From Asthma Severity to Asthma Control: Identification of New Guidelines

BACKGROUND: Managed care pharmacists utilize clinical guidelines to assist themin identifying acceptable treatment and management of chronic diseases. For thepast several months, an expert panel representing the National Asthma Education andPrevention Program (NAEPP) has discussed updated asthma guidelines, which couldchange future medication management of patients with asthma. OBJECTIVES: To identify new guidelines for asthma control. SUMMARY: Taking control of asthma is a foremost objective in managed care, asnew guidelines are debated, and asthma disease management programs identifymore effective models of asthma disease management. Within an effective asthmadisease management program, managed care practitioners, including but not limitedto physicians and pharmacists, must have a thorough understanding of the mostrecent clinical guidelines as well as current knowledge of the current consensusamong opinion leaders regarding asthma treatment and management. This ongoingeducation is critical in order to allow managed care practitioners to perform theirrole and function within their organizations. With the advent of changing asthmapractice guidelines and processes, managed care practitioners need current informationto recognize and apply guidelines-based asthma control. CONCLUSIONS: Appropriate asthma treatment requires adherence to currentmanagement guidelines. Approximately one third of asthmatics will not respondto standard asthma treatment and may even worsen while on therapy. Monitoring,close tracking, and regular evaluation of asthma patients at every health careprovider visit is a strategy that can be utilized to generate improved asthma control.


Table of Contents
From Asthma Severity to Asthma Control: Identification of New Guidelines A sthma is a chronic inflammatory airway disorder identified clinically by recurring episodes of wheezing, chest tightness, dyspnea, and coughing, particularly in the night or early morning. 1 It is characterized by airway hyperresponsiveness and inflammation. Airway obstruction can worsen when an asthmatic individual is exposed to various triggers or stimuli, which leads to airflow limitations via bronchoconstriction, mucus plugs, and increased inflammation.
Risk factors for asthma include host factors (e.g., genetic predisposition, atopy, airway hyper-responsiveness, gender, and/or race/ethnicity) as well as environmental factors that increase susceptibility for asthma development in predisposed individuals (e.g., respiratory infections, exposure to cigarette smoke, indoor/outdoor allergens, family size, occupational sensitizers, or air pollution). Asthma triggers vary from person to person and may include allergen exposure (e.g., housedust mites, animal dander, cockroaches, fungal spores, airborne pollens like grass, trees, and weeds), other environmental factors (e.g., cold air, smoke, and air pollution), strong emotional expressions (e.g., anxiety, stress, and laughter), exercise (particularly in cold air or in dry climates), medications or preservatives (e.g., aspirin, beta-blockers, nonsteroidal anti-inflammatory agents, sulfites, and benzalkonium chloride), respiratory infections, chemical irritants, occupational stimuli (e.g., flour dust for bakers; hay mold for farmers; and formaldehyde and other chemicals for plastics, rubber, and wood workers), or comorbid conditions (e.g., gastroesophageal reflux disease, rhinitis, and sinusitis). [1][2][3] A clinical history of asthma symptoms increases the likelihood that asthma is present, but other important tools should be used to make an accurate diagnosis. A physical examination may show wheezing when patients are symptomatic, and in many patients, findings consistent with allergic rhinitis and/or ocular allergies may be present. 2 A thorough patient assessment is essential, and it is important to consider the patient' s entire clinical picture when making the diagnosis. [1][2][3] The following are critical in establishing a clear asthma diagnosis: • Of primary significance is a history that suggests a pattern of respiratory symptoms such as recurrent episodes of chest tightness, difficulty breathing, shortness of breath, and coughing. A history may also suggest a pattern of symptoms that occur with exercise and exertion and that are improved with a bronchodilator (i.e., they are reversible). A clinical history should also evaluate whether these symptoms are related to environmental triggers, such as allergens, viral infections, and irritants. • Objective measurements of airflow limitations, reversibility, and variability are helpful in establishing a diagnosis. Spirometry is useful in diagnosing asthma and monitoring response to therapy. An increase in forced expiratory volume in 1 second (FEV 1 ) of 12% or greater or an increase in peak expiratory flow (PEF) of 15% or greater after an inhaled short-acting bronchodilator or after a trial of an inhaled corticosteroid (ICS) is diagnostic for asthma. Periodic assessments and ongoing monitoring are recommended to determine if asthma management goals are being met. 2,3 ss Prevalence Asthma is a common chronic disease that causes substantial economic, social, and public health burdens on society. According to the Centers for Disease Control and Prevention, asthma prevalence has increased over the past 2 decades. The overall annual age-adjusted prevalence rate of self-reported asthma increased by 209% from 1980 to 1996. Mortality from asthma also increased during this time period. 4 In the 2002 National Health Interview Survey (NHIS), 30.8 million people in the United States (11.1%) reported having been diagnosed with asthma by a health professional during their lifetimes (21.9 million adults and 8.9 million children). 5 The NHIS survey estimates of current asthma prevalence included people who were diagnosed with asthma by a health professional and who still had asthma. In 2002, 20 million people (7.2% of respondents) had asthma, including 6.1 million children and 14 million adults. 5 Approximately 60% of people with a current asthma diagnosis reported having at least 1 asthma attack in the previous year. This population may represent patients with suboptimally controlled asthma at risk for a poor outcome, such as hospitalization.
Uncontrolled asthma can interfere with daily activities, including work and school attendance. Among individuals who reported at least 1 asthma attack in the previous year (2001), children aged 5 to 17 years missed 14.7 million school days, and adults aged at least 18 years who were employed at the time of the survey missed 11.8 million work days. Health care use among this population included 13.9 million outpatient asthma visits to private physician offices and hospital outpatient departments, 1.9 million visits to emergency departments (EDs), and 484,000 hospitalizations. Among children aged 0 to 17 years, there were 196,000 hospitalizations (27 per 10,000 people), with the highest rate among children aged 0 to 4 years (59 per 10,000). In that same year, 4,261 people died from asthma. It is clear that despite the presence of well-studied, generally effective treatments for asthma, the condition remains a key public health problem in the United States. 5 ss Managing Care Evidence-based clinical practice guidelines for asthma have been available for 15 years. The National Heart, Lung, and Blood Institute/National Asthma Education and Prevention Program (NHLBI/NAEPP) guidelines were initially published in 1991 and updated in 1997, with a second update on selected topics in 2002. A complete update is expected in late 2006 or early 2007. The Global Initiative for Asthma (GINA) guidelines were originally published in 1995 and have been updated several times, most recently in October 2005. The GINA guidelines are a result of collaboration between the NHLBI and the World Health Organization. 2,3 However, despite the availability of these asthma clinical guidelines, optimal control of asthma symptoms remains an elusive goal for many patients. There are numerous reasons for this, including poor guideline compliance.

ss Costs
The costs associated with asthma are also a concern. Studies have shown that 20% of all patients with asthma account for 80% of the direct costs of treating the disease. 9 Direct medical costs include hospital outpatient services, hospitalizations, ambulatory care visits, ED visits, physician and facility payments, and prescription medications. Indirect costs include those resulting from missed work or school days and days with restricted work activity. The estimated annual per-patient costs are $2,584 for high-cost patients and $140 for low-cost patients. 9 In the United States alone in 2002, direct and indirect expenditures for all forms of asthma totaled $14 billion, which included $9.4 billion in direct costs and $4.6 billion in indirect costs. 10 However, despite these statistics and the wealth of information on the management of asthma, numerous studies have shown that a large portion of patients fall short of acceptable asthma treatment goals. Strategies aimed at optimizing asthma control for this subset of patients who consume the majority of resources should result in improved outcomes and decreased cost of care.
ss Guideline Use in the Management of Asthma (Current and Future) National asthma guidelines recommend that asthma be classified and treated in accordance with a patient' s baseline severity rating (while not using controller therapy). Severity in an individual can change over time, as can the frequency and severity of an individual' s asthma symptoms, whether the person is on or off therapy. 1 A stepwise approach to asthma treatment exists with a goal of achieving and maintaining asthma control. 2,3 The overall management of asthma involves 4 components: (1) monitoring response to therapy, including disease activity and disease control (symptoms and lung function); (2) controlling/avoiding asthma triggers; (3) patient and caretaker education, and (4) pharmacologic therapy. Pharmacotherapy includes the use of short-acting reliever medications for acute symptoms and the use of controller medications for long-term control of symptoms. There are also some options for difficult-to-treat, moderate-to-severe asthmatics, such as those with severe persistent asthma whose disease is poorly controlled with routine use of controller medications. These are usually patients with severe persistent asthma despite conventional pharmacotherapy. Some of the options for these patients include daily oral corticosteroids (e.g., prednisone) and oral methylxanthines (e.g., theophylline). Both of these drug classes require careful monitoring to prevent toxicities. Additional approaches to managing these patients include use of the oral leukotriene modifier zileuton; subcutaneous allergen-specific immunotherapy; and/or use of advanced therapeutics/new technologies such as subcutaneous injection of the anti-immunoglobulinE (IgE) monoclonal antibody omalizumab. 2 Management strategies should be individualized to meet each patient' s needs.
Ongoing monitoring and periodic assessments are necessary to determine whether the goals of asthma therapy are being met. These goals include the following: 1. Preventing chronic and troublesome symptoms 2. Maintaining near-"normal" pulmonary function 3. Preventing recurrent asthma exacerbations and minimizing the need for emergency department visits or hospitalizations 4. Maintaining normal levels of activity, including exercise and other physical activity 5. Providing optimal pharmacotherapy with the least amount of adverse effects 6. Meeting patients' and families' expectations of and satisfaction with asthma care The NAEPP guidelines classify asthma severity into 1 of 4 symptom categories, which are depicted in Table 1. 3 It is recommended that severity assessment and classification be performed during the initial evaluation and before the initiation of treatment. Spirometric tests are recommended at the time of the initial assessment, after treatment is initiated until patient symptoms and PEF have stabilized, and at least every 1 to 2 years thereafter. Periodic assessments and ongoing monitoring are recommended to ascertain that therapy goals are being met. Patients should be given a written action plan based on signs and symptoms and/or PEF, and patients should be trained to recognize signs of decreasing control and appropriate actions to take. The stepwise approach for the treatment of asthma in adults and children older than 5 years is also described in the NAEPP guidelines and can be seen in Figure 1. 3 Going forward to the "next generation" of asthma clinical practice guidelines, the cornerstone will likely be the monitoring and tracking of patient outcomes with a focus on asthma control. Key tenets may include routine monitoring and tracking of asthma control at each visit to the health care provider, whether that is well care or urgent care, or whether that is with a specialist or a general practitioner. During the past several months, an expert panel representing NAEPP has considered the results of recent  asthma studies to determine if changes in asthma treatment recommendations are warranted, with a goal to release updated guidelines in late 2006 or early 2007. 11 In an asthma disease management program in Los Angeles, one of the most important elements is to optimize control with thorough assessment at regular visit intervals. 12 Asthma patients are seen by their health care teams on a regular basis, including when they are well, in order to reinforce the patients' understanding of their daily management plans and to encourage adherence to the daily use of controller medications. Commonly and unfortunately, the health care team sees asthma patients only when they are sick. It is therefore important during the patient/family education process to reaffirm the importance of regular visit intervals (e.g., every 1-6 months as appropriate, depending on level of control). Patient compliance with visits to the provider' s office to receive ongoing assessment and asthma education may be improved by the office making simple reminder calls the day before every visit. Health care providers should provide ongoing self-management education, including control of environmental factors that exacerbate symptoms (e.g., allergens and irritants), at each visit. 13 With change in asthma treatment and management occurring at the point of care, updated guidelines may assist providers in identifying and applying new strategies for asthma control.

ss Tracking Asthma Therapy
Since asthma is a variable disease, symptom control should be evaluated at each visit so the daily management plan can be modified accordingly. A systematic method of tracking control for each patient should be established and completed at each visit to document response to asthma therapy. Treatment plans can then be individualized, with the goal of optimal control.
An example of how this model of care has been successfully implemented using a population-based model is the mobile clinics in Los Angeles, California, that supplement traditional clinics by visiting schools to monitor and treat children. In each of these clinical settings, a method has been developed to identify and track poorly controlled asthma in the general population in Los Angeles. The Health Risk Assessment survey software provides a systematic way to identify poorly controlled asthma or well-controlled asthma. Risk assessment software systems may be accessed in a kiosk located in a clinic or a physician' s office waiting area. These systems offer an automated survey consisting of a list of questions that the parent or asthma patient can complete while waiting to see the provider. Survey results can be used to help identify patients with suboptimally controlled asthma. Survey answer patterns are evaluated using validated scoring algorithms to help predict 3 things: (1) whether a child has asthma (yes or no); (2) if yes, whether the asthma is well controlled or poorly controlled; and (3) if poorly controlled, whether it is moderate to severe at the current time. 14,15 A copy of the full report may be provided directly to the provider for review before the patient enters the examination room.
An individualized checklist provides a review of the education and treatment needed either to improve poorly controlled asthma or to maintain good control. The report may also be transmitted to a care manager, risk manager, or a care coordinator at the patient' s health plan or managed care organization. These are simple methods to track control and help the multidisciplinary team coordinate care using uniform assessment tools.

ss Monitoring Asthma Control
Identifying poorly controlled asthma patients, regardless of the setting, is the first step in an effective asthma disease management program. The long-term process of achieving and maintaining asthma control is complex and requires continual focus on the following 3 areas: (1) thorough evaluation of disease activity or severity, (2) evaluation of the response to therapy, and (3) routine follow-up. This process is depicted in Figure 2. 16  behavior, follow-up of patients at close intervals, and reinforcement are all time intensive. Through technology like the kiosk survey, simple assessment before the patient visit can be performed. Information on asthma control, recent ED visits, hospitalizations, missed school days, and the frequencies of symptoms (if any) can be collected in a time-efficient manner and can be tracked and trended over time. More time can then be spent developing management plans and educating patients. This approach at the point of care facilitates a more efficient and productive delivery of health care services.
If health care providers follow the NAEPP guidelines and classify asthma severity at baseline and initiate therapy according to each patient' s severity presentation, the guidelines suggest that patients will be able to meet the goals of therapy. .2,3 Goals of therapy are the following: 1. Minimal symptom frequency 2. Intermittent symptom frequencies of ≤2 days per week and ≤2 nights per month 3. No severe asthma flare-ups between patient visits (i.e., no ED visits, hospitalizations, or systemic steroids) 4. Lung function normal or optimal for patient 5. No asthma-related activity or exercise limitations It is anticipated that the next version of the NAEPP guidelines will go a step further to address the need to track asthma control in patients. The current guidelines suggest that, over time, asthma is a highly variable condition. Symptoms wax and wane. At times, a patient may be at a low level, with intermittent or mild disease, while at other times, the symptom presentation may be more moderate to severe. The current NAEPP and GINA guidelines suggest that if appropriate therapy is started for a particular patient' s symptom frequency at the time of the evaluation, adherence is good, and the health care team has provided the appropriate education for the patient to be well controlled. Studies have provided evidence on what happens when asthma patients are treated according to these clinical practice guidelines. This evidence also provides a rationale for the need to identify and track asthma control. 14,15 ss Clinical Evidence

Inhaled Corticosteroids: Adults (>18 Years)
In an NIH/NHLBI-sponsored pilot study, 17 Szefler et al. compared the relative beneficial and systemic effects of 2 different ICSs in a dose-response relationship in a 24-week, parallel, open-label, multicenter trial in patients with mild-to-moderate persistent asthma using serial dosing (low to moderate to high) of ICSs. Thirty adult subjects were randomized to receive either beclomethasone dipropionate (BDP) in a metered-dose inhaler (MDI) with a spacer (OptiChamber, Respironics), or fluticasone propionate (FP) MDI with a spacer for 24 weeks. The goal was to determine which ICS dose provided the most beneficial response without introducing risk. There were 15 patients randomized to each treatment group. Doses were selected to result in comparable overnight plasma cortisol concentration suppression. The serial dosing for BDP-MDI was 168, 672, and 1,344 mcg/day using the Vanceril chlorofluorocarbon (CFC) 84-mcg/actuation (Schering). The serial dosing for FP-MDI was 88, 352, and 704 mcg/day using the Flovent CFC 44-mcg/actuation (GlaxoSmithKline). Each study dose was administered for 6 weeks. To assess maximal response, FP dry-powder inhaler (DPI, Flovent Diskhaler) 2,000 mcg/day was then administered to all patients for 21 days. Inflammatory markers and airway function were measured at baseline (before treatment initiation), after each of the 3 ICS-MDI study doses was administered for 6 weeks, and, finally, after week 3 of the FP-DPI dosing. Tests included spirometry, methacholine challenge, asthma control assessment, exhaled nitrous oxide (eNO, which is a noninvasive marker for airway inflammation), exercise testing, and induced sputum values. Overnight plasma cortisol levels were measured at these same times, except they were not measured after week 3 of the FP-DPI dosing. The primary outcome variable for assessing comparative effectiveness was FEV1. The secondary outcomes were methacholine PC 20 , (a provocative concentration of methacholine that reduces the FEV 1 by 20% from baseline; this test is used to determine airway reactivity 2 ), exercise-induced fall in FEV 1 , eNO, and sputum eosinophils.
There were no significant differences between the 2 study groups at baseline. Overnight plasma cortisol was suppressed in a dose-dependent manner. The FEV 1 maximally increased (n = 13), and the near-maximal methacholine PC 20 change (n = 7) occurred at the medium dose (672 mcg) for BDP-MDI. The FEV 1 maximally increased (n = 12), and the near-maximal methacholine PC 20 change (n = 7) occurred at the low dose (88 mcg) for FP-MDI. In neither of the treatment groups did the FEV 1 or the maximum PC 20 increase following the 2,000-mcg/day dosing of FP-DPI. In the exercise challenge, BDP-MDI (n=9) maximal inhibition of the absolute fall in FEV1 and of the area-under-the-curve (AUC) from baseline also occurred following medium-strength dosing. The number of FP-MDI-treated patients completing the exercise challenge was too small to assess. However, the data identified maximum inhibition of the absolute fall in FEV1 (n=6) and of the AUC decline (n = 5) occurring at the high dose and only slightly  higher than with medium dosing of FP-MDI. Sputum eosinophils were maximally reduced following high dosing with BDP-MDI (n = 5) and following medium dosing with FP-MDI (n = 6). Comparison of this effect was difficult due to baseline differences in sputum eosinophils, which tended to be higher for BDP-MDI. Patient variability in FEV1 response can be seen in Table 2 and Figure 3.
A variable response to ICSs was observed in this study but could not be attributed to poor medication adherence, as this was closely monitored. Good patient response was also associated with a high eNO level (median, 17.6 vs. 11.1 part per billion), higher median maximal bronchodilator reversibility (25.2% vs. 8.8%), and a lower median FEV1/FVC (forced vital capacity) ratio (0.63 vs. 0.73) before treatment, than with poor-response patients. Excellent improvement in PC 20 versus poor improvement was found to be associated with high sputum eosinophil levels (3.4% vs. 0.1%) and older age at onset of asthma (20-29 years vs. <10 years).
This study showed significant variability in patient response to ICSs. Most patients with moderate persistent asthma gleaned response from low-to-medium doses of ICSs. High-dose therapy with ICSs did not improve efficacy, but it did result in an increase in systemic effects, particularly suppression of overnight cortisol production. About one third of patients had a poor response to treatment as measured by methacholine PC 20 and FEV1. The authors note that a larger patient population needs to be studied in order to identify a more identifiable dose-response relationship in efficacy measures of ICSs. Additionally, further studies are needed to characterize and identify so-called "poor responders" so that alternative treatment strategies can be identified.

Inhaled Corticosteroids, Leukotriene Modifiers: Adults (>15 Years)
In a similar but larger study, Malmstrom et al. compared the clinical benefit of either an ICS (beclomethasone) or an oral leukotriene receptor antagonist (LTRA, montelukast) in patients aged 15 to 85 years with chronic mild-to-moderate asthma and an FEV1 50% to 85% of predicted value. 18 This was a 12-week, multicenter, international, randomized, double-blind, double-dummy, placebocontrolled, parallel-group study. These patients were nonsmokers, had asthma for at least 1 year before entry in the study, had an average beta-agonist use of at least 1 puff daily, had a daytime asthma score of at least 64 (of a possible 336), and had an increase of at least 15% in absolute FEV1 after inhaled beta-agonist use on at least 2 of 3 visits during period 1 of the study. The study consisted of a 2-week, single-blind, placebo run-in period (period 1) followed by a 12-week, double-blind treatment period (period 2) and a 3-week, double-blind placebo washout period (period 3). Of 2,253 patients who were screened, 895 eligible patients were randomized to 1 of 3 treatment groups in a ratio of 3:2:2, to receive (1) montelukast 10 mg once daily in the evening; (2) inhaled BDP, 200 mcg twice daily; or (3) placebo. To assess withdrawal from therapy, during period 3, a subset of patients originally receiving active treatment (n = 40) were switched to placebo in a blinded fashion. The patients originally assigned to receive placebo continued to receive it. Placebo and montelukast 10 mg tablets were identical, as were BDP (100 mcg/puff) and placebo inhalers. The short-acting beta-agonist, albuterol, 100 mcg/puff, was allowed if needed. If asthma worsened and required additional treatment, oral corticosteroids were administered according to protocol. If 2 worsening asthma episodes occurred, patients were discontinued from the study. The primary endpoints were daytime asthma symptom score and FEV1. Secondary endpoints were PEF rates in the morning and evening, as-needed beta-agonist use, nocturnal awakenings, asthma-specific quality of life (QOL), and worsening asthma episodes.
Spirometry training and assessment were done according to the American Thoracic Society guidelines. Spirometry was completed at baseline (before the start of the study) and at each visit after inhaled beta-agonist therapy, theophylline therapy, or antihistamine therapy had been withheld for at least 6, 24, or 48 hours, respectively. Answers to specific questions on a diary card regarding daily asthma symptoms, nocturnal awakenings, and morning and evening PEF rates were specifically queried, as was daily use of as-needed salmeterol. Three assessments of PEF rates were requested in the morning and evening before study medication dosing, and the best of the 3 measurements was recorded. Other endpoints were evaluated and included (1) peripheral blood eosinophil counts, (2) global patient and physician evaluation scores (7-point scale: 6 means "very much worse" and 0 means "very much better"), and (3) evaluation of asthma QOL. Asthma outcome endpoints that were evaluated included (1) asthma attacks (worsening defined as requiring oral corticosteroids, a physician' s office visit, an ED visit, or a hospitalization); www.amcp.org Vol. 12

From Asthma Severity to Asthma Control: Identification of New Guidelines
(2) percentage of days with exacerbations of asthma; and (3) percentage of asthma-control days (ACDs). Patient compliance was evaluated. For each efficacy endpoint, only patients who had baseline measurements and at least 1 measurement after randomization were included.
The percentage of patients who discontinued the study during periods 2 and 3 was significantly greater in the placebo group (18.3%) than in the BDP group (9.6%) or the montelukast group (10.6%). Reasons for discontinuation included protocol deviations, withdrawal of consent, pregnancy, and loss to follow-up. Additionally, 11 placebo-treated patients (4%), 8 montelukasttreated patients (2%), and 5 BDP-treated patients (2%) discontinued treatment due to clinical adverse effects; a worsening of their asthma was the most common of these effects (8 placebo-treated patients, 4 montelukast-treated patients, and 1 BDP-treated patient). Other adverse effects associated with treatment discontinuation included menstrual disorder, eyelid edema, and pharyngeal discomfort (placebo group); epididymitis, urticaria, vasovagal reaction, and trauma (montelukast group); and rash, dyspepsia, neoplasm, and asthenia or fatigue (BDP group). Mean compliance (± SD) with inhaled study medication in period 2 was 89.6% ± 36.3% in the placebo group, 87.6% ± 30.9% in the montelukast group, and 88.6% ± 34.8% in the BDP group. Mean compliance with oral medication in this same period was 99.6% ± 2.6% in the placebo group, 99.8% ± 0.9% in the montelukast group, and 99.3% ± 3.4% in the BDP group.
Both montelukast and BDP treatment, compared with placebo, significantly improved FEV1, daytime symptom scores, and the secondary endpoints (P <0.001). The average percentage change from baseline in FEV1 was 13.1%, 7.4%, and 0.7% for BDP, montelukast, and placebo, respectively (P <0.001 for each active treatment vs. placebo; P <0.010 for BDP vs. montelukast). The daily symptom score average change from baseline was -0.62 for BDP, -0.41 for montelukast, and -0.17 for placebo (P <0.001 for each active treatment vs. placebo; P <0.010 for BDP vs. montelukast). The same statistical significance was noted when comparing PEF rates, QOL, reduced nocturnal awakenings, reduced asthma attacks, increased number of ACDs, and decreased number of asthma exacerbation days. Overall, BDP had a larger mean clinical benefit than montelukast. The differences between montelukast and BDP were 5.8% for FEV1, -0.21 for daytime symptom scores, -0.67 for puffs/day of beta-agonist use, 15.3 L/min for morning PEF, 11.2 L/min for evening PEF, and -0.70 for nocturnal awakenings. However, montelukast had a faster onset of action and a greater initial response than BDP. Seven to 10 days after therapy initiation, the BDP effect bypassed that of montelukast. Physician global assessments were similar to the patient global assessments. Patients felt better after 12 weeks of treatment: 94.2% for BDP, 89.1% for montelukast, and 73.4% for placebo (P <0.001 for each active treatment vs. placebo). Similar decreases in peripheral blood eosinophil counts (P <0.050 vs. placebo) were also noted. Patients treated with placebo, montelukast, and BDP had asthma exacerbations on 26.1%, 15.2%, and 9.7% of days, respectively. The percentage of days with asthma exacerbations was decreased by 63% with BDP vs. placebo (P <0.050), and 42% with montelukast vs. placebo (P <0.050). The percentage of patients with at least 1 asthma attack during treatment was 27.3% for placebo, 15.6% for montelukast, and 10.1% for BDP. Discontinuation of both medications caused slow but definite decreases in effect. After drug discontinuation, patients showed no rebound worsening of asthma. Worsening asthma, headache, and upper respiratory tract infection (URI) were the most commonly reported adverse effects during treatment.
Medication doses used in this study mimic initial clinical dosing for asthmatic patients. In this study of mild-to-moderate asthmatics (treated with montelukast or inhaled BDP, some patients had little or no improvement (~30%), some patients got worse (~30%), and others had larger improvements (~40%). The distribution of treatment responses for FEV1 can be seen in Figure 4. This study showed a similar distribution of response to that conducted by Szefler et al. These studies showed that by using currently available asthma therapies in clinically approved doses, some patients would have an adequate response while others would not, and it is difficult to predict nonresponders. Malmstrom et al. recommended that future asthma studies generate response distributions in different patient groups (e.g., different demographic groups, severity groups) to better understand how these populations respond to differing asthma therapies. Additionally, long-term effects of inhaled BDP and montelukast were unable to be identified due to the short course of treatment in this study. They did, however, show that both inhaled BDP and montelukast each protected patients against worsening of asthma episodes.
In addition, Israel

From Asthma Severity to Asthma Control: Identification of New Guidelines
effects determined by ACDs. 19 This was a randomized, multicenter, double-blind, placebo-controlled, parallel-group study of asthmatics (n = 782) defined as having (1) an FEV1 percent predicted value between 50% and 85% at rest, and at least a 15% increase in FEV1 after albuterol administration; (2) a weekly average beta-agonist use of >2 puffs/day; and (3) an age span between 15 and 85 years. Additionally, patients had to have had at least a 1-year history of clinical asthma symptoms and be nonsmokers. Patients were randomized in a 3:3:1 ratio to receive either montelukast 10 mg daily, inhaled BDP 200 mcg twice daily, or placebo treatment for 6 weeks in a double-dummy manner. Week 1 was a prestudy screening period, followed by a 2-week, single-blind, placebo baseline period, followed by a 6-week treatment period. Clinic visits were scheduled at the end of the prestudy week 1, at the end of the week 2 single-blind, placebo baseline period, and every 3 weeks during double-blind therapy. The primary study objective was to compare the effects of the 2 active treatments, as evaluated by ACDs. Secondary endpoints included (1) change from baseline in FEV1, (2) beta-agonist use, (3) asthma attacks, (4) asthma flares, (5) sustained asthma control, (6) rescue corticosteroid use, and (7) adverse effects. Daily diary cards with information regarding asthma attacks (yes or no), nighttime awakening, and albuterol use (number of puffs per day) were completed throughout the single-and double-blind treatment periods. All patients with a baseline evaluation and at least 1 postrandomization measurement were included in the analysis. There were 782 patients who entered the double-blind period: 332 received BDP-ICS, 339 received montelukast, and 111 received placebo. There were 752 patients who completed the 6-week, double-blind study.
The percentage of ACDs for the active treatment groups was almost identical (97.7% overlap in the distribution), with montelukast being at least equal to inhaled BDP. Both were greater than placebo when frequency of asthma attacks, asthma flare-ups, and rescue oral corticosteroid use were compared. Inhaled BDP had a greater effect than montelukast at improving FEV1, and both active treatments were better than placebo at improving FEV1. The difference between the 2 active treatments (montelukast vs. inhaled BDP) from placebo in the percentage of days of asthma control was significant (14.6% [P <0.001] and 14.4% [P <0.001], respectively). The effect of both active treatments on ACDs gradually increased with time. The effects of montelukast and inhaled BDP on airway function and asthma control can be seen in Figure 5. Both active treatments, compared with placebo, reduced average beta-agonist use (P <0.001 for each comparison), and both did not differ from each other in this effect. The percentage of patients receiving active treatments with at least 1 episode of sustained asthma control was not significantly different (P = 0.324), and both active treatments were significantly superior to placebo for this endpoint. The mean percentages of sustained asthma control were 33.4% (montelukast), 32.1% (inhaled BDP), and 19.3% (placebo). The percentage of overlap between the active treatments for ACDs was 100% (95% confidence interval, 91.2%-100%). The active treatments were generally well tolerated. The most frequently reported adverse effects in >5% of patients were URIs, headache, sinusitis, and asthma, and did not differ between the active treatment groups. Laboratory test abnormalities were similar among the treatment groups.
In this study, although the extent of clinical control achieved was not the primary purpose of the study, both montelukast and inhaled BDP treatment did not achieve a high level of asthma control. 19 Neither treatment group differed in measures of clinical effectiveness, including reduction in beta-agonist use or the need for oral corticosteroids, and they did not differ in the extent of sustained asthma relief. It was therefore surprising that these authors identified that both montelukast and inhaled BDP differed in their effects on airway function, as can be seen in Figure 6

FIGURE 6
As noted, the change from baseline in FEV1 for montelukast and inhaled BDP showed the same distribution as the 2 studies previously reviewed. 17,18 Additionally, since fixed dosing for a finite period of time is not ideal treatment in the clinical setting, other therapeutic interventions are required to achieve full asthma control. Higher doses of ICSs can be used, but as noted by Szefler et al. in the previously reviewed study, a flat dose-response relationship can occur with inhaled BDP if used at doses higher than those used in this study. 17 The authors note that this study demonstrates similar response rates for montelukast and inhaled BDP when evaluating broad and clinically relevant measures of asthma control in chronic asthmatics, which supports the use of montelukast as initial therapy as an alternative to ICSs. In patients who are not controlled with either montelukast or an ICS, the combination of both agents may provide additional control when symptoms are not controlled with either agent alone. Asthmatics treated with ICSs, compared with leukotriene modifiers, might yield effects on FEV1 that are disproportionate to clinical effects achieved. Therefore, when evaluating treatment response, practitioners may need to place greater importance on clinical indices than on measurements of airway quality.
This distribution of asthma response has been observed in the pediatric population as well. Zeiger et al. studied the intraindividual and interindividual response profiles and predictors of response to ICSs and an LTRA in school-age children. 20 This was a multicenter, double-masked, 2-sequence, 16-week crossover trial in children aged 6 to 17 years with mild-to-moderate persistent asthma using only "as needed" bronchodilators. The patients (n = 144) were randomized to receive an ICS, fluticasone propionate (100 mcg/inhalation, twice daily) or montelukast (5 to 10 mg nightly, age dependent). A 5-to 10-day run-in period after enrollment characterized each patient' s asthmatic features while off controller therapy. During each 8-week treatment period each patient received an active drug and a matching placebo for the alter-native drug. The first 4 weeks of the second treatment period were felt to be an adequate washout period for the first drug, and the second 4 weeks of each treatment period were therefore used to compare outcomes between the 2 active treatments. Clinical, pulmonary, and inflammatory responses were evaluated.
Outcomes were evaluated from daily patient diaries. The principal secondary outcome, ACD, was defined as a day with no daytime or nighttime asthma symptoms, no rescue albuterol use for asthma symptoms, a PEF <80% of personal best, no asthma health care use, and no asthma-related absences from work or school. The PEF was evaluated upon awakening and before bedtime. The following outcomes were determined at randomization and each 4-week treatment evaluation as part of the Asthma Control Questionnaire (ACQ): symptoms, albuterol use, and FEV 1 . The ACQ is scored on a 7-point scale (0 = good control, 6 = poor control). Additionally, eNO, FEV 1/FVC, and FEV1 were determined, as was resistance of the respiratory system at 5 Hz (R5) and area of reactance (AX). These last 2 tests are potential indicators of small-airway function. To help characterize each patient' s phenotype, some of the following procedures (but not all) were performed prior to randomization: asthma history, allergen skin tests, total blood eosinophil count, serum IgE measurement, spirometry, methacholine P20, and eNO measurement. Only patients who completed both treatment periods were included in the analyses.
Of the 144 patients initially randomized, only 127 (88.2%) completed both treatment periods. Improvements in most clinical outcomes (ACDs, ACQ, albuterol use), pulmonary responses (FEV1/FVC, PEF variability, morning PEF, and measures of impedance), and inflammatory biomarkers (eNO) significantly improved more with ICSs than montelukast treatment. The eNO test was a predictor of ACDs (P = 0.011) and a response indicator in selecting the difference in ACD response between ICS and montelukast (P = 0.003). The ACDs both significantly increased for ICS (2.8 days/week) and montelukast (2.1 days/week), P <0.001. The change in ACDs per week from baseline can be seen in Figure 7. In the ICS-treated patients, 59% improved ≥2 days/week compared with 47% for montelukast-treated patients, and 13% of ICS patients improved 1 to <2 days/week compared with 12% of montelukast-treated patients. When individual responses were compared, 29.3% of ICS-treated patients had ≥1 ACD per week during treatment compared with 12.2% of montelukast-treated patients. Compared with baseline scores, both treatments showed significant improvements in ACQ scores, but better control was achieved with ICSs (P = 0.001). Albuterol use decreased significantly with both treatments, but a greater decrease was observed with ICS treatment. The ICS therapy had significant improvements in prebronchodilator FEV 1 /FVC, PEF variability, morning PEF, R5, and AX, in addition to a 6.8% FEV 1 mean improvement. Montelukast treatment led to significant improvements from baseline in morning PEF, and a 1.9% increase in FEV 1 . Comparing prebronchodilator effects of

FIGURE 7
ICS and montelukast, the former had greater effects in FEV 1/ FVC (P <0.001), PEF variability (P=0.003), and AX (P=0.001). A greater decrease in eNO was noted following ICS treatment (P = 0.003), but both drugs decreased eNO. At the baseline measurement, higher eNO levels (P = 0.036), more positive aeroallergen skin test responses (P = 0.008), greater albuterol use (P = 0.029), and fewer ACDs predicted more ACDs following ICSs. During ICS treatment, eNO decreases were associated with positive clinical responses.
This study resulted in more positive clinical, pulmonary, and inflammatory responses in mild-to-moderate persistent asthma in children treated with ICSs compared with montelukast, supporting the first-line use of ICSs. As a predictor of patient response, eNO might help to identify individual children not receiving asthma controllers who would achieve a greater improvement in ACDs with an ICS compared with an LTRA. Like the previously evaluated studies, there was a similar proportion of patients with no change in ACDs from baseline following treatment with either an ICS or an LTRA.

Inhaled Corticosteroids, Nedocromil: Children (5-12 Years)
Finally, the Childhood Asthma Management Program (CAMP) Research Group of NHLBI conducted a study that evaluated 2 active treatments, or placebo, on the long-term treatment of mild-to-moderate asthma. 21 Children (n = 1,041) aged 5 through 12 years were randomized to receive 200 mcg budesonide (n = 311), 8 mg nedocromil (n = 312), or placebo (n = 418) twice daily in a placebo-controlled manner. The objective of the study was to evaluate whether the long-term, continuous treatment (for 4-6 years) with either an ICS (budesonide) or an inhaled noncorticosteroid drug (nedocromil) produced an improvement in lung growth and whether it was safe as compared with only symptomatic treatment (e.g., albuterol and prednisone, if needed). All patients used albuterol for symptoms. The primary outcome was lung growth, assessed by the change in FEV1 after bronchodilator administration, and the secondary outcomes were the degree of airway responsiveness, physical growth, psychological development, and morbidity.
Spirometry was performed twice yearly and a methacholine challenge was performed yearly. The children (or their parents or guardians) completed a daily diary card that recorded nighttime awakenings due to asthma, morning and evening peak flow measurements, use of study medication, use of albuterol for symptoms and for prevention of exercise-induced bronchospasm, prednisone use, school absences due to asthma, visits to a physician' s office or hospital due to asthma, and severity of symptoms. Patient' s height and weight were recorded at each follow-up visit; total bone mineral density of the spine from L1 to L4 and the Tanner stage of sexual development were assessed annually. Bone age (skeletal maturation) was assessed during the last 8 months of follow-up and used to estimate projected final height. Additionally, 4 neurocognitive tests were administered, 1 at baseline and then again 3 years later; 8 psychosocial questionnaires (including the Children' s Depression Inventory) were completed at baseline and at each follow-up visit. Skin testing for a core of allergens was performed at baseline and 4 years later. Anterior and posterior eye lens images were obtained during the last 8 months of follow-up, to be observed for posterior subcapsular cataracts.
The 3 study groups were similar at baseline except there were more boys in the nedocromil group. The follow-up duration was also similar in all groups, with a mean of 4.3 years. There was no significant difference in the primary outcome or the change in FEV 1 after bronchodilator administration between the treatment groups and placebo. Compared with the placebo-treated patients, the budesonide-treated patients had a significantly smaller decline in the FEV1/FVC before bronchodilator administration (0.2% vs. 1.8%). Budesonide-treated patients also had lower airway responsiveness to methacholine (P <0.001 vs. placebo), fewer hospitalizations (43% lower rate compared with placebo, and 2.5 vs. 4.4/100 person-years), fewer urgent care visits (45% lower rate, 12 vs. 22/100 person-years; P <0.001), a greater reduction in albuterol use for asthma symptoms, fewer prednisone courses (43% lower rate of use, P <0.001), and a smaller percentage of days on which additional asthma medications were needed. The FEV 1 PC 20 response to methacholine in children treated with budesonide and nedocromil can be seen in Figure 8. Budesonide treatment improved the FEV 1 after bronchodilator administration by 103.2% of the predicted value to a mean of 106.8% within 2 months, but this value diminished to 103.8% by the end of the treatment period. Nedocromil-treated patients versus placebo-treated patients had significantly reduced urgent care visits (27% fewer visits, 16 vs. 22/100 person-years; P = 0.020, and 16% fewer prednisone courses (P = 0.010. The mean height increase that was mostly evident during the first year of treatment was 1.1 cm less in the budesonide group compared with the placebo group (22.7 vs. 23.8, P = 0.005). The height  difference was similar between the placebo group and the nedocromil group. One nedocromil-treated patient died; she had been receiving supplemental treatment, including ICSs, for several months before her death. One placebo-treated patient required intubation for an asthma exacerbation. Better asthma control was achieved in the budesonide treatment group as indicated by significantly fewer asthma symptoms (P = 0.005), less albuterol use (P <0.001), and more episode-free days (P = 0.010. Changes in morning peak flow and the number of nighttime awakenings per month were similar in all groups. Compliance with treatment was similar between the budesonide group and the placebo group (73.7% vs. 76.2%, respectively), but was lower in the nedocromil group (70.2%). When comparisons were made over the entire follow-up time, the budesonide group significantly differed from the placebo group in all respiratory (spirometric) measures (P <0.001), whereas the nedocromil group did not significantly differ from the placebo group in any measure. Changes in growth velocity and bone density in the active treatment groups were similar to placebo at the end of the study. The only psychosocial measure that was different from placebo was a greater improvement in the total score of the Children' s Depression Inventory in the budesonide group, indicating less depression compared with placebo (a decline of 3.2 vs. 2.2; P = 0.010. One child in the budesonide group was classified as having a questionable posterior subcapsular cataract.
This study found that neither treatment with nedocromil nor budesonide improved lung function, as measured by the percentage of the predicted value for FEV1 after bronchodilator administration, which was unexpected. The widely accepted, clinically useful, highly reproducible and predictive measure of lung function, FEV1, was used because it correlates well with disease progression, health care use, and asthma severity, while accurately describing the natural history of childhood asthma. As a result of normal lung growth, the FEV 1 /FVC before bronchodilator administration declined over time in all 3 groups. There was also a lack of longterm improvement in the active treatment groups compared with placebo. The findings of improved airway responsiveness and health outcomes in this study favored budesonide but were not statistically significant, similar to results of short-term studies, including the one previously reviewed. After discontinuation of the study medication, no differences were observed among the 3 groups in lung function or growth from baseline to the final measurement, except for FEV 1 /FVC before bronchodilator administration. Continuous daily treatment with inhaled budesonide led to better asthma control than symptomatic treatment or nedocromil treatment in mild-to-moderate asthmatic children, with side effects limited to small but transient growth velocity reductions. This study revealed the importance of long-term, controlled trials in the treatment of asthmatics.
The CAMP study followed patient response over an average of 4 to 5 years of therapy in ideal conditions with follow-up. 21 On average, the ICS-treated patient' s airway reactivity improved. Tantisira et al. mapped out the 4-year change in airway responsiveness of ICS-treated patients in the CAMP study, as measured by PC change, and found a normal distribution. 22 Using the same CAMP study data, which looked "good," Tantisira et al. demonstrated that some patients have a good outcome, some a poor outcome, and some a worse response outcome, the same bell curve as previously described. This deception of averages can be seen in Figure 9. 22 This response suggests that other factors aside from treatment type contributed to this response. Therefore, it becomes increasingly more important to track asthma control and response to therapy, because there are patients who either in the short term or the long term and, even under ideal conditions, do not adequately respond.

FIGURE 9
A New Way to Predict the Response to Inhaled Corticosteroids in Asthma II 20,21

FIGURE 10
Geometric mean value of PC20 at various time points over the 4-year follow-up, stratified by treatment group assignment and H33Q genotype. The mean PC20 of individuals on corticosteroids who also possessed a copy of the 33Q variant improved significantly by the end of the first year of the study and continued to improve substantially with time and corticosteroid usage. The mean PC20 value at 4 years is within the range normally ascribed to individuals without the diagnosis of asthma. PC = methacholine provocative concentration.

From Asthma Severity to Asthma Control: Identification of New Guidelines
Tantisira et al. also evaluated and revealed that there was a certain gene involved in immune regulation. 22 Of all the CAMP patients studied, 701 were successfully genotyped at H33Q and had data for the outcomes of interest. Only genotypic information from whites was analyzed, due to concerns over possible population stratification. Approximately 30% of these patients had been randomized to the ICS treatment group. Baseline gender, age, ethnic distributions, FEV1, and PC 20 were similar between steroid and nedocromil groups. Use of ICSs over time was associated with significant improvements in FEV1 and PC 20 . There were no significant differences between children randomized to placebo or nedocromil. This study basically demonstrated that patients on ICSs with coding for the H33Q variant allele gene will have a dramatic therapeutic response (PC 20 reactivity) over 4 to 5 years, as depicted in Figure 10. If a patient has any other form of this gene (e.g., H33H), the response will be limited to zero, suggesting a biological basis for asthma treatment response, which needs to be considered along with social issues, compliance issues, and behavioral issues.

Summary of Clinical Trial Experience
1. Asthma control can be achieved and maintained by many patients.

Response to all classes of controller medication is variable, and
approximately one third of patients will not respond to a particular class and may even worsen while on conventional pharmacotherapy. 3. Pharmacogenetics may help to predict asthma response patterns; however, until that time comes, it is not known who will respond to therapy and who will not, requiring close tracking of patient responses. 4. It is important to carefully track the response to therapy at each patient visit. 5. Currently, there is no clear, universal objective measure of asthma control because clinical control of asthma is a construct of both subjective and objective measures. A subjective measure is needed that could include measures for daytime symptoms, nighttime symptoms, albuterol use, ED visits, hospitalizations, and missed days from work or school. This thorough asthma assessment is needed at every visit. Lung function is included, but it is not enough.

ss Conclusions
Asthma should be appropriately diagnosed and treated according to current clinical practice guidelines. 2,3 Most asthmatics achieve and maintain asthma control on conventional pharmacotherapy; however, response to controller medications is variable. Approximately one third of asthmatics will not respond to standard asthma treatment and may even worsen while on the therapy. Other options to manage these patients (e.g., those with severe persistent asthma despite conventional pharmacotherapy) may include subcutaneous omalizumab, subcutaneous allergen-specific immunotherapy, oral theophylline, oral zileuton, and/or oral daily corticosteroids. In the future, pharmacogenetics may also help to predict asthma response patterns. Therefore, close tracking of patient responses is necessary to determine who will and who will not respond to standard therapies. Monitoring and reevaluating each patient' s response and outcomes should occur continually with follow-up every 1-6 months, or once control occurs, depending on disease severity. Adherence to treatment (and avoidance of triggers) can be enhanced with education at regular intervals and phone call reminders of office visits. A health risk assessment should be performed and may be obtained in a technological manner, either as part of a disease management program or a stand-alone program in a provider' s office. A clear universal objective measure of asthma control is needed. A universal subjective measure is also needed. New asthma guidelines are expected in less than a year and will likely focus on asthma control. Monitoring, tracking, and evaluating an individual patient' s asthma outcomes, at each and every patient visit, have been shown to improve asthma control; therefore, strategies to accomplish these goals must be identified and implemented to generate best practices at the point of care delivery.

DISCLOSURES
Funding for this research was provided by Genentech, Inc. and Novartis Pharmaceuticals Corporation. Author Craig Jones discloses that he receives grant/research support from Genentech and Merck, is a member of the speaker' s bureau for Genentech, Novartis, Merck, sanofi-aventis, and Critical Therapeutics, and is a consultant for sanofi-aventis, Genentech,Novartis, Merck, and Critical Therapeutics. He has received an honorarium from PRIME for participation in this supplement. Author Linda Endicott discloses that she received an honorarium from PRIME for participation in this supplement. Author Michele B. Kaufman, is senior medical writer at PRIME; she discloses no other potential bias or conflicts of interest relating to this article. Jones served as principal author of the study. Study concept and design were contributed by Endicott and Kaufman. Data collection and interpretation and writing of the manuscript and its revision were the work of all authors.