The study also included all patients with a diagnosis of asthma who filled at least 1 prescription for FSC or ICS and at least 3 months of follow-up care after the initial prescription claim (as determined through a 12-month washout period). The date of the first prescription claim for FSC or ICS was considered an individual’s index date; patients who were

R.H. Stanford et al.

65 to 79 years of age on the index date were included in the study. Patients were required to have continuous enrollment in a health plan for at least 12 months before the index date and 3 to 12 months after the index date and to have at least 1 asthma-related encounter (outpatient visit, ED visit, or hospitalization with a primary or secondary ICD-9-CM diagnosis code of 493.xx) during the 12-month preindex observation period. Individuals were also required to have at least 1 prescription claim in both the preindex period and the postindex period to ensure that all study patients had prescription drug coverage through their health plan, thereby limiting unobserved prescription drug use. Patients with an index date based on an FSC prescription claim did not have an ICS prescription claim in the 12-month preindex period, and, similarly, patients with an index date based on an ICS prescription claim did not have an FSC prescription claim. Patients were excluded if they had a medical claim indicating a diagnosis of chronic obstructive pulmonary disease (COPD) (ICD-9-CM codes 491.xx, 492.xx, 496.xx), any respiratory tract cancer (ICD-9-CM codes 161, 161.x, 162, 163, 163.x, 231, 231.x), or bronchiectasis (ICD-9-CM code 494.xx).

Measures The main outcome measure of interest was the combined event of either an asthma-related hospitalization or an ED visit that occurred ⱖ30 days post index. This 30-day time period was chosen to ensure that the index treatment was associated with the events of interest. However, asthma rates were observed from day 1 post index to assess whether a similar pattern was observed. Based on previous research and due to the low incidence of asthma-related hospitalizations a combined exacerbation end point was chosen.26 Secondary outcome and utilization measures included asthma-related inpatient hospitalizations (IPs), ED visits, prescription claims for oral corticosteroids (OCS), use of SABA, (a marker for asthma control),29 and total asthma-related costs. Baseline characteristics of the sample were assessed from the preindex period (to include index date events, with the exception of FSC or ICS prescription fills) and included demographic (age category, geographic region, and sex) and medical (comorbidities and utilization) characteristics, as well as asthma-related costs for medical services and pharmacy products. Preindex asthma-related costs were defined as the amount paid by the health plan; costs paid by other sources were not included. Medical service utilization and associated costs were similarly assessed in the period

The American Journal of Geriatric Pharmacotherapy

post index, although it should be noted that patients did not have uniform follow-up time periods in the postindex period.

Statistical Analysis Descriptive statistics were assessed for preindex demographic and medical characteristics and for costs for both treatment groups. Counts and percentages were calculated for categorical variables, and measures of central tendency were used for continuous variables. Means and frequencies for health-care utilization and costs were calculated for both treatment groups during the pre- and postindex date periods (Table I). Comorbid conditions that were reviewed were chosen based on the classifications by Elixhauser et al30 and included obstructive sleep apnea, heart disease (rheumatic heart failure, hypertensive heart disease, ischemic heart disease, and unspecified cardiovascular disease), serious lung conditions other than asthma, and COPD, as these have previously been identified as risk factors in obstructive lung disease.31 Cox proportional hazard regression models were used to estimate the time to first event for each asthma-related health-care utilization category within the 3- to 12-month postindex period. The measurement of time to first event began 30 days after the first prescription claim for FSC or ICS. Patients who had an asthmarelated event (IP, ED, or OCS use) within 30 days of the index date were excluded from each of the corresponding analyses (eg, patients with an ED visit within 30 days were excluded from the ED analyses, but not from the OCS or the IP alone analyses). Patients were observed until 1 or more of the following occurred: patients had an asthma-related event such as IP, ED, ED/IP, and OCS or was no longer eligible in the database. The Breslow method was used for evaluating ties.32 Covariates included in the Cox proportional hazard regressions were sex; age; season in which the index date occurred; geographic region; selected preperiod comorbidities; any fill in the preindex period for SABA, leukotriene receptor antagonists, theophylline, anticholinergics, or OCS; asthma-related hospitalizations/ED visits; and outpatient visits. Comorbidities were selected for inclusion in analyses if the comorbidity frequency was significantly different between the 2 treatment groups using a P value ⱕ0.20, and the comorbidity was present in ⬎2% of the treatment groups. All statistical tests were 2 sided with an ␣ level set at 0.05 for significance. Analyses were conducted with SAS version 9.2 for Windows (SAS Institute, Cary, North Carolina).

345

The American Journal of Geriatric Pharmacotherapy

R.H. Stanford et al.

Table I. Baseline demographic and comorbidity characteristics (total subjects ⫽ 10,837). Parameter*

FSC (n ⫽ 5994)

ICS (n ⫽ 4843)

P

Age, y, mean (SD) 65–69 70–74 75–79

70.5 (4.26) 2902 (48.4) 1539 (25.7) 1553 (25.9)

70.4 (3.95) 2227 (46.0) 1612 (33.3) 1004 (20.7)

0.10 0.01 ⬍0.0001 ⬍0.0001

Male Region Mid-Atlantic Mountain National New England North Central Pacific South Atlantic South Central

2108 (35.2)

1670 (34.5)

0.46

1812 (30.2) 467 (7.8) 148 (2.5) 875 (14.6) 1005 (16.8) 226 (3.8) 851 (14.2) 610 (10.2)

1542 (31.8) 369 (7.6) 240 (5.0) 1133 (23.4) 699 (14.4) 135 (2.8) 457 (9.4) 268 (5.5)

0.07 0.74 ⬍0.0001 ⬍0.0001 ⬍0.001 0.005 ⬍0.0001 ⬍0.0001

Season of asthma diagnosis Winter Spring Summer Fall Preindex comorbid conditions Arrhythmia Congestive heart failure Deficiency anemias Depression Diabetes (complicated) Diabetes (uncomplicated) Fluid disorders Heart disease Hypertension (complicated) Hypertension (uncomplicated) Hypothyroidism Other lung conditions Obesity

Parameter* †

Obstructive sleep apnea Other neurological disease Peripheral vascular Psychoses Renal failure Rheumatoid arthritis Valvular disease

FSC (n ⫽ 5994)

ICS (n ⫽ 4843)

P†

256 (4.3)

110 (2.3)

⬍0.0001

143 (2.4) 369 (6.2) 135 (2.3) 154 (2.6)

120 (2.5) 271 (5.6) 112 (2.3) 96 (2.0)

0.76 0.22 0.83 0.04

262 (4.4) 587 (9.8)

200 (4.1) 415 (8.6)

0.54 0.03

FSC ⫽ fluticasone-salmeterol combination; ICS ⫽ inhaled corticosteroid. *Values shown are number (%) unless otherwise stated. † P values were calculated using the ␹2 test for categorical data and t test for continuous data.

RESULTS 1994 (33.3) 1687 (28.1) 1229 (20.5) 184 (18.1)

1618 (33.4) 1393 (28.8) 914 (18.9) 918 (19.0)

0.88 0.48 0.03 0.25

692 (11.5)

529 (10.9)

0.31

431 (7.2) 462 (7.7) 217 (3.6)

299 (6.2) 344 (7.1) 161 (3.3)

0.04 0.23 0.40

268 (4.5)

195 (4.0)

0.26

1135 (18.9) 258 (4.3) 1143 (19.1)

724 (14.9) 198 (4.1) 885 (18.3)

⬍0.0001 0.58 0.29

286 (4.8)

195 (4.0)

0.06

3118 (52.0) 638 (10.6)

2334 (48.2) 444 (9.2)

⬍0.0001 0.01

768 (12.8) 159 (2.7)

494 (10.2) 127 (2.6)

⬍0.0001 0.92 (continued)

346

Table 1. (continued).

There were initially 46,571 individuals 65 years of age or older with a diagnosis of asthma and a prescription claim for either FSC or ICS who met the enrollment criteria before and after the prescription claim date; 10,837 study subjects (5994 FSC, 4843 ICS) were identified who further met the criteria of a pharmacy claim preand postindex date and had no exclusionary conditions (Figure). Demographic and comorbid characteristics in the baseline preindex period are shown in Table I. Similarities between the FSC and ICS groups included mean age (70.5 and 70.4 years, respectively), proportion of female subjects (64.8% and 65.5%, respectively), and the percentage living in the Mid-Atlantic (30.4% and 33.5%, respectively) and Mountain (7.8% and 7.6%, respectively) regions. A higher percentage of subjects in the ICS treatment group came from the New England area compared with the FSC group (23.4% vs 14.6%, P ⬍ 0.0001) and fewer from the South Atlantic and South Central regions (14.9% vs 24.4%, P ⬍ 0.0001). Asthma diagnosis occurred in the winter or spring for ⬎60% of patients in each group. The most prevalent comorbidities in each group included uncomplicated hypertension, heart disease, and uncomplicated diabetes. Treatment groups were similar in baseline asthmarelated utilization (Table II), with similar asthma-related ED visits, IP, or OCS claims (FSC ⫽ 33.0%, ICS ⫽ 32.6%; P ⫽ 0.69); however, the ICS group had a higher percentage with any ED visit (3.9% vs 2.9%, P ⫽ 0.007). Patients in the ICS group had slightly higher utilization

R.H. Stanford et al.

690,244 individuals with at least 1 asthma claim (primary or secondary DX) in the study period (between July 1, 2001 and June 30, 2008)

365,679 individuals with a prescription claim for FSC (n = 206,340) or ICS (n = 159,339)

The American Journal of Geriatric Pharmacotherapy

similar to those occurring ⱖ30 days. After adjustment for covariates, initial treatment with FSC was associated with a 23% lower risk of an IP/ED visit (HR ⫽ 0.77; 95% CI, 0.62– 0.96; P ⫽ 0.02) compared with ICS (Table IV).

Secondary Outcomes 46,571 individuals with continuous enrollment for 12 months prior to and 3 months following the index date and ≥ 65 on index date (FSC = 24,602, ICS = 21,969)

19,536 individuals with an asthma diagnosis (primary or secondary) prior to index date (FSC = 11,153, ICS = 8,383)

18,404 individuals with any pharmacy fill pre- and postindex date (FSC = 10,484, ICS = 7,920)

17,448 individuals without respiratory cancers and bronchiectasis and with values for all variables (FSC = 9,868, ICS = 7,580)

10,837 asthmatic subjects without a COPD diagnosis (FSC = 5,994, ICS = 4,843)

Figure. Flow diagram for group selection. COPD ⫽ chronic obstructive pulmonary disease; DX ⫽ diagnosis; FSC ⫽ fluticasone propionate/salmeterol combination; ICS ⫽ inhaled corticosteroids.

Asthma-related ED visits occurred in 1.5% of the FSC group compared with 1.8% in the ICS group (P ⫽ 0.18). Conversely, an asthma-related IP event rate was 1.7% in the FSC group compared with 2.6% in the ICS group (P ⬍ 0.001). After adjustment for covariates, initial treatment with FSC was associated with a 23% lower risk of an IP visit (HR ⫽ 0.72; 95% CI, 0.54 – 0.95; P ⫽ 0.02). No differences in risk were observed for ED visits (HR ⫽ 0.94; 95% CI, 0.68 –1.29; P ⫽ 0.69) or for filling an OCS prescription (HR ⫽ 1.02; 95% CI, 0.94 – 1.11; P ⫽ 0.64) during the follow-up period (Table IV). Use of medications differed widely by group. More ICS patients used a SABA (40.6% vs 32.4%, P ⬍ 0.001), LABA (16.9% vs 2.0%, P ⬍ 0.001), and MTYs (6.1% vs 3.5%, P ⬍ 0.001) than FSC patients. The ICS group had a higher percentage of patients with any use of ipratropium (12.4% vs 10.0%, P ⬍ 0.001), but lower percentages of iotropium bromide (2.2% vs 4.3%, P ⬍ 0.001) and LTM (25.3% vs 27.6%, P ⫽ 0.008) use. The percentage of ICS patients who later used FSC was also higher (9.3% vs 5.7%). There were no differences in mean total asthma-related costs between the 2 treatment groups.

DISCUSSION of respiratory-related medications except for use of leukotriene modifiers/inhibitors (LTM) in the preindex baseline period. Treatment groups were similar in baseline asthma-related utilization, although the ICS group had a higher percentage with any ED visit (3.9% vs 2.9%, P ⫽ 0.007). There was no difference between groups in mean total medical cost, the sum of medical services cost and pharmacy cost, before the index date.

Primary Outcomes During the follow-up observation period, unadjusted percentages of patients with any asthma-related events occurring ⱖ30 days post index in the FSC group were compared with the ICS group and are shown in Table III. Percentages with any combined IP/ED event for FSC and ICS were, respectively, 2.8% and 4.0% (P ⬍ 0.001). The proportion of patients in each treatment arm who experienced events beginning 1 day post index are also shown, and differences between the groups are

In this retrospective longitudinal study of Medicare-eligible asthma patients initiating either FSC or ICS therapy, patients who initiated FSC had a lower risk of the combined event of a IP/ED visit ⬎30 days after treatment initiation. Similarly, FSC patients had a lower risk of hospitalization alone compared with ICS patients. However, after censoring use of alternative medications, the statistical significance of the difference between the 2 groups for either of these outcomes was lost. This may have been due to a reduction in the number of events after imposing the more strict censoring criteria. No significant differences were found on ED visit alone or for a prescription claim for an OCS between FCS and ICS. In addition, a similar effect was observed in a sensitivity analysis when events were allowed to occur between day 1 and 29, in addition to those ⬎30 days in the primary analysis. Nearly half of the ICS group used SABA during the index date and preindex period compared with

347

The American Journal of Geriatric Pharmacotherapy

R.H. Stanford et al.

Table II. Preindex asthma-related health care resource utilization and costs. FSC (n ⫽ 5994)

ICS (n ⫽ 4843)

P*

220 (3.7) 175 (2.9) 1841 (30.7) 1976 (33.0)

162 (3.3) 187 (3.9) 1448 (29.9) 1579 (32.6)

0.36 0.007 0.32 0.69

2305 (38.5) 227 (3.8) 1333 (22.2) 33 (0.6) 199 (3.3) 116 (1.9) 584 (9.7)

2369 (48.9) 819 (16.9) 1013 (20.9) 64 (1.3) 290 (6.0) 57 (1.2) 588 (12.1)

⬍0.0001 ⬍0.0001 0.10 ⬍0.0001 ⬍0.0001 0.002 ⬍0.0001

548.50 (3171) 156.70 (307) 705.20 (3190)

514.10 (3246) 191.60 (324) 705.70 (3271)

0.58 ⬍0.0001 0.99

221 (3.7) 184 (3.1) 161 (2.7) 204 (3.4) 216 (3.6) 216 (3.6) 233 (3.9) 244 (4.1) 218 (3.6) 4097 (68.4)

140 (2.9) 132 (2.7) 128 (2.6) 128 (2.6) 136 (2.8) 137 (2.8) 142 (2.9) 153 (3.2) 151 (3.1) 3596 (74.3)

0.02 0.29 0.89 0.02 0.02 0.02 0.007 0.01 0.14 ⬍0.0001

Patients with any utilization history by category Asthma-related utilization Hospitalization Emergency department Oral corticosteroid claim Any of above events Asthma-related pharmacy utilization Short-acting ␤-agonists Long-acting ␤-agonists Leukotriene modifiers/inhibitors Mast-cell stabilizers Methylxanthines Tiotropium bromide Ipratropium (IPR and IPA) Preindex costs, mean (SD) for population), $ Asthma-related utilization Medical services Pharmacy Total medical charges Follow-up, mo 3 4 5 6 7 8 9 10 11 12

FSC ⫽ fluticasone propionate/salmeterol combination; ICS ⫽ Inhaled corticosteroids; IPR ⫽ ipratropium bromide. *P values were calculated using the ␹2 test for categorical data and t test for continuous data.

slightly less than 40% of the FSC group (ICS ⫽ 48.9%, FSC ⫽ 38.5%; P ⬍ 0.001). Use of SABA in the postindex period decreased to 40.6% in the ICS group and to 32.4% in the FSC group. Guidelines for management of persistent asthma suggest initiating therapy with low-dose ICS and then moving to a combination of ICS and LABA if needed. Previous studies have found that FSC should be considered in patients with asthma whose symptoms are inadequately controlled with ICS therapy alone.33 Evidence has also supported the use of combination therapy such as FSC over ICS.26 This switch to FSC among the ICS group may indicate that the use of combination ICS

348

and LABA therapy might prove more beneficial in managing patients’ asthma. Similarly, a switch in FSC users to ICS may indicate a step-down in therapy after asthma stabilization. LABA use among asthma patients can be a concern when used as monotherapy. In this longitudinal study, concurrent use of medications was not examined, nor was LABA monotherapy use explored because this was beyond the scope of this study. However, use of LABA was low in the preindex period: 3.8% in the FSC group and 16.9% in the ICS group. There were LTM or MTY prescriptions in 36% of each group.

R.H. Stanford et al.

The American Journal of Geriatric Pharmacotherapy

Table III. Postindex (unadjusted) asthma-related health care resource utilization and costs. Parameter*

FSC (n ⫽ 5994)

ICS (n ⫽ 4843)

P†

Asthma-related utilization, 30⫹ days‡ Hospitalization ED Oral corticosteroid claim Either hospital or ED

100 (1.7) 87 (1.5) 1366 (22.8) 167 (2.8)

128 (2.7) 86 (1.8) 1101 (22.7) 196 (4.1)

⬍0.001 0.18 0.95 ⬍0.001

Asthma-related utilization (from 1 day post index) Hospitalization ED Oral corticosteroid claim Either hospital or ED

127 (2.1) 110 (1.8) 1755 (29.3) 216 (3.6)

154 (3.2) 100 (2.1) 1416 (29.2) 235 (4.9)

⬍0.001 0.39 0.96 0.001

Other asthma-related pharmacy utilization Use of alternate treatment (ICS or FSC) Short-acting ␤-agonists Long-acting ␤-agonists Leukotriene modifiers/inhibitors Mast-cell stabilizers Methylxanthines Tiotropium bromide Ipratropium

339 (5.7) 1945 (32.4) 121 (2.0) 1654 (27.6) 25 (0.4) 210 (3.5) 260 (4.3) 602 (10.0)

449 (9.3) 1965 (40.6) 819 (16.9) 1227 (25.3) 62 (1.3) 296 (6.1) 107 (2.2) 600 (12.4)

⬍0.0001 ⬍0.0001 ⬍0.0001 0.008 ⬍0.0001 ⬍0.0001 ⬍0.0001 ⬍0.0001

326.70 (2471) 1017.40 (955) 1344.10 (2716)

464.50 (3006) 807.20 (991) 1271.80 (3243)

0.01 ⬍.0001 0.21

Costs, mean (SD) for population, $ Asthma-related costs Medical services Pharmacy Total asthma-related costs

ED ⫽ emergency department; FSC ⫽ fluticasone-salmeterol combination; ICS ⫽ inhaled corticosteroid. *Values shown are number (%) unless otherwise stated. † P values were calculated using the ␹2 test for categorical data and t test for continuous data. ‡ Note that denominators for percentages for the 30-day events do not include individuals having events in the first 30 days. Denominators are as follows: hospitalization, FSC ⫽ 5967, ICS ⫽ 4817; ED visit, FSC ⫽ 5971, ICS ⫽ 4829; oral corticosteroid claim, FSC ⫽ 5595, ICS ⫽ 4509; hospitalization or ED, FSC ⫽ 5945, ICS ⫽ 4804.

Effective pharmacotherapy in older adults with asthma can be difficult to achieve due to physiologic issues related to aging and increased comorbidity.7,25 One asthma medication that has received increased attention in the past decade is theophylline. Although it is considered an effective bronchodilator that may have anti-inflammatory effects, its use has decreased over the past 10 years. Older adults may have the most difficulty with theophylline clearance and be at greater risk for drug– drug interaction adverse events.5 In this study of older adults with asthma, use of MTYs (which include theophylline) was low (⬍7%) both before and after initiation of ICS or FSC, although use was somewhat higher among the ICS group in both periods (P ⬍ 0.0001).

This study is not without limitations inherent to observational study designs using administrative data. In population-based research, the recommendation is to infer asthma severity based on the lowest level of treatment needed for control. Information was not available on initial severity measurements for patients or on presenting symptoms; thus, the basis for treatment therapy is unknown. Additionally, we could not determine how much medication (FSC, ICS, OCS, SABA) was actually used because exposure data were derived from prescription drug claims. It is possible that the eligibility criteria of our study may have caused selection bias because we did not capture uninsured patients and those with no pharmacy claims for the medications of interest during the observational period. Furthermore, we rely on the

349

The American Journal of Geriatric Pharmacotherapy

R.H. Stanford et al.

Table IV. Adjusted risk of asthma-related health-care utilization: FSC compared with ICS.*

Time to hospitalization Time to ED visit Time to hospitalization/ED visit Time to OCS fill

HR (95% CI)

P

0.72 (0.54–0.95) 0.94 (0.68–1.29) 0.77 (0.62–0.96) 1.02 (0.94–1.11)

0.02 0.69 0.02 0.64

ED ⫽ emergency department; FSC ⫽ fluticasone-salmeterol combination; ICS ⫽ inhaled corticosteroid; OCS ⫽ oral corticosteroid. *Adjusted for age, sex, age, season in which index date occurred, geographic region, selected preperiod comorbidities, any fill in the preperiod for shortacting ␤-agonists, leukotriene receptor antagonists, theophylline or anticholinergics, quantities of OCS, asthma-related hospitalizations/ED visits, outpatient visits.

health care system and the accuracy of coding to measure exposures and outcomes, which may be inherently flawed. Another potential limitation is that we first identified the asthma sample by the use of 1 ICD-9 code, and the sensitivity/specificity of this approach compared with chart review detection of this disease is unknown. The incidence of asthma exacerbations reported here may underestimate the true incidence of the event if asthma-related exacerbations are coded as another respiratory disease event, such as COPD and respiratory failure. Likewise, although OCS is the main treatment for an asthma-related exacerbation, these medications are used for a variety of other diseases, and it is not possible to determine whether the medication was in response to a specific event using these data.

CONCLUSIONS Our results suggest that in Medicare-eligible older adults with asthma from a managed care population, patients who initiated FSC therapy experienced lower rates of asthma-related events over a subsequent 12month period, with no increase in asthma-related costs, compared with patients who initiated ICS therapy. In addition, FSC initiation was associated with lower rates of asthma-related serious exacerbations and better asthma control, as measured by SABA use versus patients who initiated ICS therapy.

ACKNOWLEDGMENTS Dr. Stanford was responsible for the study design, data interpretation, and data analysis and was an investigator and writer/reviewer. Dr. Blanchette was responsible for the study design and was an investigator and writer/ reviewer. Ms. Roberts and Mr. Peterson were responsi-

350

ble for data analysis and data interpretation and were writers/reviewers. Dr. Fuhlbrigge was an investigator and was responsible for data interpretation. This study, ADA112605, was funded by GlaxoSmithKline.

CONFLICTS OF INTEREST Dr. Stanford is an employee of and owns stock in GlaxoSmithKline. Drs. Blanchette and Fuhlbrigge are paid investigators for GlaxoSmithKline and Ms. Roberts and Mr. Petersen receive payment from GlaxoSmithKline for analytical support. The authors have indicated that they have no other conflicts of interest regarding the content of this article.

REFERENCES 1. Pleis J, Lucas J, Ward B. Summary Health Statistics for U.S. Adults: National Health Interview Survey, 200810. Atlanta, Ga: National Center for Health Statistics; 2009. 2. Rodrigo GJ, Rodrigo C, Hall JB. Acute asthma in adults: a review. Chest. 2004;125:1081–1102. 3. Weiss KB, Sullivan SD. The health economics of asthma and rhinitis. I. Assessing the economic impact. J Allergy Clin Immunol. 2001;107:3– 8. 4. American Lung Association. Asthma in adults fact sheet. 2010; www.lungusa.org/lung-disease/asthma/resources/ facts-and-figures/asthma-in-adults.html. Accessed June 3, 2010. 5. National Heart Lung and Blood Institute. National Asthma Education and Prevention Program, Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma. Bethesda, Md: National Institutes of Health: National Asthma Education and Prevention Program (NAEPP) Coordinating Committee, 2007. 6. Paltiel AD, Fuhlbrigge AL, Kitch BT, et al. Cost-effectiveness of inhaled corticosteroids in adults with mild-tomoderate asthma: results from the Asthma Policy Model. J Allergy Clin Immunol. 2001;108:39 – 46. 7. Braman S. Asthma in the elderly. Clin Geriatr Med. 2003; 19:57–75. 8. Barua P, O’Mahony M. Overcoming gaps in the management of asthma in older patients: new insights. Drugs Aging. 2005;22:1029 –1059. 9. Global Initiative for Asthma. Global strategy for asthma management and prevention. 2009. www.ginasthma.org. Accessed June 5, 2010. 10. National Asthma Education and Prevention Program (NAEPP) Expert Panel. National Asthma Education and Prevention Program (NAEPP) expert panel report 3: guidelines for the diagnosis and management of asthma. J Allergy Clin Immunol. 2007;120:S94 –S138.

R.H. Stanford et al.

11. Shrewsbury S, Pyke S, Britton M. Meta-analysis of increased dose of inhaled steroid or addition of salmeterol in symptomatic asthma (MIASMA). BMJ. 2000;320:1368 – 1373. 12. Woolcock A, Lundback B, Ringdal N, Jacques L. Comparison of addition of salmeterol to inhaled steroids with doubling the dose of inhaled steroids. Am J Respir Crit Care Med. 1996;153:1481–1488. 13. Greening A, Ind P, Northfield M, Shaw G. Added salmeterol versus higher-dose corticosteroid in asthma patients with symptoms on existing inhaled corticosteroid. Allen & Hansburys Limited UK Study Group. Lancet. 1994; 344:219 –224. 14. Lalloo U, Malolepszy J, Kozma D, et al. Budesonide and formoterol in a single inhaler improves asthma control compared with increasing the dose of corticosteroid in adults with mild-to-moderate asthma. Chest. 2003;123: 1480 –1487. 15. Lamanske R Jr, Sorkness C, Mauger E, et al. Inhaled corticosteroid reduction and elimination in patients with persistent asthma receiving salmeterol: a randomized controlled trial. JAMA. 2001;285:2594 –2603. 16. Masoli M, Weatherall MW, Holt S, Beasley R. Moderate dose of inhaled corticosteroids plus salmeterol versus higher doses of inhaled corticosteroids in symptomatic asthma. Thorax. 2005;60:730 –734. 17. Postma D, Kerstjens H, ten Hacken N. Inhaled corticosteroids and long-acting beta-agonists in adult asthma: a winning combination in all? Naunyn-Schmiedebergs Arch Pharmacol. 2008;378:203–215. 18. Randall T. Demographers ponder the aging of the aged and await unprecedented looming elder boom. JAMA. 1993;269:2331–2332. 19. Plaza V, Serra-Batlles J, Ferrer M, Morejon E. Quality of life and economic features in elderly asthmatics. Respiration. 2000;67:65–70. 20. Burrows B, Barbee R, Cline M, et al. Characteristics of asthma among elderly adults in a sample of the general population. Chest. 1991;100:935–942. 21. Diette G, Krishnan J, Dominici F, et al. Asthma in older patients: factors associated with hospitalization. Arch Intern Med. 2002;162:1123–1132.

The American Journal of Geriatric Pharmacotherapy

22. Nejjari C, Tessier J, Barberger-Gateau P, et al. Functional status of elderly people treated for asthma-related symptoms: a population based case-control study. Eur Respir J. 1994;7:1077–1083. 23. Quadrelli S, Roncoroni A. Is asthma in the elderly really different? Respiration. 1998;65:347–353. 24. Bailey W, Richards J Jr, Brooks C, et al. Features of asthma in older adults. J Asthma. 1992;29:21–28. 25. Braman S. Drug treatment of asthma in the elderly. Drugs. 1996;51:415– 423. 26. Stanford RH, Fuhlbrigge A, Riedel A, et al. An observational study of fixed dose combination fluticasone propionate/salmeterol or fluticasone propionate alone on asthma-related outcomes. Curr Med Res Opin. 2008;24: 3141–3148. 27. Currie GP, Lee DKC, Wilson AM. Effects of dual therapy with corticosteroids plus long acting beta(2)-agonists in asthma. Respir Med. 2005;99:683– 694. 28. Jilinskaia E, Norton S, Johnson C. Comparison of Methods for Imputation of Paid Amounts in Medical Insurance Data. Watertown, Mass: Joint Statistical Meetings, 2002. 29. Silver HS, Blanchette CM, Kamble S, et al. Quarterly assessment of short-acting ␤2-adrenergic agonist use as a predictor of subsequent health care use for asthmatic patients in the United States. J Asthma. 2010;47:660 – 666. 30. Elixhauser A, Steiner C, Harris D, Coffey R. Comorbidity measures for use with administrative data. Med Care. 1998;36:8 –27. 31. Mapel D, Robinson S, Lydick E. A comparison of healthcare costs in patients with chronic obstructive pulmonary disease using lightweight portable oxygen systems versus traditional compressed-oxygen systems. Respir Care. 2008;53:1169 –1175. 32. Breslow NE. Discussion of the paper by D.R. Cox. J R Statist Soc B. 1972;34:216 –217. 33. Delea TE, Hagiwara M, Dalal AA, et al. Healthcare use and costs in patients with chronic bronchitis initiating maintenance therapy with fluticasone/salmeterol vs other inhaled maintenance therapies. Curr Med Res Opin. 2009;25:1–13.

Address correspondence to: Richard H. Stanford, PharmD, MS, GlaxoSmithKline, Five Moore Drive, Research Triangle Park, NC 27709. E-mail: [email protected]

351