COPD-Related Healthcare Utilization and Costs After Discharge From a Hospitalization or Emergency Department Visit on a Regimen of Fluticasone Propionate-Salmeterol Combination Versus Other Maintenanc

, , ,
The American Journal of Managed Care, March 2011, Volume 17, Issue 3

Initiation of fluticasone propionate–salmeterol after a COPD-related hospitalization or ED visit decreased the risk of a recurrent event and reduced COPD-related medical costs.

Objectives: To quantify healthcare use and costs associated with chronic obstructive pulmonary disease (COPD) among patients discharged from a COPD-related hospitalization or emergency department (ED) visit on a regimen of fluticasone propionate—salmeterol combination versus other inhaled maintenance therapies.

Study Design: Retrospective cohort study.

Methods: Managed care enrollees with an index hospitalization (with a primary or secondary [ie, in the second position] diagnosis of COPD) or ED visit (with a primary diagnosis of COPD) were identified for placement into study cohorts during a 60-day period following the index date. Time to COPD-related events and healthcare costswere compared during up to 1 year of follow-up between the 2 cohorts.

Results: The sample comprised 5677 patients (1291 in the fluticasone propionate—salmeterol cohort and 4386 in the other maintenance therapies cohort). The adjusted rate of COPD-related hospitalizations or ED visits was 35% lower in the fluticasone propionate–salmeterol cohort (P <.05). Adjusted COPD-related total (medical plus pharmacy) costs were lower in the fluticasone propionate–salmeterol cohort ($240 vs $279 per patient per month, P <.05), mostly because of lower medical costs ($113 vs $160 per patient per month, P <.05). Pharmacy costs did not differ between fluticasone propionate–salmeterol and other maintenance therapies. Results were similar in the subset of patients 65 years or older.

Conclusions: Initiation of fluticasone propionate— salmeterol after discharge from a COPD-related hospitalization or ED visit significantly reduced the risk of a recurrent event during the ensuing months and decreased COPD-related medical costs, without an increase in COPD-related pharmacycosts, in a real-world setting.

(Am J Manag Care. 2011;17(3):e55-e65)

The occurrence of an exacerbation of chronic obstructive pulmonary disease (COPD) increases the risk of subsequent exacerbations.

  • Risk of rehospitalization for a recurrent exacerbation is particularly high during the months after an initial exacerbation-related hospitalization.

  • Compared with other maintenance therapies, initiation of fluticasone propionate—salmeterol after discharge from a COPD-related hospitalization or emergency department (ED) visit significantly reduced the risk of a COPD-related return to the hospital or ED during the ensuing months and decreased COPD-related medical costs, without an increase in COPD-related pharmacy costs, in a real-world setting.

The fourth leading cause of death (after coronary artery disease, cancer, and stroke), chronic obstructive pulmonary disease (COPD) affects approximately 12 million adults in the United States, where COPD accounts for approximately 1.5 million emergency department (ED) visits, 726,000 hospitalizations, and 119,000 deaths annually and an estimated $32 billion in direct and indirect costs.1,2 The personal and economic effects of COPD in developed countries derive primarily from acute exacerbations, which cause significant morbidity and mortality and are associated with poor prognosis.3-9 Because acute exacerbations of COPD often require intensive care, inpatient costs of COPD are particularly high. For example, per capita expenditures for hospitalization of patients with COPD exceed those for patients without COPD by almost 3-fold.10

Given the large human and economic tolls of acute exacerbations of COPD, prevention and treatment of acute exacerbations are primary goals of therapy.8,11,12 Of several categories of maintenance therapy available for the management of COPD, combination inhaled corticosteroid—long-acting beta-agonist (LABA) treatment has recently emerged as offering significant promise in the control of acute exacerbations. Combination inhaled corticosteroid–LABA therapy in a single inhaler was found in Cochrane reviews of randomized controlled clinical trials to be more effective than placebo,13 LABAs,14 and inhaled corticosteroids15 at reducing the rate of exacerbations of COPD. In the Towards a Revolution in COPD Health (TORCH) study, a landmark trial included in the Cochrane reviews, the inhaled corticosteroid—LABA combination fluticasone propionate–salmeterol significantly reduced the annual rate of exacerbations by 25% versus placebo over a 3-year period and was associated with a 17.5% reduction in the risk of death (P = .05 vs placebo).16 All-cause mortality rates were 12.6% in the fluticasone propionate— salmeterol group, 15.2% in the placebo group, 13.5% in the salmeterol group, and 16.0% in the fluticasone group. It has been suggested that LABAs may increase the effectiveness of inhaled corticosteroids by amplifying their anti-inflammatory effects.17

The efficacy of combination inhaled corticosteroid—LABA therapy at reducing exacerbations seems to translate to less healthcare resource use and lower total costs (medical plus pharmacy) compared with component monotherapies or other classes of COPD maintenance therapy.18-26 In a retrospective analysis, combination inhaled corticosteroid—LABA therapy was associated with a lower risk of rehospitalization within 1 year of an index COPD-related hospitalization than treatment with LABAs only, inhaled corticosteroids only, or short-acting beta-agonists (SABAs) only.20 Furthermore, in observational studies19,21-24,26 using medical and pharmacy claims, combination inhaled corticosteroid—LABA therapy, particularly fluticasone propionate–salmeterol, was associated with a significantly lower risk of COPD-related hospitalizations and with lower healthcare costs compared with the use of anticholinergic bronchodilators in various populations of patients with COPD. Cost-benefit fluticasone propionate–salmeterol data have also been reported from controlled clinical trials. For example, the estimated cost per quality-adjusted life-year reflected greater cost-effectiveness of fluticasone propionate–salmeterol than either component monotherapy alone in an analysis of data from the TORCH trial.25

Current guidelines recommend that patients with COPD who have exacerbations should be started on regular treatment with an inhaled medication proven to reduce the risk of exacerbations.2 What is unknown is whether starting with a combined therapy, such as fluticasone propionate—salmeterol, has advantages over monotherapies, such as tiotropium bromide, LABAs, ipratropium bromide, and inhaled corticosteroids. Furthermore, while health outcomes and costs with fluticasone propionate–salmeterol vs other maintenance therapies have been thoroughly elucidated, the effect of fluticasone propionate–salmeterol therapy vis-à-vis other classes of maintenance therapy (besides its monotherapy components) on health outcomes and costs has not been systematically studied during the months immediately after an exacerbation. The occurrence of a COPD exacerbation increases the risk of subsequent exacerbations, and the risk of rehospitalization for a recurrent exacerbation is particularly high during the months after an initial exacerbation-related hospitalization.7-9 In a prospective study8 of 129 patients hospitalized for an acute exacerbation of COPD, 58.5% were readmitted to the hospital over the ensuing year. The study reported herein was conducted to quantify healthcare utilization and costs attributed to COPD after discharge from a COPD-related hospitalization or ED visit on a regimen of fluticasone propionate—salmeterol combination versus other maintenance therapies.

METHODS

This observational retrospective cohort study was conducted to compare postdischarge COPD-related healthcare use and costs among patients discharged from a COPD-related hospitalization or ED visit on a regimen of fluticasone propionate—salmeterol combination versus other inhaled maintenance therapies. Figure 1 shows the study design.

Data Source

Data were obtained from the IMS LifeLink Health Plans Claims Database,27 an integrated source of fully adjudicated managed care claims containing data from more than 90 managed healthcare plans covering over 60 million lives across the United States. Data available for each patient include inpatient and outpatient diagnoses (by International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis code) and procedures (in Current Procedural Terminology, Version 4 and Health Care Financing Administration Common Procedural Coding System formats), as well as retail and mail-order prescription records, which include the National Drug Code and the quantity dispensed. Charged, allowed, and paid amounts are available for all services rendered, as are dates of service for all claims. Additional data elements include demographic variables (age, sex, and geographic region), payer type (ie, private [such as health maintenance organization or preferred provider organization] or public [such as Medicaid]), provider specialty, and start and stop dates for plan enrollment. The data are fully deidentified and are compliant with the Health Insurance Portability and Accountability Act of 1996.

Study Sample

The target population was managed care enrollees who had an index event of at least 1 hospitalization with a primary or secondary [ie, in the second position] diagnosis of COPD or at least 1 ED visit with a primary diagnosis of COPD between January 1, 2003, and July 31, 2008 (the study period) and who initiated fluticasone propionate—salmeterol combination at 250/50 mcg or other maintenance therapies (ie, inhaled corticosteroids, LABAs, tiotropium, or ipratropium alone or in combination with albuterol) during the treatment assessment period (ie, within 60 days after the date of discharge from the index event). The index date was defined as the date of discharge from the index event. To be eligible for analysis, patients had to be at least 40 years old and have continuous eligibility during the preindex period (defined as the 1-year period before the admission date of the index event), the treatment assessment period, and the follow-up period (a maximum of 1 year after the end of the treatment assessment period). Patients were excluded from the sample who during the preindex, treatment assessment, or follow-up periods had any of the following comorbid conditions: respiratory tract cancer, cystic fibrosis, fibrosis from tuberculosis, bronchiectasis, pneumonoconiosis, pulmonary fibrosis, pulmonary tuberculosis, or sarcoidosis. Patients were also excluded if during the treatment assessment period they used or switched to other maintenance therapies different from the one that they initiated. For example, if patients started a tiotropium regimen but then switched or added fluticasone propionate–salmeterol, they were excluded from the study. Finally, patients were excluded if they had a hospitalization, COPD-related ED visit during the treatment assessment period, or physician visit with a prescription for an oral corticosteroid or antibiotic within 3 days of the visit during the treatment assessment period.

Outcomes

Outcomes of interest were compared between the following cohorts, defined by the maintenance therapies initiated during the treatment assessment period: the fluticasone propionate—salmeterol combination cohort (which initiated fluticasone propionate–salmeterol) and the other maintenance therapies cohort (which initiated inhaled corticosteroids, LABAs, tiotropium, or ipratropium). Outcomes of interest were COPD-related events (defined as hospitalizations, COPDrelated ED visits, or physician visits with a prescription for an oral corticosteroid or antibiotic within 3 days of the visit) and COPD-related total (medical plus pharmacy), medical, and pharmacy costs. Emergency department visits, physician visits, and hospitalizations were assumed to be COPD related if the claim included a primary diagnosis of COPD (discharge diagnosis of COPD for hospitalizations).

For COPD-related events, time to event (defined as the number of days between the start of the follow-up period and the date of the first COPD-related event) was computed. The COPD-related medical costs were computed from the paid amounts of medical claims with a primary diagnosis code for COPD. The COPD-related pharmacy costs were computed from the paid amounts of COPD-related prescription medications (including anticholinergics, SABAs, LABAs, inhaled corticosteroids, combination inhaled corticosteroids—LABAs, methylxanthines, oral corticosteroids, and antibiotics for respiratory tract infections), identified using the National Drug Code.

Study outcomes were assessed during a follow-up period of variable duration. The follow-up period was defined as the period from the index date to the first date of any of the following events: the end of the study period, the end of continuous eligibility in the health plan, the end of the follow-up period (which lasted a maximum of 1 year), a switch to any study medication different from the medication dispensed during the treatment assessment period, more than a 60-day gap between the end of days’ supply of the preceding prescription and the fill date of the next consecutive prescription, or a COPD-related hospitalization or ED visit, or physician visit with a prescription for an oral corticosteroid or antibiotic within 3 days of the visit. To accurately capture resource use attributable to both cohorts, costs were assessed during the entire length of variable follow-up and were not censored with a COPD-related event. Costs were computedon a per-month basis because of interpatient differences in the length of follow-up and were standardized to 2008 US dollars using the medical component of the consumer price index.

Cohorts were also compared with respect to pretreatment characteristics, including the following: demographics (age, sex, and US census region), comorbidities during the preindex period (Charlson Comorbidity Index28 and asthma), and proxies for COPD severity2 during the preindex period (number of canisters of inhaled SABAs, number of canisters of inhaled ipratropium, number of prescriptions for oral corticosteroids, number of subclasses of maintenance therapy, use of home oxygen therapy, number of hospitalizations or ED visits for COPD, presence of an intensive care unit stay for COPD, and number of physician visits for COPD). Age, sex, and US census region at the index date were obtained from enrollment files. The Dartmouth-Manitoba adaptation of the Charlson Comorbidity Index28 (a weighted index of 19 chronic medical conditions that predict mortality, postoperative complications, and length of hospital stay) was calculated for each patient based on diagnoses reported during the preindex period. The COPD codes that are generally included in computing the Charlson Comorbidity Index were excluded. Patients were classified as having asthma if they had at least 1 hospitalization or ED visit or at least 2 physician visits with a diagnosis of asthma in any field. The number of canisters of SABAs was computed by dividing the quantity dispensed in milligrams by the milligrams per canister. The use of home oxygen therapy was categorized as a binary variable (use vs no use) based on Current Procedural Terminology codes for home oxygen therapy in medical claims. The presence of an intensive care unit stay with a primary diagnosis code for COPD was identified via revenue codes.

Statistical Analysis

Data were analyzed for the sample as a whole and for the subset of patients who were 65 years or older as of the index date. Pretreatment characteristics were summarized using descriptive statistics. Inferential statistics (X2 test for categorical variables and t test or Mann-Whitney test for continuous variables) were used to quantify differences between cohorts. Unadjusted outcomes of COPD-related events assessed during the follow-up period were reported as the rate per 100 person-years of follow-up. Log-rank test was used to assess statistical differences in unadjusted time to COPD-related events, and P values were used to evaluate statistical differences in the unadjusted rate per 100 person-years of follow-up. Survival analysis techniques were used to determine differences in time to COPD-related events after controlling for all other baseline covariates. The proportional hazards assumption was tested using the global test of proportional hazards to assess the appropriateness of a Cox proportional hazards regression. Hazard ratios from survival analysis models provided an estimate of the adjusted differences in time to COPD-related events.

Unadjusted costs per month were summarized using descriptive statistics for all costs and cost components. Differences between cohorts in adjusted COPD-related total, medical, and pharmacy costs were assessed using multivariate regression models. Because a significant proportion of the sample had zero costs and because the cost distribution was positively skewed for those with costs, the following 2-part model was adopted: (1) a logistic model and (2) a generalized linear model using a gamma distribution with a log link. Adjusted costs estimated from 2-part models were computed by multiplying the adjusted probability obtained from the logistic regression model (part 1) by the predicted cost from the generalized linear model (part 2). Monthly costs were used as the dependent variable in the generalized linear model, and the other maintenance therapies cohort was used as the reference cohort in the model.29

RESULTS

Sample

A total of 31,759 persons in the database had a hospitalization or COPD-related ED visit during the study period and at least 1 prescription claim for a study medication during the treatment assessment period (Table 1). Of this group, 5677 patients (1291 in the fluticasone propionate—salmeterol cohort and 4386 in the other maintenance therapies cohort) met all eligibility criteria and comprised the study sample. Most of the 4386 patients in the other maintenance therapies cohort (almost 80%) were using short-acting or long-acting anticholinergics, and the remaining patients (a small percentage) were using inhaled corticosteroids alone and LABAs alone. The subset of patients 65 years or older included 713 patients in the fluticasone propionate–salmeterol cohort and 2567 patients in the other maintenance therapies cohort.

Pretreatment Characteristics

Compared with the other maintenance therapies cohort, the fluticasone propionate—salmeterol cohort was significantly younger and included more women (Table 1). More patients in the fluticasone propionate–salmeterol cohort compared with the other maintenance therapies cohort had asthma during the preindex period (25.5% vs 17.4%, P <.05) and were using SABAs (46.6% vs 42.1%, P <.05), but the fluticasone propionate–salmeterol cohort had fewer oral corticosteroid prescriptions (0.8 vs 1.0, P <.05). The use of home oxygen therapy was less frequent in the fluticasone propionate–salmeterol cohort than in the other maintenance therapies cohort (14.3% vs 22.3%, P <.05). The fluticasone propionate–salmeterol cohort used fewer subclasses of maintenance therapy (0.1 vs 0.2, P <.05). During the 1-year preindex period, COPD-related hospitalizations or ED visits were significantly less frequent in the fluticasone propionate–salmeterol cohort than in the other maintenance therapies cohort (5.7% vs 9.5%,

P <.05), primarily because of differences in the frequency of COPD-related ED visits (5.0% vs 9.0%, P <.05). In the subset of patients 65 years or older, the pattern of results was generally similar.

COPD-Related Events

Table 2 gives the adjusted multivariate outcomes in the sample as a whole and in the subset of patients 65 years or older. The adjusted rate of COPD-related hospitalizations or ED visits was 35% lower in the fluticasone propionate—salmeterol cohort than in the other maintenance therapies cohort (hazard ratio [95% confidence interval], 0.649 [0.455-0.926]), a difference driven by a 40% lower adjusted rate of hospitalizations (hazard ratio [95% confidence interval], 0.601 [0.326- 1.109]) and a 35% lower unadjusted rate of COPD-related ED visits (hazard ratio [95% confidence interval], 0.651 [0.434- 0.977]) (P <.05 for the initial and latter hazard ratios). The adjusted rate of COPD-related physician visits combined with a prescription for oral corticosteroids or antibiotics within 3 days was 21% lower in the fluticasone propionate–salmeterol cohort than in the other maintenance therapies cohort (hazard ratio [95% confidence interval], 0.785 [0.646-0.954]) (P <.05). Results in the subset of patients 65 years or older were generally similar to those in the sample as a whole. Figure 2 shows the unadjusted rate of COPD-related hospitalizations or ED visits per 100 person-years in the sample as a whole, and Figure 3 shows the time to any COPD-related event and to any COPD-related hospitalization or ED visit. The switch rate to a study medication different from the medication dispensed on the index date was fairly balanced between the cohorts (270 [20.9%] in the fluticasone propionate—salmeterol cohort vs 1023 [23.3%] in the other maintenance therapies cohort) (P = .07).

COPD-Related Costs

In the sample as a whole, the monthly COPD-related adjusted total costs were significantly lower in the fluticasone propionate—salmeterol cohort than in the other maintenance therapies cohort ($240 vs $279 per patient per month, P < .05), mostly because of lower medical costs ($113 vs $160 per patient per month, P <.05) (Table 2). Similarly, the total monthly unadjusted COPD-related costs were lower in the fluticasone propionate–salmeterol cohort than in the other maintenance therapies cohort ($205 vs $297 per patient per month), mostly because of lower medical costs ($84 vs $171 per patient per month), with no increase in pharmacy costs (Figure 2). Hospitalization costs were a primary driver of the difference in medical costs. As in the sample as a whole, the monthly COPD-related total costs and medical costs in the subset of patients 65 years or older were significantly lower in the fluticasone propionate–salmeterol cohort than in the other maintenance therapies cohort (Table 2).

DISCUSSION

Most of the medical costs of COPD are attributed to hospitalizations and ED visits for acute exacerbations, which occur twice yearly, on average, and are associated with significant risk of persistent disability and death.3,5,6,17 The occurrence of an acute exacerbation of COPD is a major risk factor for subsequent exacerbations requiring hospitalization.7-9 Control of COPD-related morbidity and medical costs depends on reducing the incidence of acute exacerbations and the associated risk of hospitalization and ED visits. In this retrospective cohort study using claims data, initiation of fluticasone propionate—salmeterol after discharge from a COPD-related hospitalization or ED visit was associated with a 35% lower risk of a recurrent event than initiation of other maintenance therapies, including (collectively) inhaled corticosteroids, LABAs, tiotropium, and ipratropium. Most members of the other maintenance therapies cohort were using anticholinergics. This study constitutes a real-world comparative assessment of effectiveness of these medications that differs from less realistic comparator clinical trials. While initiation of fluticasone propionate–salmeterol was associated with a 40% lower risk of hospitalization than initiation of other maintenance therapies in the sample as a whole, this difference was not statistically significant, probably because of the few hospitalizations captured during the postindex period in the context of the censoring rules for the study. The clinical reduction in the risk of subsequent or repeat exacerbations (operationalized as COPD-related hospitalizations, ED visits, or physician visits with a prescription for an oral corticosteroid or antibiotic) with fluticasone propionate–salmeterol vs other maintenance therapies translated to lower monthly COPD-related medical costs, without an increase in COPD-related pharmacy costs. Costs were assessed for the entire length of the variable follow-up period, which did not end with a COPD-related event.

In the subset of patients 65 years or older, fluticasone propionate—salmeterol use was associated with a 39% lower risk of a COPD-related hospitalization or ED visit than other maintenance therapies. In those 65 years or older (as in the sample as a whole), the clinical reduction in the risk of subsequent or repeat exacerbations with fluticasone propionate– salmeterol use also translated to lower monthly COPD-related medical costs, without an increase in COPD-related pharmacy costs. Morbidity associated with COPD seems to increase with age.2 A particularly important target group of efforts to reduce disease burden and to control COPD-related costs is the older Medicare population. This study shows that fluticasone propionate—salmeterol is cost-effective not only among a commercial population 40 years or older but also among an older population that is Medicare eligible.

This study did not restrict the definition of COPD exacerbations captured as outcomes to hospitalizations and ED visits but also considered physician visits with a prescription for an oral corticosteroid or antibiotic within 3 days of the visit as a proxy marker of moderate COPD exacerbation. A similarly inclusive definition was used in recent pivotal trials of COPD exacerbations.11,30 The broad definition of COPD exacerbations provides a sensitive measure of treatment outcomes and reflects the spectrum of manifestations of COPD exacerbations in clinical practice.

A rigorous approach to censoring variables in the risk analysis was taken to ensure that outcomes were not unfairly attributed to treatment. The risk analysis used censoring variables to account for treatment switch or discontinuation, loss of enrollment, and events of interest to ensure that outcomes were not inappropriately attributed to the treatment groups even after the drug was discontinued or treatment failure occurred. The censoring approach also helps alleviate confounding by other treatments. If censoring is not performed on treatment change (a switch or a discontinuation), outcomes may be unfairly associated with the index treatment in either direction. The approach of censoring is analogous to the practice of disallowing concomitant therapies in randomized controlled clinical trials to help ensure that outcomes can be attributed to an effect of study medication rather than of concomitant treatments. Besides censoring, other measures were taken to help ensure the validity of the results. To reduce selection bias in the absence of randomization, the analyses controlled for differences in baseline characteristics in adjusted risk and cost models. In addition, patients having COPD events during the 60-day treatment assessment period were excluded to avoid issues of immortal time bias. This approach also allowed capture of events after the treatment assessment period to allow for a reasonable time for treatment to take effect. The switch rate was found to be fairly balanced between the cohorts, a result suggesting that the probability of an endogeneity bias is small.

To our knowledge, this study is the first to assess COPD-related healthcare utilization and costs with combination inhaled corticosteroid—LABA therapy vs other major classes of maintenance therapy (besides only combination monotherapies) among patients immediately after discharge from a COPD-related hospitalization or ED visit in a real-world observational setting. Furthermore, this study is one of few economic assessments to include patients on a regimen of tiotropium, which was introduced in 2004 in the United States. These results extend findings of a retrospective study20 in which combination inhaled corticosteroid—LABA therapy was associated with a lower risk of rehospitalization within a year of an initial COPD-related hospitalization than treatment with LABAs only, inhaled corticosteroids only, or SABAs only. The results of the present study are also consistent with clinical observations showing that fluticasone propionate–salmeterol compared with other maintenance therapies significantly reduced the rates of exacerbations and exacerbations requiring hospitalization in samples not selected on the basis of their having had a recent COPD-related hospitalization or ED visit.11,16 The present results are also in agreement with previous health economics data showing that fluticasone propionate—salmeterol therapy is associated with lower healthcare resource use and costs than other classes of maintenance therapy in samples not selected on the basis of their having had a recent COPD-related hospitalization or ED visit.19,26 The data from the subset of patients 65 years or older in the present study are also consistent with results of a retrospective observational study26 among 14,689 Medicare beneficiaries 65 years or older in whom initiation of maintenance therapy with fluticasone propionate—salmeterol was associated with a significant reduction in total costs (medical plus pharmacy) relative to costs associated with the use of tiotropium (or with the short-acting bronchodilator therapies ipratropium bromide and ipratropium bromide–albuterol).

As with any observational study, the conclusions herein must be drawn in the context of study limitations. In this study, residual confounding could be operating given the lack of random assignment to a treatment condition; that is, differences between cohorts in study outcomes might be attributed to differences in patient characteristics that were not controlled for in the multivariate analyses. Patients in the fluticasone propionate—salmeterol cohort significantly differed from those in the other maintenance therapies cohort on several pretreatment characteristics. While these factors were adjusted for in the multivariate analyses, the possibility of residual confounding remains. Other limitations include potential errors in the coding of claims, the inability to verify the accuracy of diagnosis codes, and the absence of a means of assessing patient compliance. Because of the lack of reversibility information in claims data, the presence of asthma (which is commonly comorbid with COPD and is often similarly treated) could not be excluded based on diagnosis codes. In the event that these biases (the potential for which is common among observational claims studies) were present, they are likely to have been operating similarly between cohorts. In this case, study outcomes would have been biased toward the null hypothesis of no difference between the cohorts.

In summary, the results of this study show that initiation of fluticasone propionate—salmeterol (compared with other maintenance therapies) after discharge from a COPD-related hospitalization or ED visit significantly reduced the risk of a return to the hospital or ED during the ensuing year in the sample as a whole (comprising patients >40 years old) and in the subset of patients 65 years or older. This reduction in healthcare resource use was realized as a decrease in monthly COPD-related medical costs, without an increase in monthly COPD-related pharmacy costs, associated with fluticasone propionate–salmeterol use compared with other maintenance therapies. Therefore, fluticasone propionate–salmeterol maintenance therapy seems to be a cost-effective treatment option for patients recently experiencing an exacerbation of COPD. The findings support the recommendations of the Global Initiative for Chronic Obstructive Lung Disease2 and other international professional societies for the treatment of patients with COPD. The Global Initiative for Chronic Obstructive Lung Disease2 recommends that all patients having moderate COPD should be treated with a long-acting bronchodilator and that inhaled corticosteroids should be added for patients with severe COPD and frequent exacerbations or hospitalizations. These recommendations are based largely on randomized clinical trials that have proven the efficacy of these treatments in reducing the risk of exacerbations. The results of the present study suggest that the treatment benefits proven in selected clinical trial cohorts extrapolate to patients with COPD who are treated in the general population. The findings of this study also support the current thinking that COPD is a treatable disease and that aggressive therapy results in tangible benefits to the patient.31

Acknowledgments

We acknowledge Jane Saiers, PhD (The WriteMedicine Inc) for assistance with preparing the manuscript. Dr Saiers’ work was funded by GlaxoSmithKline. She prepared the first draft of the manuscript, worked on the references, and helped with the introduction. She also helped collate comments from authors.

Author Affiliations: From GlaxoSmithKline (AAD), Research Triangle Park, NC; Xcenda (MS, AOD), Palm Harbor, FL; and Lovelace Clinic Foundation (DWM), Albuquerque, NM.

Funding Source: The study (ADC112609) was funded by GlaxoSmithKline. Xcenda conducted the study.

Author Disclosures: Dr Dalal reports being an employee of GlaxoSmithKline and owning company stock. Drs Shah and D’Souza report being employees of Xcenda, who conducted this study, and report receiving funding from GlaxoSmithKline. Dr Mapel reports serving as a consultant to and receiving funding from GlaxoSmithKline, Pfizer Inc, Boehringer-Ingelheim, and AstraZeneca. Dr Mapel also reports that the Lovelace Clinic Foundation has received grants for COPD-related research.

Authorship Information: Concept and design (AAD, MS, AOD, DWM); acquisition of data (AAD, MS, AOD); analysis and interpretation of data (AAD, MS, AOD, DWM); drafting of the manuscript (AAD, DWM); critical revision of the manuscript for important intellectual content (AAD, AOD, DWM); statistical analysis (AAD, MS, AOD, DWM); obtaining funding (AAD); and supervision (AAD).

Address correspondence to: Anand A. Dalal, PhD, MBA, GlaxoSmithKline, Research Triangle Park, c/o 5 Moore Dr, Mail Stop 17.1355B, Durham, NC 27709. E-mail: anand.a.dalal@gsk.com.

1. National Heart, Lung, and Blood Institute. Data fact sheet: chronic obstructive pulmonary disease. NIH publication 03-5229. March 2003. http://www.uptakemedical.com/pdfs/copd_fact.pdf. Accessed February 10, 2011.

2. GOLD: the Global Initiative for Chronic Obstructive Lung Disease. Global strategy for diagnosis, management, and prevention of COPD. Updated 2009. http://www.goldcopd.com. Accessed February 15, 2010.

3. Halpern MT, Higashi MK, Bakst AW, Schmier JK. The economic impact of acute exacerbations of chronic bronchitis in the United States and Canada: a literature review. J Manag Care Pharm. 2003;9(4): 353-359.

4. Halpern MT, Polzin J, Higashi MK, Bakst A. The workplace impact of acute exacerbations of chronic bronchitis (AECB): a literature review. COPD. 2004;1(2):249-254.

5. Schmier JK, Halpern MT, Higashi MK, Bakst A. The quality of life impact of acute exacerbations of chronic bronchitis (AECB): a literature review. Qual Life Res. 2005;14(2):329-347.

6. Sin DD, McAlister FA, Man SF, Anthonisen NR. Contemporary management of chronic obstructive pulmonary disease: scientific review. JAMA. 2003;290(17):2301-2312.

7. Bahadori K, FitzGerald JM. Risk factors of hospitalization and readmission of patients with COPD exacerbation: systematic review. Int J Chron Obstruct Pulmon Dis. 2007;2(3):241-251.

8. Almagro P, Barreiro B, Ochoa de Echaguen A, et al. Risk factors for hospital readmission in patients with chronic obstructive pulmonary disease. Respiration. 2006;73(3):311-317.

9. Terzano C, Conti V, Di Stefano F, et al. Comorbidity, hospitalization, and mortality in COPD: results from a longitudinal study. Lung. 2010; 188(4):321-329.

10. Sullivan SD, Strassels S, Smith DH. Characterization of the incidence and cost of COPD in the US [abstract]. Eur Respir J. 1996;9 (suppl 23):S421.

11. Anzueto A, Ferguson GT, Feldman G, et al. Effect of fluticasone propionate/salmeterol (250/50) on COPD exacerbations and impact on patient outcomes. COPD. 2009;6(5):320-329.

12. Suh DC, Lau H, La HO, Choi IS, Geba GP. Association between incidence of acute exacerbation and medication therapy in patients with COPD. Curr Med Res Opin. 2010;26(2):297-306.

13. Nannini L, Cates CJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta-agonist in one inhaler versus placebo for chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2007;(4):CD003794.

14. Nannini LJ, Cates CJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta-agonist in one inhaler versus long-actingbeta-agonists for chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2007;(4):CD006829.

15. Nannini LJ, Cates CJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta-agonist in one inhaler versus inhaled steroids for chronic obstructive pulmonary disease. Cochrane Database Syst Rev. 2007;(4):CD006826.

16. Calverley PM, Anderson JA, Celli B, et al. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med. 2007;356(8):775-789.

17. Barnes PJ. Scientific rationale for inhaled combination therapy with long-acting b2-agonists and corticosteroids. Eur Respir J. 2002; 19(1):182-191.

18. Delea TE, Hagiwara M, Dalal AA, Stanford RH, Blanchette CM. 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):1-13.

19. Simon-Wastila L, Yang HW, Blanchette CM, Zhao L, Qian J, Dalal AA. Hospital and emergency department utilization associated with treatment for chronic obstructive pulmonary disease in a managed-care Medicare population. Curr Med Res Opin. 2009;25(11): 2729-2735.

20. Soriano JB, Kiri VA, Pride NB, Vestbo J. Inhaled corticosteroids with/without long-acting beta-agonists reduce the risk of rehospitalization and death in COPD patients. Am J Respir Med. 2003;2(1):67-74.

21. Rascati KL, Stanford RH, Borker R. A comparison of the risk of hospitalizations due to chronic obstructive pulmonary disease in Medicaid patients with various medication regimens, including ipratropium, inhaled corticosteroids, salmeterol, or their combination. Clin Ther. 2005; 27(3):346-354.

22. Rascati KL, Akazawa M, Johnsrud M, Stanford RH, Blanchette CM. Comparison of hospitalizations, emergency department visits, and costs in a historical cohort of Texas Medicaid patients with chronic obstructive pulmonary disease, by initial medication regimen. Clin Ther. 2007;29(6):1203-1213.

23. Akazawa M, Hayflinger DC, Stanford RH, Blanchette CM. Economic assessment of initial maintenance therapy for chronic obstructive pulmonary disease. Am J Manag Care. 2008;14(7):438-448.

24. Blanchette CM, Akazawa M, Dalal A, Simoni-Wastila L. Risk of hospitalizations/ emergency department visits and treatment costs associated with initial maintenance therapy using fluticasone propionate 500 mg/salmeterol 50 mg compared with ipratropium for chronic obstructive pulmonary disease in older adults. Am J Geriatr Pharmacother. 2008;6(3):138-146.

25. Briggs AH, Glick HA, Lozano-Ortega G, et al; Towards a Revolution in COPD Health (TORCH) Investigators. Is treatment with ICS and LABA cost-effective for COPD? multinational economic analysis of the TORCH study. Eur Respir J. 2010;35(3):532-539.

26. Dalal AA, Petersen H, Simoni-Wastila L, Blanchette CM. Healthcare costs associated with initial maintenance therapy with fluticasone-propionate 250 μg/salmeterol 50 μg combination versus anticholinergic bronchodilators in elderly US Medicare-eligible beneficiaries with COPD. J Med Econ. 2009;12(4):339-347.

27. International Society for Pharmacoeconomics and Outcomes Research. ISPOR international digest of databases. IMS LifeLink Health Plan Claims Database. Updated February 23, 2009. http://www.ispor.org/DigestOfIntDB/Default.aspx?rcd=426. Accessed February 12, 2010.

28. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-383.

29. Blough DK, Ramsey SD. Using generalized linear models to assess medical care costs. Health Serv Outcomes Res Method. 2000; 1(2):185-200.

30. Ferguson GT, Anzueto A, Fei R, Emmett A, Knobil K, Kalberg C. Effect of fluticasone propionate/salmeterol (250/50 μg) or salmeterol (50 μg) on COPD exacerbations. Respir Med. 2008;102(8):1099-1108.

31. Cooper CB, Dransfield M. Primary care of the patient with chronic obstructive pulmonary disease, part 4: understanding the clinical manifestations of a progressive disease. Am J Med. 2008;121(7 suppl):S33-S45.