Treatment Patterns in the Months Prior to and After Asthma-Related Emergency Department Visit

Treatment Patterns in the Months Prior to and After Asthma-Related Emergency Department Visit* David A. Stempel, MD; Craig S. Roberts, PharmD, MPA; and Richard H. Stanford, PharmD, MS

Background: There are 2 million asthma-related emergency department (ED) events each year in the United States. The underrecognition and undertreatment of asthma is believed to be associated with this high level of morbidity. This study was designed to describe the treatment patterns in the year prior to the ED event and for 2 months after the event. Methods: This retrospective observational study utilized an integrated managed care database that contained administrative claims from > 20 managed care plans across the United States. All patients with at least one ED visit for asthma during 2001 were included. Patients were required to have data available 12 months prior to and 2 months following the ED visit of interest, and were excluded if they had made an asthma-related ED visit within 12 months of the identified event. Results: There were 12,636 patients identified with an asthma-related ED visit. In the year prior to the ED event, 25.1% of the patients received an inhaled corticosteroid (ICS), 29.9% received an oral corticosteroid (OCS), and 53.5% received a short-acting ␤-agonist (SABA). Overall, there were three albuterol units dispensed for every ICS unit dispensed in the 12-month period prior to the ED event. Ninety-four percent of patients had made an office visit in the prior year, but only 13.3% underwent spirometry testing. Prescriptions dispensed for ICSs and OCSs increased 2.6-fold and 7.5-fold, respectively, in the month after the ED event, and dispensing rates reverted approximately to baseline rates by the second month after the index ED event. Conclusion: This study demonstrates the dependence of this population on the use of rescue medications, including SABA and OCS, to treat their asthma. Furthermore, the ED event resulted in only an incremental short-term improvement in ICS-containing controller treatment. (CHEST 2004; 126:75– 80) Key words: asthma; emergency department; inhaled corticosteroids; spirometry Abbreviations: ED ⫽ emergency department; ICS ⫽ inhaled corticosteroid; OCS ⫽ oral corticosteroid; SABA ⫽ shortacting ␤-agonist

department (ED) visits and hospitalizaE mergency tions for asthma represent evidence of treatment failure. One of the primary goals of the National Asthma Education and Prevention Program is the reduction of exacerbations, which are defined as hos*From Infomed Northwest (Dr. Stempel), Seattle, WA; NDCHealth (Dr. Roberts), Yardley, PA; and GlaxoSmithKline (Dr. Stanford), Research Triangle Park, NC. This study was supported by GlaxoSmithKline, Research Triangle Park, NC. Dr. Stempel is a consultant for GlaxoSmithKline, Dr. Roberts was an employee of NDCHealth at the time of article submission, and Dr. Stanford is an employee of GlaxoSmithKline. Manuscript received September 15, 2003; revision accepted February 13, 2004. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: David A. Stempel, MD, 9121 Northeast Sixteenth St, Bellevue, WA 98004; e-mail: [email protected] www.chestjournal.org

pitalizations, ED visits, and use of oral corticosteroids (OCSs).1 Evidence-based guidelines2– 4 promote the use of inhaled corticosteroids (ICSs) as the primary and preferred component of controller therapy for the treatment of persistent asthma. ICS controller therapy has been consistently associated with a reduction in ED visits, hospitalizations, and mortality.5–10 Adherence to guideline recommendations reduces unnecessary variations in care and should be associated with a decrease in disease morbidity. Data from US national statistics11 demonstrate that over the past decade the numbers of asthma-related ED events has remained elevated despite the recommendations of the national guidelines. Wolfenden and colleagues12 have recently reported that physician underdiagnosis of asthma severity may lead to undertreatment. In addition, Adams and coworkers13 have demonstrated that the use of antiCHEST / 126 / 1 / JULY, 2004

75

inflammatory medications for the treatment of persistent asthma is inadequate, and may be related to lower income, less education, present unemployment, and active smoking. To date, there are limited data investigating the pattern of resource utilization prior to ED attendance. The National Asthma Education and Prevention Program guidelines1,2 have recommended routine office visits and the objective measurement of lung function to promote a reduction in disease morbidity. The identification of treatment patterns of patients prior to an ED visit may help to identify potential opportunities for patient care interventions that may target and improve treatment regimens. The present study was designed to describe the patterns of care observed in patients prior to the ED treatment for acute asthma. The primary area of interest was the exploration of adherence to guideline recommendations and, specifically, the use of ICSs, the preferred controller treatment, in the year prior to the ED visit. In addition, the study quantified the use of rescue medications, including OCSs and short-acting ␤-agonists (SABAs). Furthermore, the impact of the acute care intervention in the ED on altering the prescription of ICSs and other asthma medications in the 2 months after the ED event was investigated. The frequency of office visits and the use of spirometry before and after the event were secondary measures of interest.

Materials and Methods This study is a descriptive retrospective analysis focusing on the 12 months prior to and the 2 months following an initial asthma-related ED visit. During this time period, patient demographics, comorbid conditions, asthma-related medical claims (eg, medications, office visits, and spirometry procedures for patients ⱖ 6 years of age), and office visits associated with other diagnoses were collected by month. The number of claims for asthma medications dispensed was calculated as well as the number of patients receiving new asthma medications following the event. Prescription claims identified as asthma medications included SABAs, ICSs, long-acting ␤-agonists, leukotriene receptor antagonists, cromones, theophylline, and OCSs. In this analysis, patients receiving fluticasone propionate and salmeterol in a single device were included in the ICS controller cohort. Patients were identified from an integrated managed care database (PharMetrics; Watertown, MA), which is a collection of administrative claims from a number of managed care organizations distributed across the United States. These plans encompass a number of different models, including health maintenance organizations, preferred provider organizations, and point of service plans, and can be considered to be representative of the current landscape of commercial health insurance in the United States. The database is compliant with the Health Insurance Portability and Accountability Act of 1996, and has encrypted unique patient identifiers to integrate pharmacy and medical claims. Patients were included in the study if they had made an ED visit during 2001 with a primary diagnosis indicating treatment for asthma (International Classification of Diseases, ninth revision, code 493.xx). Asthma-related ED visits were defined as any medical claim having a procedure code 76

indicative of ED care (Current Procedural Code, 99281–99288) or a hospital claim that included a UB-92 revenue code for ED care (RC45x). The date of the first ED visit of 2001 was defined as the patient’s index date. Patients were required to have been enrolled in the managed care plan for 12 months prior to and 2 months following the initial ED event of interest. Patients who had made any asthma-related inpatient or ED visit 12 months prior to the index event were excluded. These selection criteria were established to provide a standardized population and to define the effect of an initial ED event. Patients with an inpatient visit within 24 h of the initial ED visit were grouped into an “admitted” cohort, and the remaining patients were classified into a treated-and-released cohort. Statistical comparisons of the cohorts were conducted using t tests for continuous variables and ␹2 tests for categoric variables.

Results There were 12,636 patients identified with an asthma-related ED visit. Hospitalizations were reported in 10.5% of these patients. Children aged 0 to 17 years comprised 43.5% of the total number of subjects. The mean age of patients in the released cohort was 25.8 years. Patients who were subsequently hospitalized were slightly older, with a mean age of 31.0 years. Women comprised 58.8% of the total cohort. During the observation period, comorbid acute upper respiratory infections were reported in 7.3% of the patients, and allergic rhinitis was identified in 7.1% of the patients (Table 1). The pharmacy records indicated that ICSs were dispensed to 25.1% of the patients who filled a mean amount of 2.97 units in the year prior to the ED visit and that OCSs were dispensed to 29.9% of the patients prior to the ED event. The mean number of prescription claims for ICSs and OCSs was comparatively constant for all 12 study months prior to the claim for ED service, with ICS claims ranging from 0.056 to 0.078 prescription claims per patient per month, and OCS claims ranging from 0.040 to 0.051 prescription claims per patient per month (Fig 1). In addition, leukotriene receptor antagonists and long-acting ␤agonists were dispensed to 14.6% and 10.8%, respectively, of the overall population in the year prior to the ED event, with mean supplies of 4.6 and 3.3 months, respectively, of medication dispensed. There were 1,471 new users (15.5% of prior nonusers) of an ICS-containing controller in the 2 months after the ED event. Despite the presence of initial new users of an ICS, the monthly rate of ICSs dispensed in the second month after the ED event declined nearly 50% to a monthly fill rate of 0.107 prescriptions per patient (p ⬍ 0.001 [first month after the index event vs second month after the index event]) [Fig 1]. Twenty percent of patients received an ICS-containing medication in the month following the ED visit, but ⬍ 11% of patients were given an ICS in the second month after the index ED event. As expected, there was a significant increase (7.5-fold) in the dispensing of OCSs Clinical Investigations

Table 1—Demographics* Variables

ED Only

ED Then Admitted

Total

Patients Age, yr Age group 0–17 yr 18–30 yr 31–40 yr 41–50 yr 51–60 yr 61⫹ yr Sex Female Male Secondary diagnosis Acute URI Allergic rhinitis Acute bronchitis Otitis media Acute pharyngitis

11,305 (89.5) 25.8 ⫾ 19.0

1,331 (10.5) 31.0 ⫾ 22.9

12,636 26.4 ⫾ 19.5

4,962 (43.9) 1,686 (14.9) 1,578 (14.0) 1,635 (14.5) 1,098 (9.7) 346 (3.1)

539 (40.5) 86 (6.5) 125 (9.4) 227 (17.1) 252 (18.9) 102 (7.6)

5,501 (43.5) 1,772 (14.0) 1,703 (13.5) 1,862 (14.7) 1,350 (10.7) 448 (3.5)

6,617 (58.5) 4,688 (41.5)

815 (61.2) 516 (38.8)

7,432 (58.8) 5,204 (41.2)

825 (7.3) 803 (7.1) 487 (4.3) 540 (4.8) 490 (4.3)

97 (7.3) 95 (7.1) 71 (5.3) 74 (5.6) 54 (4.1)

922 (7.3) 898 (7.1) 558 (4.4) 614 (4.9) 544 (4.3)

*Values given as No. (%) or mean ⫾ SD. URI ⫽ upper respiratory infection.

in the month of the ED visit (0.583 prescriptions per patient). In the second month after the ED occurrence, the observed dispensing rate was similar to the rate observed in the months prior to the ED event (0.078 prescriptions per patient). The ICS prescriptions dispensed within 2 months after the ED event were prescribed by physicians based at hospitals or EDs for only 26% of the patients. In contrast, these physicians prescribed 38% of SABA and 42% of OCS prescriptions that were dispensed following the ED event (Fig 2). Patients admitted to the hospital compared to those seen in the ED and then released had higher controller

use (39% vs 34%, respectively; p ⬍ 0.0001) and OCS use (39.0% vs 28.8%, respectively; p ⬍ 0.0001) in the preevent period. At least one rescue SABA was dispensed to 53.5% of patients in the 12 months prior to the event. These patients received on average 4.11 canisters per year. Pharmacy records for the overall population revealed that there were three albuterol canisters dispensed for every ICS unit dispensed. In addition, the dispensing of a rescue medication (ie, a SABA) increased 330% in the month of the ED event and was increased by only 30% in the second month after an acute care episode. Figure 3 demonstrates the changes in prescriptions for

Figure 1. Mean monthly claims for ICS and OCS prior and post ED event. www.chestjournal.org

CHEST / 126 / 1 / JULY, 2004

77

Figure 2. Location of the prescribing physician of dispensed medication after the ED visit.

controller and rescue medications in the 2 months prior and the 2 months after an ED visit. As shown in Table 2, office visits for any reason occurred in 94% of the patients in the 12 months prior to the ED visit and in 81% of the patients in the 2 months after the ED visit. Office visits associated with an International Classification of Diseases, ninth revision, code 493.xx occurred in 43% of patients before the ED event and in 54% after the event. In 31% of the patients, there were no asthmaspecific office visits either before or after the epi-

sode. Spirometry was performed in only 13.3% and 10.0%, respectively, of the patients (ⱖ 6 years of age) in the 12 months before and the 2 months after the ED visit. There were no significant differences in the admitted and released cohorts for this parameter.

Discussion This analysis demonstrates the greater dependence for this population on rescue medications

Figure 3. The mean monthly medication claims 1 and 2 months prior to and following the index ED event. 78

Clinical Investigations

Table 2—Use of Asthma Medications, Spirometry, and Outpatients Visits* Variables

Patients With at Least 1 Event 12 Mo Before ED Visit

New Users 2 Mo After ED Visit

Patients With at Least 1 Event Over 14 Mo

Outpatient visit ICS use† Any Controller OCS use SABA use

93.6 25.1 34.8 29.9 53.5

55.7 15.5 19.7 44.5 37.7

97.2 36.8 47.6 61.0 71.1

*Values given as %. New users ⫽ calculated as number of patients with new post-index events as the numerator over the number of patients without an event in the period before an ED event as the denominator. †Includes all ICSs (including Advair; GlaxoSmithKline; Research Triangle Park, NC).

including SABAs and OCSs. OCSs were dispensed to 30% of the patients in the year prior to an asthmarelated ED event and to 39% of patients in the 12 months prior to a hospitalization. The mean number of rescue medications was consistent over the 12 months prior to the event and did not appear to be dispensed as an intervention to prevent the index ED event. Although pharmacy claims do not indicate whether an OCS prescription was prescribed for asthma, the 7.5-fold increase in OCSs dispensed during the month of the event and the return to the baseline rate after the ED visit describes a pattern that is consistent with a treatment of asthma. The use of OCSs by 30% of the patients prior to the ED event suggests that many of these patients should have been considered candidates for ICS controller therapy. Over 50% of the patients received, on average, 4.11 canisters per year of a SABA during the 12 months prior to the ED event. This number of SABA canisters provides adequate medication for daily use, well above the guideline threshold for routine controller therapy. National and international guidelines1–3 advocate ICSs as the preferred component of controller therapy for patients with persistent asthma for all ages and all levels of disease severity. This recommendation is based on an evidence-based review of the literature that demonstrates consistency in the following clinical end points: reduction in symptoms; fewer exacerbations; less dependency on rescue medication; and greatest improvement in lung functions.4,14 –17 Furthermore, the use of ICSs has been identified as reducing exacerbations, including ED events and hospitalizations, in multiple observational studies.5–9 This association may be strongest when refill persistence approaches 4 to 6 canisters per year.8,9 The magnitude of reduction in these resource-intensive events has not been noted with alternative controllers. The present study demonstrates that 75% of patients treated in the ED for asthma did not receive ICS controllers in the year prior to this event. In addition, in those patients who were dispensed an ICS, the mean refill persistence was less than three canisters per year. www.chestjournal.org

One may have hypothesized that a severe acute exacerbation would have altered long-term treatment. Appropriately, patients received an increase in rescue medications of SABA and OCS in the month of the ED event. During the 2 months after the acute episode, approximately 20% of the patients started receiving ICS controller therapy. The initial increase in ICS use is still below guideline recommendations, but the decline by the second month after the index event demonstrates that the acute exacerbation had no sustained impact on dispensing of ICS controller medications. The dispensing of ICSs and OCSs was consistent in the 12 months preceding the ED visit and in the second month after the acute episode. In the 12 months preceding the ED visit, ⬎ 90% of patients made an office visit, and nearly 50% of these visits were associated with an asthma-specific diagnosis prior to the acute event. However, this physician contact did not result in the implementation of guideline-recommended care. The recognition of the signs and symptoms of asthma by patient, family, or health-care providers may have identified patients requiring ICS controller therapy. The infrequent use of spirometry both before and after the ED visit may be indicative of a lack of constant monitoring of the disease, although future studies are needed to confirm this hypothesis. Furthermore, the underreporting of asthma symptoms by patients may lead to an underrecognition of the need for disease control by the physician, and can result in undertreatment with controller medications. An inaccurate assessment by the health-care provider further decreases the likelihood of appropriate controller therapy.12 There were three patient– health-care provider events described in this analysis in which ICSs or other controller medications could have been prescribed. The first was in the health-care provider’s office prior to the ED event. Most patients had encounters with health-care providers before the ED visits, and the majority received one or more rescue medications prior to the ED event. Second was in the ED. Although typically not a place for maintenance medications to be prescribed, it is a site CHEST / 126 / 1 / JULY, 2004

79

where the patient and family may focus on the severity of the acute asthma and may be more receptive to preventive care. Finally, after the ED event controller therapy could have been instituted at the follow-up visit. The majority of these patients had these three contacts with health-care providers, but, unfortunately for many of these patients, these interactions did not result in an increased use of controller medications. The reason why controller medication was not dispensed as a result of these contacts cannot be discerned from these data. However, it does suggest a need for increased awareness by health-care professionals and patients with asthma of the benefit of starting patients on controller therapy when there is evidence of persistent signs or symptoms, or evidence of disease morbidity. This study has several limitations. The data records only the prescriptions dispensed. It does not report the prescriptions written and not filled, nor does it describe actual patient compliance with therapy. In addition, this initial ED visit may have been the sentinel asthma event for many of these patients, although the majority of patients were being given asthma medications in the 12 months before the acute flare-up of their respiratory symptoms. The 2-month follow-up time after the ED event may not be long enough to assess the impact of the ED event on medication use. However, the use of asthma medications at 2 months after the ED event were similar to monthly use before the event, suggesting a return to preindex event rates. Also, this analysis describes the treatment patterns of patients admitted to the ED for an asthma episode. The treatment patterns of asthma patients not admitted to the ED were not identified. The patients in this study were observed for the 12 months before the ED event with little variation from month to month, and this may possibly reflect the treatment patterns of those individuals not seen in the ED. In conclusion, this study demonstrates the inadequate use of controller medications and the overreliance on rescue medications in patients attending the ED for treatment of acute asthma. Evidence-based guidelines have given clinicians and patients recommendations for care. Attention now needs to be focused on interventions that enhance implementation. Strategies that identify patients who are at risk need to focus on the ability of patient, family, and health-care workers to recognize the signs and symptoms that warrant ICS controller therapy. In addition, spirometry is important, especially in the patient who underreports symptoms. Guideline implementation needs to focus on concise and elegant strategies to decrease unnecessary variations in care and to assure that the preferred treatments regimens are prescribed and dispensed to

80

patients who are at risk in order to decrease the number of patients treated for acute exacerbations of asthma in the ED.

References 1 National Asthma Education and Prevention Program. Expert panel report 2: guidelines for the diagnosis and management of asthma. Bethesda, MD: National Institutes of Health, April 1997; NIH Publication No. 97– 4051 2 National Asthma Education and Prevention Program. Expert panel report: guidelines for the diagnosis and management of asthma update on selected topics-2002. J Allergy Clin Immunol 2002; 110:S141–S219 3 Global Strategy for Asthma Management and Prevention. National Institutes of Health and National Hearty, Lung and Blood Institute, revised 2002. Available at: http://www.ginasthma.com. Accessed April 10, 2002 4 Ducharme FM, Hicks GC. Anti-leukotriene agents compared to inhaled corticosteroids in the management of recurrent and and or chronic asthma in adults and children. Cochrane Database Syst Rev 2002; issue 4 5 Donahue J, Weiss S, Livingston J, et al. Inhaled steroids and the risk of hospitalization for asthma. JAMA 1997; 277:887–891 6 Wennergren G, Kristja´ nsson S, Stannegård IL. Decrease in hospitalizations for treatment of childhood asthma with increased use of anti-inflammatory treatment, despite an increase in the prevalence of asthma. J Allergy Clin Immunol 1996; 97:742–748 7 Sin DD, Man SF. Low-dose inhaled corticosteroid therapy and risk of emergency department visits for asthma. Arch Intern Med 2002; 162:1591–1595 8 Schatz M, Cook EF, Nakahiro R, et al. Inhaled corticosteroids and allergy specialty care reduce emergency hospital use for asthma. J Allergy Clin Immunol 2003; 111:503–508 9 Suissa S, Ernst P, Kezouh A. Regular use of inhaled corticosteroids and the long term prevention on hospitalization for asthma. Thorax 2002; 57:880 – 884 10 Suissa S, Ernst P, Benayoun S, et al. Low-dose inhaled corticosteroids and the prevention of death from asthma. N Engl J Med 2000; 343:332–336 11 Mannino DM, Homa DM, Akinbami LJ, et al. Surveillance for asthma: United States, 1980 –1999. MMWR Surveill Summ 2002; 51:1–13 12 Wolfenden LL, Diette GB, Krishman JA, et al. Lower physician estimate of underlying asthma severity leads to undertreatment. Arch Intern Med 2003; 163:231–236 13 Adams RJ, Fuhlbrigge A, Guilbert T, et al. Inadequate use of asthma medication in the United States: results of the asthma in America national population survey. J Allergy Clin Immunol 2002; 110:58 – 64 14 Childhood Asthma Management Program Research Group. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med 2000; 343:1054 –1063 15 Pauwels RA, Pedersen S, Busse WW, et al. START Early intervention with budesonide in mild persistent asthma: a randomised, double-blind trial. Lancet 2003; 361:1071–1076 16 Pauwels RA, Lofdahl CG, Postma DS, et al. Formoterol and corticosteroids establishing therapy (FACET) International study group: effect of inhaled formoterol and budesonide on exacerbations of asthma. N Engl J Med 1997; 337:1405–1411 17 O’Bryne PM, Barnes PJ, Rodriquez-Roisen R, et al. Low dose inhaled budesonide and formoterol in mild persistent asthma. Am J Respir Crit Care Med 2001; 164:1392–1397

Clinical Investigations