- Email: [email protected]
Comprehensive Long-Term Management Program for Asthma: Effect on Outcomes in Adult African-Americans
Comprehensive Long-Term Management Program for Asthma: Effect on Outcomes in Adult African-Americans TIFFANY M. KELSO, PHARMD,* NABIL ABOU-SHALA, MD, t GRETA M. HEILKER, PHARMD,f KRISTOPHER L. ARHEART, EoD,§ TRACY S. PORTNER, PHD,* TIMOTHY H. SELF, PHARMD*
ABSTRACT: To determine if a comprehensive long-term management program, emphasizing inhaled corticosteroids and patient education, would improve outcomes in adult AfricanAmerican asthmatics a nonrandomized control trial with a 2-year intervention was performed in a university-based clinic. Inclusion criteria consisted of (~5) emergency department (ED) visits or hospitalizations (~2) during the previous 2 years. Intervention patients were volunteers; a comparable control group was ident\fied via chart review at hospitals within the same area and time period as the intervention patients. Individualized doses of beclomethasone with a spacer, inhaled albuterol "as needed," and crisis prednisone were the primary therapies. Environmental control, peak flow monitoring, and a partnership with the patient were emphasized. Detailed patient education was an integral part of management. Control patients received usual care from local physicians. ED visits and hospitalizations for 2 years before and 2 years during the intervention period were compared. Quality of life (QOL) measurements were made at baseline and every 6 months in the intervention group. Study group (n = 21) had a significant reduction in ED visits (2.3 ± 0.2 pre-intervention versus 0.6 ± 0.2 post-intervention; P = 0.0001). Control group (n = 18) did not have a significant change in ED visits during the 2-year post-inFrom the *Department of Clinical Pharmacy, University of Tennessee, Memphis, the t Department of Medicine, King Faisal Hospital & Research Center, Riyadh, Saudi Arabia, tMethodist Hospital, M emphis, Tennessee, and the §Cleveland Clinic, Cleveland, Ohio. Supported in part by an investigator-initiated grant from Glaxo W ellcome Inc. and by a grant (M01-RR002211) from General Clinic Research Center. Submitted November 5, 1995; accepted in revised form February 6, 1996. Correspondence: Dr. Timothy H. Self, Professor of Clinical Pharmacy, 26 S . Dunlap, University of Tennessee, M emphis, TN 38163.
272
tervention period (2.6 ± 0.2 pre-intervention versus 2.0 ± 0.2 post-intervention; P = 0.11). Both groups had significant reductions in hospitalizations, but the study group had a greater reduction. Sixty-two percent of study patients had complete elimination of ED visits and hospitalizations, whereas no control patients had total elimination of the need for institutional acute care. QOL in the study patients revealed significant improvements for most parameters. A comprehensive long-term management program emphasizing inhaled corticosteroids combined with other state-of-the-art management, including intensive patient education, improves outcomes in adult African-American asthmatics. KEY INDEXING TERMS: Asthma; Outcomes; Minorities; Inhaled corticosteroids; Patient education. [Am J Med Sci 1996;311(6):272-280.]
D
espite the avai.l ability of effective methods of long-term asthma management, morbidity and mortality rates that reflect this common condition are rising. 1 •2 In the United States, morbidity and mortality due to asthma are more common in African-Americans than Caucasians. 1 Asthma is estimated to result in costs of $6.2 billion in the United States each year, primarily due to inadequate long-term management that results in excessive acute care costs. 3 As a response to the problem of human suffering and economic costs associated with this condition that affects at least 10 million Americans, the National Institutes of Health (NIH) launched the National Asthma Education Program in 1991. The NIH publications Guidelines for the Diagnosis and Management of Asthma1 and International Consensus Report on Diagnosis and Management of Asthma2 are landmark documents that helped redefine the approach to optimal asthma management. Using the principles of stressing inhaled antiinflammatory therapy and in tenJune 1996 Volume 311 Number 6
Kelso et al
sive patient education, Mayo et al 4 reported reduced hospitalization rates in a group of adult asthmatics in New York City. Mayo and associates 4 studied primarily Hispanics, but also some African-Americans and Caucasians. The goal of our study was to further validate that a comprehensive asthma management program results in improved outcomes. Our program emphasized the basic principles in the NIH Guidelines. 1·2 In addition, our goal was to improve outcomes exclusively in adult African-Americans in a different geographic area of the United States as compared to Mayo et al. 4 Primary outcome measures for this study were emergency department (ED) visits and hospitalizations. The study was approved by the university Institutional Review Board. Methods
Beginning in August 1992, patients were recruited from local teaching hospitals, referrals, and through local public service announcements, including television, radio, and newspapers. Patient recruitment continued through April 1993. Inclusion criteria included being an African-American adult (~ 18 years of age), meeting NIH 1 criteria for moderate to severe asthma, and frequent ED visits (~ 5 in the previous 2 years or 3 in the previous 12 months) or hospitalizations (~ 2 during the previous 2 years), or at least 1 intensive care unit (ICU) admission in the previous 2 years. Exclusion criteria included chronic bronchitis/emphysema (chronic obstructive pulmonary disease [COPD]), pregnancy, inability to perform correct use of peak flow meter or metered dose inhaler plus Aerochamber (Monaghan Medical, Plattsburg, NY) despite adequate instruction, clinically significant cardiac disease, psychosis, or substance abuse. The university-based, board-certified pulmonologist investigator was careful to rule out COPD in any patient by using NIH criteria for the differential diagnosis of asthma. 1 After obtaining written informed consent, we obtained routine history, physical examination, blood chemistry, electrocardiogram, and spirometry for each patient. Special attention was given to rule out other medical conditions that can mimic asthma and to identify and treat endogenous triggers of asthma such as gastroesophageal reflux and sinusitis. Because we could not in good conscience deny stateof-the-art1·2 treatment and education for patients with whom we came in contact, our control group was identified via chart review, primarily at local hospitals. The medical record department from each hospital generated lists of all adult African-American asthmatics at least 18 years of age with the same criteria for numbers of ED visits or hospitalizations in the same time period as the study group. Once a diagnosis of asthma was verified via chart review by an investigator (eg, COPD ruled out), each ED visit and hospitalization was verified as being due to asthma. Exclusion criteria for the study group were applied to the control group. In reTHE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
viewing charts of potential control patients, if there was any significant smoking history (eg, smoking from teenage years to mid 30s), patients were not included in the control group. Furthermore, if there were any physician notes even questioning possible COPD, or any lung disease other than asthma, these patients were not included in the control group. In addition, the control group was selected to assure it was comparable to the study group for current employment, gender, and onset of asthma. Severity of asthma was obviously more difficult to assess, but was likely quite similar based on ED visits and hospitalizations. Because five of our patients were recruited from the surrounding region, we included five control patients from another hospital outside the Memphis area in West Tennessee. In Table 1, we summarize characteristics of the study and control groups. Educational Interventions. On enrollment in the study, each patient received 1 hour of asthma education from one of the doctor of pharmacy investigators. Education was provided to individual patients versus group educational sessions. Educational content was consistent with the recommendations of the NIH, 1·2 stressing that asthma is primarily an inflammatory condition and that inhaled antiinflammatory medication is absolutely critical to successful long-term management in moderate to severe disease. The four major components of asthma management were covered with each patient, including (1) environmental control, (2) objective monitoring at home with peak flow meters, (3) asthma education and partnership 1 between the patient and the health-care professionals in our clinic, and (4) medications. Throughout the initial encounter and subsequent visits, the concept of asthma prevention was stressed. Patients were told with enthusiasm, "Your asthma can be controlled: expect nothing less!" and the NIH National Asthma Education Program booklet with this title was given to each patient. Asthma triggers for each patient were reviewed, with techniques to eliminate them or minimize exposure. An Assess peak flow meter (HealthScan, Cedar Grove, NJ) was given to each patient with green (80-100% of personal best), Table 1. Demographics of Study and Control Patients
Age (mean) Gender( % female) Onset of asthma Childhood Adult Employedt (%) ICS* prescribed( %)
Study Group (n = 21)
Control Group (n = 18)
41 ± 2
35 ± 3 p = 0.12 72% p = 0.43
86% 48%
52% 95% 48%
72% p 28%
=
0.12
94% NS 39% p = 0.54
* ICS =Inhaled corticosteriod (prior to study initiation).
t Employed or in college or homemaker with spouse employed.
273
Management Program for Asthma
yellow (50-80%), and red (<50%) zones established, 1 including colored stickers placed directly on the peak flow meter. The investigator would always demonstrate use of the peak flow meter and have the patient demonstrate correct use on the initial and each subsequent visit. With each visit and telephone contact, the investigators made a concerted effort to maintain a partnership with the patient. As part of showing a genuine concern for each patient, active listening was an important part of our clinic. On each visit, we would ask the following question: "What is bothering you the most about your asthma right now?" Patients were involved in decision making with regard to therapy and how they felt about changes (eg, dosage adjustments). As part of self-management, patients were taught how to deal with increased symptoms early. In addition, each patient was given written instructions about crisis management (eg, procedure to follow if peak expiratory flow rate (PEFR) is in the "red zone"). Education with regard to medications received great emphasis in our program. Special care was given to stress the absolute importance of inhaled corticosteroids. Patients were told that their inhaled beclomethasone was the key to optimal long-term control of asthma, and that it must be used every day at the dose prescribed without ever missing a dose. An albuterol metered dose inhaler was given to each patient to prevent exercise-induced asthma and to treat symptoms "as needed." If two puffs of albuterol were needed more than three times daily, patients were instructed to contact the investigators for a dosage adjustment of their inhaled beclomethasone. Other medications that were required were explained to each patient verbally and in writing. Inhalation technique was carefully taught on the initial visit and checked on each subsequent visit. Teaching included discussion of each step, demonstrations by the investigators, and observation of each patient's technique on each visit. 5 •6 Patients were told that correct technique was vitally important to successful therapy. The frequency of clinic visits in the study group was initially every month, but as patients improved, we saw most patients every 2-3 months. Obviously, a few patients needed more frequent visits, whereas others did so well that visits were less frequent. Based on a review of hospital medical records in the control group, we could not determine the frequency of office visits. Therapeutic Interventions. Inhaled beclomethasone (Beclovent, Allen & Hanburys, Research Triangle Park, NC) was the only oral inhalation corticosteroid used in this study and was always administered via an Aerochamber. Individualized dosing of the inhaled corticosteroid was viewed by the investigators as being of major importance in optimizing therapy for these patients. Each patient's dose was tailored based on symptom control as well as peak flow monitoring and
274
inhaled albuterol requirements. The maximum dose used in our clinic was 40 puffs daily (42 ~tg/puff). Albuterol metered dose inhaler was used "as needed" and before exercise. An emergency supply of prednisone (40 mg daily for 3 days was based on the Mayo et al4 study) was given to each patient. If the PEFR was in the "red zone" and two puffs of albuterol did not bring the PEFR back to at least the "yellow zone" in 15 minutes along with some subjective relief, the patient was to call the investigators immediately and start the emergency supply of prednisone. Early use of systemic corticosteroids was clearly shown to prevent further dangerous escalation of asthma symptoms in many patients,4·7 and is recommended by the NIH Guidelines. 1•2 For routine clinic visits, patients were seen by one of the doctor of pharmacy investigators. Medication adjustments were done by NIH Guidelines. 1•2 Objective assessment of asthma was done at each clinic visit as well as review of the PEFR patient diary, symptom frequency, and "as needed" albuterol requirement. For unusual presentations or if patients were not clinically stable, the pulmonologist investigator was always available to see patients. Other medications required by some patients included occasional antibiotics for sinusitis, histamine (H 2 ) receptor antagonists for gastroesophageal reflux, and intranasal beclomethasone for allergic rhinitis. A low dose of sustained release theophylline (Uniphyl, Purdue-Frederick, Norwalk, CT) in the evening only was used as adjunctive therapy for nocturnal asthma only if tailored doses of inhaled beclomethasone did not eliminate nighttime or early morning symptoms. Oral beta agonists were not used in our patients. When the study was initiated, salmeterol was not available in the United States, and was not allowed until completion of the 2-year intervention. Outcomes Assessment. In both the intervention and control groups, numbers of ED visits and hospitalizations were the primary outcome indicators. For the intervention group only, three quality of life measures were also used, including the SF-36,8 •9 the Asthma Bother Profile, 10 and the Asthma Sleep Scale. 11 Each questionnaire was self-administered by the patient at baseline and every 6 months for the 2-year study period. Statistical Analysis. The data were extremely skewed because many patients had zero ED visits and zero hospitalizations for years 1 and 2 after intervention. However, conclusions were similar for parametric and non parametric analyses. Because mean numbers of ED visits and hospitalizations are directly interpretable, parametric results are reported, but levels of significance are taken from nonparametric tests. P values of 0.05 or less were considered statistically significant. For each group, the ED and hospitalization data were analyzed using nonparametric univariate analysis of variance for repeated measures (Friedman's test). For each group, preplanned comparisons were made within the group between the pre-intervention and post-inJune 1996 Volume 311 Number 6
Kelso et al
3.0
2.5 ~
·v;
> !il '0 .8e z"
2.0
1.5
1.0
0.5
0.0 Baseline
lst Year
2nd Year
Figure 1. Number of ED visits at baseline and after one and two years in clinic for both study and control patients. Baseline values were taken an average of two years before initiation of study (or if all visits in one year, then that number used). The values are mean ± standard error. *Significant at P < 0.05 in favor of the study group at first and second years.
tervention. In addition, preplanned comparisons were made between groups for the pre-intervention and postintervention periods. At each time, groups were compared by the Wilcoxon rank sum test. The Statistical Analysis System (SAS Institute, Cary, NC) version 6.08 software was used for these analyses. Mean scores from the SF -36, Asthma Bother Profile, and Asthma Sleep Scale were analyzed by repeated measures multivariate analysis of variance. Post hoc paired t-tests were performed to identify where specific differences existed. The Statistical Package for the Social Sciences was used for this analysis. Results
The study and control groups were comparable for all demographic variables assessed (Table 1). Both the study and control groups had similar numbers of ED visits and hospitalizations in the 2 years before the study interventions. Twenty-five study patients were initially enrolled in our clinic. Three were excluded due to diagnoses of co-existing sarcoidosis, vocal cord dysfunction, and a significant restrictive component. One patient was excluded after repeatedly seeing other
health providers who prescribed medications not included in our Institutional Review Board-approved protocol. Fourteen patients had been seeing a primary care physician and six patients had been seeing a specialist (ie, pulmonologist, allergist) before enrollment in our clinic. One patient had been relying solely on ED care. In the control group, 14 patients were under the care of a primary care physician and 4 patients were seeing a pulmonologist or allergist. A highly significant reduction in ED visits was found for the study patients after both the first and second years of the clinic (Figure 1). There was a 91% reduction of ED visits in the first year and an 83% reduction in the second year. Study patients had an 83% reduction in hospitalizations for the first and second years of our clinic (Table 2). In addition, 62% of patients in our clinic had total elimination of the need for institutional acute care management. In comparison, none of the patients in the control group had complete elimination of ED visits or hospitalizations. There was no reduction in ED visits in the control group after the first year of the intervention period (Figure 1). During the second year, there was a significant reduction in ED visits in the control group, but the reduction in the study group was greater (P < 0.05). Although the average number of hospitalizations for control patients was reduced after 1 year (Table 2), a comparison with the study group revealed a significant difference in the amount of reduction in favor of the clinic patients (P < 0.05)). Similarly, during the second year, the number of hospitalizations for both groups decreased, but the study group had a significantly greater reduction (P < 0.05). In Table 3, we summarize a comparison of 2-year mean ED visits and hospitalizations before and after intervention for both groups. Four of the patients in the intervention group had a history of ICU admission with intubation before admission to the study. Three of these patients had an excellent response and completed the 2-year trial. After having a good response to therapy for 18 months in the study, the other patient (age 55 at the time of this event with a history of ICU admission and intubation before enrollment) experienced sudden, severe asthma at home. Despite promptly calling 911, she died later that day in the ED. Two patients in the control group
Table 2. Hospitalizations Before and After Asthma Clinic* Before* Study group (n = 21) Control group (n = 18)
0.6 ± 0.1 [1 (0, 1)) 1.3 ± 0.2 [1 (0, 2))
Year 1 0 .1 ± O.lt 0.8 ± 0.2t
*t
[0 (0, 0)) [1 (0, 1))
Year 2 0 .1 ± O.lt 0.4 ± 0.2t
*t
[0, (0, 0)) (0 (0, 1))
* Median number of visits during 2 years before clinic (if all visits in one year, then that value used) [median (25th percentile, 75th percentile)]. Data are means ± standard error. t Significant at P < 0.05 compared with before intervention period. :j: Significant at P < 0.05 between groups during Year 1 and Year 2. THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
275
Management Program for Asthma
Table 3. Emergency Department Visits and Hospitalizations 2 Years Before and 2 Years After Asthma Clinic* Before Clinic Study group (n = 21) Emergency department visits Hospitalizations Control group (n = 18) Emergency department visits Hospitalizations
P Value
After Clinic
2.3 ± 0.2 [3 (1, 3)) 0.6 ± 0.1 [1 (0, 1))
0.6 ± 0.2t [0 (0, O)) 0.1 ± O.lt [0 (0, 0))
0.0001 0.002
2.6 ± 0.2 [3 (2, 3)) 1.3 ± 0.2 [1 (0, 2))
2.0 ± 0.2t [2 (1.5, 2)) 0.6 ± O.lt [0.5 (0, 1))
0.14 0.004
• Data are mean ± standard error. Brackets indicate median (25th percentile, 75th percentile). t Significant at P < 0.05 between groups.
had a history of ICU admission before the study intervention period, and both of these patients had an ICU admission during the intervention period. Table 4 is a description of the medical treatment and knowledge of asthma for the intervention group before and after enrollment in the clinic. Because our education was reinforced at each clinic visit and because the primary outcome was change in ED visits and hospitalizations, we did not formally evaluate patient knowledge with extensive testing. Although we did not calculate exact rates of clinic appointments kept in the study group, the rate of appointments kept was very high. Likewise, we did not use sophisticated methods to assess medication compliance, but the best measure of this parameter was the dramatic improvement in outcomes, which strongly suggests excellent medication compliance. Patients always brought their medications to the clinic so we could see them, and they would call us consistently for refills if they were getting low in supply of any medication. During the study period, 78% of control patients had inhaled corticosteroids prescribed (frequently in low doses), whereas all study patients were given individ-
Table 4. Study Group Patient Characteristics• Before Clinic Medical treatment Using daily ICS Average number puffs ICS Using spacer device Daily oral corticosteroids Using theophylline bid Using theophylline PM only Using home peak flow meter Seeing clinician regularly Seeing a specialist Knowledge of asthma Correct inhalation technique Difference in ICS and beta agonist Self-management, crisis prednisone
• % of patients (n
=
21).
ICS = inhaled corticosteroids.
276
After Clinic
48% 5.4 14% 14% 24 % 10% 0% 67 % 24 %
100% 25 100% 0% 0% 14% 100% 100% 100%
52% 0% 0%
100% 100% 100%
ualized doses of inhaled corticosteroids. In study patients, the mean inhaled beclomethasone dose was 25 puffs daily or 1,050 J,Lg daily (doses ranged from 4 puffs twice daily to 10 puffs four times daily). The mean number of times that the 3-day emergency supply of prednisone was required was 4.4. Fifteen of21 patients required 0-3 uses of emergency prednisone in the 2year study period. Although three patients began the study taking oral corticosteroids, all three were successfully started on inhaled beclomethasone and tapered from oral agents. Theophylline products were discontinued in five patients but were continued in three others because of persistent nocturnal symptoms. Eleven patients required intranasal corticosteroids for allergic rhinitis, and 6 patients required a 3-week course of appropriate antibiotics to treat sinusitis. In patients suspected of having gastroesophageal reflux, confirmation was obtained by 24-hour gastric pH monitoring. Subsequently, five patients were placed on Hz antagonists. Some patients required addition of metoclopramide or substitution with omeprazole. Ipratropium was added in two patients who had a significant psychogenic triggerY Two patients required premenstrual increase in inhaled corticosteroid dose. Peak expiratory flow rates improved significantly after a few months of clinic involvement. Baseline mean PEFR was 58% of eventual personal best versus 83% after 9-12 months in the clinic and 86% at the end of the trial. Each of our patients expressed appreciation for having an objective measure of their condition. Quality of life measures also showed significant improvements for most parameters for both years of intervention (Tables 5-7). Improvements generally occurred within the first 6 months and were maintained at roughly the same scores throughout the 2 years for most patients. Reductions in nocturnal asthma and the ability to exercise again are excellent examples of improved quality of life for our patients. No side effects of inhaled corticosteroids were reported by our patients, despite high doses. One patient reported gastrointestinal intolerance of pulse prednisone requiring treatment with an Hz antagonist, and June 1996 Volume 311 Number 6
Kelso et al
Table 5. SF -36 Scores at Baseline, 6, 12, 18, and 24 Months
SF -36 Dimension Health perception (F(4,52) = 2.64, P = .04) Physical functioning (F(4,44) = 1.74, P = .16) Limitations - physical (F(4,48) = 10.41, P = .00) Limitations - emotional (F(4,56) = 4.22, P = .01) Social functioning (F(4,56) = 6.81, P = .00) Mental health (F(4,52) = 2.74, P = .04) Bodily pain (F(4,48) = 1.50, P = .22) Energy /fatigue (F(4,44) = 3.12, P = .02)
Baseline (n = 19)
6 Months (n = 19)
44.86 (25.03)"·b.c
52.29 (25.74) 8 *
57.50 (28.23 )b *
50.27 (32.05)
57.99 (30.68)C *
52.92 (31.58)
65.42 (36.40)
71.25 (29.71)
61.58 (29.57)
70.42 (27.00)
11.53 (24.19)"•b,c,d
80.77 (37.02)" *
59.62 (43.94)b *
60.31 (42.99)C *
63.81 (42.52)d *
42.22 (49.55 )"·b
86.67 (35.19) 8 *•C
73.33 (38.21)
58.37 (43.70)c,d
80.00 (37.37)b *·d
55.83 (24.49) 'b,c
90.00 (22.76)" *
84.17 (23.84 )b *·d
68.93 (32.10)d.e
82.50 (23.53)c *••
62.57 (23.51)"·b,c
75.43 (20.27)" *
75.42 (19.46)b *
72.07 (25.35)
78.29 (20.74)C *
71.15 (25.28)
82.31 (24.93)
77.89 (21.23)
65.50 (33.11)
81.54 (25.20)
38.75 (29.40)"•b,c
63.75 (16.39)" *
62.08 (26.24)b *
57.00 (33.85)
61.85 (33.85)c *
8
12 Months (n = 19)
18 Months (n = 18)
24 Months (n = 18)
Data are means ± SD; scored such that the higher score represents the positive health attribute. Data analyzed using repeated measures analysis of variance. Means with the same superscript letter (a, b, c, d, or e) are significantly different at P .:5 0.05. * Significantly different from baseline.
one patient had glucose intolerance secondary to pulse prednisone. One patient reported anorexia with theophylline, which resolved after discontinuation of the drug. Discussion
This clinic was offered to poorly controlled minority asthmatics to determine whether a comprehensive long-term asthma management program that stressed the principles of the NIH Guidelines 1 •2 would improve outcomes. Because African-Americans have nearly three times the mortality rate from asthma than Caucasians, as well as a greater morbidity rate, 1 this population was targeted in our study. Lang and Polansky 13 recently reported an analysis of asthma deaths in Philadelphia, and found that black women were at increased risk of mortality. Other minorities were also at greater risk of asthma mortality, and poverty was another risk factor. This study was conducted in Memphis where African-Americans constitute approximately 50% of the population. Our patients were primarily working, middle-class individuals. Mayo and associates4 demonstrated that aggressive, state-of-the-art therapy combined with a concerted effort at patient education and easy access to clinicians reduced hospital readmission rates and duration of hospital stay in a group made up primarily of Hispanics. Our results reinforce the findings of Mayo et al 4 and several other groups of investigators 14- 18 that state-ofthe-art drug therapy and patient education improve outcomes in minority asthmatics. The effectiveness of our interventions was assessed in two ways-patients as their own control subjects and comparison with a THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
contemporary and comparable nonintervention control group. Despite very careful state-of-the-art management, including application of the principles of the NIH Guidelines, one of our patients with a history of severe asthma died from a sudden attack 18 months into the study. Asthmatics with a history ofiCU admission and intubation have a high mortality rate. 19- 22 Marquette et al 19 reported a posthospitalization mortality rate of 10.1% at 1 year and 14.4% at 3 years after near fatal asthma and mechanical ventilation. Unfortunately, these deaths are often preceded only by rapid, severe airflow obstruction. 23 •24 In an analysis of 80 asthma deaths from 1988-91, Campbell et al 21 found a mean age of 52 years, with 61% female and 32% with sudden collapse. Mayo et al 4 refer to "blitz" asthma characterized by severe respiratory failure within Yz hour of the onset of attack. Such attacks occurred in 4 of their special clinic patients during 32 months of follow-up. Ea(!h of these patients required ICU admission with intubation. Our patient apparently suffered such an attack, which occurred 2 years after her ICU admission with intubation. She successfully used her emergency supply of prednisone on several prior occasions, in conjunction with peak flow monitoring and use of inhaled albuterol while contacting the investigators. We were unable to determine which aspect of the clinic was primarily responsible for the improvement in outcome in the majority of our patients. One important aspect was the "partnership" described in the NIH Guidelines. 1 We placed great emphasis on listening actively, showing real concern for the patients, and involving the patients in therapeutic decisions (eg,
277
Management Program for Asthma
Table 6. Asthma Bother Profile Scores at Baseline, 6, 12, 18, and 24 Months
Item How much does your asthma bother . you at your paid work? (F(4,52) = 2.55, P = .05) you when you do jobs around the house? (F(4,56) = 3.69, P = .01) your social life? (F(4,56) = 3.45, P = .01) your personal life? (F(4,56) = 4.92, P = .00) your involvement in leisure activities? (F(4,60) = 4.80, P = .00) you when you sleep? (F(4,60) = 5.45, P = .00) How much . . . does the cost of your asthma medicines bother you? (F(4,48) = 6.70, P = .00) does the inconvenience or embarrassment of taking your asthma medicines bother you? (F(4,56) = 1.94, P = .11) How much . . . bother is the worry that you will have an asthma attack when visiting a new place? (F(4,56) = 2.79, P = .04) bother is the worry that you will catch a cold? (F(4,56) = 3.18, P = .02) bother is the worry that you will let others down? (F(4,56) = 4.08, P = .01) bother is the worry that your health may get worse in the future? (F(4,56) = 4.03, P = .01) bother is the worry that you won't be able to cope with an asthma attack? (F(4,56) = 4.92, P = .00) do coughs and colds bother you? (F(4,56) = 3.75, P = .01) are you bothered by feeling anxious, depressed, tired, or helpless? (F(4,52) = 2.76, P = .04)
Baseline (n = 20)
6 Months (n = 20)
12 Months (n = 19)
18 Months (n = 18)
24 Months (n = 18)
4.21 (1.25)"•b,c
2.57 (1.34) 8
*
2. 71 (1.38)b
*
3.21 (2 .12)
3.14 (2 .14)' *
3.67 (l.ll) 8 'b,c
2.67 (1.23)"
*
2.67 (1.29)b
*
3.07 (1.34)
2.67 (1.29)'
*
3.60 (1.68)"•b,c
2.20 (1.08) 8
*
2.60 (1.24)b
*
2.80 (1.27)
2.53 (1.64)'
*
3.53 (1.81) 8 'b,c,d
1.87 (0.92) 8
*
2.48 (1.36)b
*
2.40 ( 1.45 )'
*
2.40 (1.64)d
*
4.00 (1.55)"•b,c,d
2 .56 (1.46) 8
*
2. 75 (1.39)b
*
3.00 (1.41)'
*
2.69 (1.62)d
*
3.75 (1.57)"·b,c,d
2.19 (1.17)
8
*
2.63 (1.46)b
*
2.44 (1.15)'
*
2.50 (1.46)d.
5.00 (1.08)"•b,c,d
2.85 (1.82)"
*
2.85 (2.04)b
*
2.85 ( 1.99 )'
*
2.92 (2.10)d.
2.73 (1.62)
1.93 (1.49)
4.07 (1.83) 8
2.47 (1.13) 8
1.67 (1.18)
1.73 (1.39)
2.27 (2.31)
*
3.07 (1.62)
3.07 (1.49)
2.93 (1.83)
4.20 (1.94) 8
2.80 ( 1.4 7)" •
3.27 (1.67)
3.27 (1.49)
3.07 (1.53)
3.93 ( 1.67) 8 'b,c,d
2.80 (1. 70)"
*
3.00 (1.77)b
4.60 (1.45) 8 'b,c,d
3.27 (1.53) 8
*
3.60 ( 1. 72)b *·'
4.60 (1.35)"·b,c,d
3.13 (1.51) 8
*
3.33 (1.50)b
4.33 (1.50)"·b,c,d
3.40 (1.68) 8
*
4.36 (1.45)"•b,c,d
3.00 (1.71) 8
*
*
2.53 (1.60)'
*
2.67 (1.50)d *
3.13 (1. 73)'
*
3.00 (1.89)d
*·'
*
3.20 (1.32)'
*
2.87 (1.73)d
*
3.13 (1.51)b
*
3.53 (1.25)'
*
3.33 (1.50)d
*
3.29 (2.05)b
*
3.21 (1.67)'
*·'
2.93 (2.02)d *·'
Data are means ± SD; rated on a six point scale where a lower number indicates less bother. Data analyzed using repeated measures analysis of variance. Means with the same superscript letter (a, b, c, d, or e) are significantly different at P :5 .05.
* Significantly different from baseline.
change in doses or initiation of prednisone). Patients had short waiting times for clinic visits, and were given written instructions at each visit and easy telephone access to us for follow-up questions. Mellins et al 25 state that these aspects of care along with meeting the patient's expectations and patient's satisfaction are associated with improved compliance. Virtually all of our patients commented that "no one has ever taken the time to explain asthma and its management like you have." Several patients were excited to tell us that they
278
could exercise (eg, brisk walking and softball) again, and the difference it made in their quality of life. This asthma clinic and study was initiated and managed by doctors of pharmacy with a pulmonologist. In other studies, it was shown that pharmacists working in clinics as a team with physicians improve outcomes in asthmatics. 17 •18•26 Pauley et al 17 applied the NIH Guidelines in minority patients and dramatically reduced ED visits by 87%. Whereas our current report involved primarily minorities who were employed, we June 1996 Volume 311 Number6
Kelso et al
Table 7. Asthma Sleep Scale Scores at Baseline, 6, 12, 18 and 24 Months
Item Sleep loss (F(4,60) = 7.49, P = .00) Work/school interference (F(4,60) = 7.44, P = .00) Nights awakened (F(4,60) = 5.70, P = .01)
Baseline (n = 20)
6 Months (n = 20)
12 Months (n = 20)
18 Months (n = 18)
3.50 (1.37)"•b,c,d
1. 75 (1.07)" *
1.88 (1.26)b *
2.13 (1.26)• *
2.13 (1.36)d *
3.60 (1.45)"•b,c
1.94 (1.48)" *
1.94 (1. 33)b *
2.09 (1.20)• *
2.41 (1.70)
4.38 (1.75)"·b,c,d
1.75 (1.07)" •.•
2.50 (2.07)b *
2.69 (1.82)• • .•
2.44 ( 1.87)d •
24 Months (n = 18)
Data are means ± SD; a higher rating indicates a more negative effect on sleep. Data analyzed using repeated measures analysis of variance. M eans with the same superscript letter (a, b, c, d, or e) are significantly different at P s .05. • Significantly different from baseline.
previously reported a 41% reduction in ED visits in a group of indigent minorities.18 An asthma clinic in Kaiser Permanente in California reduced ED visits by 66% and hospitalizations by 48%.26 An obvious benefit of close teamwork between clinical pharmacy and medicine is that once there is agreement to apply the NIH Guidelines, the pharmacist can take adequate time to educate the patient and help in self-management. The busy physician has more time for other patients, yet asthma patients have enhanced outcomes. Obviously, for difficult patients or when there are concerns regarding concomitant medical problems, the protocol is set to have the physician spend more time with these patients. For routine visits when patients are doing well objectively and subjectively, the physician is freed to spend more time with other patients. As in our study and others previously cited, other groups of investigators also found that concerted efforts at patient education help improve outcomes in adult asthmatics. 27-29 Some of these reports involve highly structured educational programs. 27·28 The study group had significantly greater reductions in hospitalizations than the control group. Still, the control group had a reduction, and one potential explanation for this decrease in hospitalizations is that the prescription of inhaled corticosteroids increased from 39% (prestudy period) to 78% during the intervention period. Certainly, this is only a potential explanation, because we could not assess inhalation technique or level of patient education given the control subjects. It is reasonable to speculate that the increased use of inhaled corticosteroids in the control group resulted from gradual incorporation of the NIH Guidelines into the practice of these clinicians. A limitation of our study is lack of randomization. There is a possible selection bias in that our study patients responded to public service announcements and were perhaps more motivated and receptive to education and treatment compliance. In addition, via the informed consent process, study patients were aware of the purpose of the study. Therefore, one could argue that participation in the study itself might have been THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
responsible for some reduction in ED use. Our view is that we absolutely urged patients to seek emergency care based on protocol previously described, and we doubt that being in the study affected patient behavior of possibly not going to the ED. Another potential limitation would have been if the control group did not have easy access to medical care. From our review of medical records of the control group, there was every indication that each patient had easy access to primary care physicians or subspecialists. Another variable to consider is that whereas we believe the two groups are comparable (Table 1), we were not able to assess salary ranges, level of education, and living conditions in the control group. Although a full discussion of our quality of life data is beyond the scope of this article, a brief comment is appropriate. For each of the three validated instruments,8-11 a majority of quality of life parameters showed significant improvements for both years of the intervention period. For example, note in Tables 5-7 improved health perception, less physical limitations, fewer nocturnal awakenings, and less bother with asthma at home, work, and in social life. These types of improvements along with decreased ED visits obviously have a highly significant impact on quality of life. Although we did not assess lost work or school days before the study intervention, some quality of life measures (eg, asthma bother at work) imply an improvement in these parameters. In addition, patients with dramatic reductions in need for acute care would be expected to have reduction in lost work or school days. Cost implications of our results are noteworthy in that $1.86 billion (in 1990 dollars) are estimated to be spent in the United States for ED visits and hospitalizations. An initial investment in time educating the patient, planning therapy, and establishing a partnership with the patient can pay very large dividends in outcomes. Our results combined with similar results of other investigators suggest the cost of institutional acute care for asthma can be reduced dramatically when state-of-the-art, long-term management is applied.
279
Management Program for Asthma
Since the initiation of our study, recent doubleblinded trials demonstrated that addition of salmeterol to low-to-moderate doses of inhaled corticosteroids provides better outcomes than further increases in inhaled antiinflammatory therapy. 30•31 If other long-term trials validate this research, that is an appealing strategy to pursue. The International Consensus Report 2 already includes this approach as an option (before in~ creasing inhaled corticosteroid dose beyond 1,000 ,.,g daily). This study extends the findings of other investigators that a comprehensive long-term asthma management program that emphasizes the principles of the NIH Guidelines can improve outcomes in adult minorities. Clearly, specific program elements in state-of-the-art, long-term asthma management are interdependent. For example, individualized inhaled corticosteroid therapy must be accompanied by repetitive patient education regarding the purpose of the medication and correct inhalation technique. Further studies are recommended in larger populations. Acknowledgments We thank Elizabeth Tolley, Ph.D. for her assistance in data analysis. References 1. Expert Panel Report. Guidelines for the diagnosis and management of asthma. National Institutes of Health. Publication No. 91-3042, 1991. 2. International Report. International consensus report on diagnosis and treatment of asthma. National Institutes of Health. Publication No. 92-3091 , 1992. 3. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. N Eng! J Med. 1992;326:862-6. 4. Mayo PH, Richman J, Harris HW. Results of a program to reduce admissions for adult asthma. Ann Intern Med. 1990;112: 864-71. 5. Newman SP, Pavia D, Clarke SW. How should a pressurized beta adrenergic bronchodilator be inhaled? Eur J Respir Dis. 1981;62:3-21. 6. Self TH, Rumbak MJ, Kelso TM. Correct use of metered dose inhalers and spacers. Postgrad Med. 1992;92:95-106. 7. Brunette MG, Lands L, Thibodeau L. Childhood asthma: Prevention of attacks with short-term corticosteroid treatment of upper respiratory tract infection. Pediatrics. 1988;81:624-9. 8. Bousquet J, Knanl J, Dhlvert H, Richard A, Chicove A, Ware JE, Jr, et al. Quality of life in asthma: internal consistency and validity of the SF -36 questionnaire. Am J Respir Crit Care Med. 1994;149:371- 5. 9. Ware JE, Sherbourne CD. The MOS 36-item short form health surve1r (SF-36) . Med Care. 1992;30:473-83. 10. Hyland ME, Ley A, Fisher DW, Woodward V. Measurement of psychological distress in asthma and asthma management programs. Br J Clin Psycho!. 1995;34:601-11. 11. Boyer G, Aaronson D, Meltzer E, Laforce C, Grossman J, Yancey S. Enhancement of a general quality of life scale: Validation of a sleep scale (abstract). J Allergy Clin Immunol. 1992;89:186.
280
12. Rumbak MJ, Kelso TM, Arheart KL, Self TH. Perception of anxiety as a contributing factor of asthma: Indigent versus nonindigent. J Asthma. 1993;30:165-9. 13. Lang DM, Polansky M. Patterns of asthma mortality in Philadelphia from 1969 to 1991. N Eng! J Med. 1994;31:1542-6. 14. Bolton MD, Tilley BA, Kuder J, Reeves T, Schultz LR. The cost and effectiveness of an education program for adults who have asthma. J Gen Intern Med. 1991;6:401-7. 15. Clark NM, Feldman CH, Evans, Levinson MJ, Wasilewski Y, Mellins RB. The impact of health education on frequency and cost of health use by low income children with asthma. J Allergy Clin Immunol. 1986;78:108-15. 16. Perla EN. A program to reduce emergency room utilization and hospitalizations for inner-city asthmatics. First National Conference on Asthma Management. Arlington, VA., 1992 (abstract). 17. Pauley TR, Magee MJ, Cury JD. Results of a pharmacy managed, physician directed program to reduce emergency department visits in a group of inner city adult asthma pat ients. Ann Pharmacother. 1995;29:5-9. 18. Kelso TM, Self TH, Rumbak MJ, Stephens MA, Garrett W, Arheart KL. Educational and long term therapeutic intervention in the emergency department: Effect on outcomes in adult indigent minority asthmatics. Am J Emerg Med. 1995;13: 632-7. 19. Marquette CH, Saulvier F, Leroy 0, Wallbert B, Chopin C, Demarq JM, et al. Long term prognosis of near fatal asthma. Am Rev Respir Dis. 1992;146:76-81. 20. Richards GN, Kolbe J, Fenwick J, Rea HH. Demographic characteristics of patients with severe life threatening asthma: Comparison with asthma deaths. Thorax. 1993;48:1105-9. 21. Campbell DA, McLennan G, Coates JR, Frith PA, Gluyas P A, Latimer KM, et al. A comparison of asthma deaths and near fatal asthma attacks in South Australia. Eur Respir J . 1994;7:490-7. 22. Westerman DE, Benatar SR, Potgieter PD, Ferguson AD. Identification of the high-risk asthmatic patient. Am J Med. 1979;66:565- 72. 23. Mackonald JB , Seaton A, Williams DA. Asthma deaths in Cardiff, 1963-74:90 deaths outside hospital. BMJ. 1976;2:7213. 24. Bateman JRM, Clarke SW. Sudden death in asthma. Thorax. 1979;34:40-4. 25. Mellins RB, Evans D, Zimmerman B, Clark NM. Patient compliance: Are we wasting our time and don't know it? Am Rev Respir Dis. 1992;146;1376-7. 26. lm J. Evaluation of the effectiveness of an asthma clinic managed by an ambulatory care pharmacist. California Journal of Hospital Pharmacy. 1993;5:5-6. 27. Wilson SR, Scamagas P, German DF, Hughes GH, Lulla S, Coss S, et al. A controlled trial of two forms of self-management education for adults with asthma. Am J Med. 1993;94: 564-75. 28. Muhlhauser Richter B, Kraut D, Weske G, Worth H, Berger M. Evaluation of a structured treatment and teaching program on asthma. J Intern Med. 1991;230:157- 64. 29. Zieger RS, Heller, Mellon S, Wald J, Falkoff R, Schatz M. Facilitated referral to asthma specialist reduces relapses in asthma emergency room visits. J Allergy Clin lmmunol. 1991;87: 1160- 8. 30. Greening AP, Ind PW, Northfield M Shaw G. Added salmeterol versus higher dose corticosteroid in asthma patients with symptoms on existing inhaled corticosteroid. Lancet. 1994;344: 219-24. 31. Woolcock A, Lundback B, Ringdal OLN, Jacques LA. Comparison of the effect of addition of salmeterol with doubling the inhaled steroid dose in asthmatic patients (abstract). Am J Resp Crit Care Med. 1994;149:A280.
June 1996 Volume 311 Number 6
ABSTRACT: To determine if a comprehensive long-term management program, emphasizing inhaled corticosteroids and patient education, would improve outcomes in adult AfricanAmerican asthmatics a nonrandomized control trial with a 2-year intervention was performed in a university-based clinic. Inclusion criteria consisted of (~5) emergency department (ED) visits or hospitalizations (~2) during the previous 2 years. Intervention patients were volunteers; a comparable control group was ident\fied via chart review at hospitals within the same area and time period as the intervention patients. Individualized doses of beclomethasone with a spacer, inhaled albuterol "as needed," and crisis prednisone were the primary therapies. Environmental control, peak flow monitoring, and a partnership with the patient were emphasized. Detailed patient education was an integral part of management. Control patients received usual care from local physicians. ED visits and hospitalizations for 2 years before and 2 years during the intervention period were compared. Quality of life (QOL) measurements were made at baseline and every 6 months in the intervention group. Study group (n = 21) had a significant reduction in ED visits (2.3 ± 0.2 pre-intervention versus 0.6 ± 0.2 post-intervention; P = 0.0001). Control group (n = 18) did not have a significant change in ED visits during the 2-year post-inFrom the *Department of Clinical Pharmacy, University of Tennessee, Memphis, the t Department of Medicine, King Faisal Hospital & Research Center, Riyadh, Saudi Arabia, tMethodist Hospital, M emphis, Tennessee, and the §Cleveland Clinic, Cleveland, Ohio. Supported in part by an investigator-initiated grant from Glaxo W ellcome Inc. and by a grant (M01-RR002211) from General Clinic Research Center. Submitted November 5, 1995; accepted in revised form February 6, 1996. Correspondence: Dr. Timothy H. Self, Professor of Clinical Pharmacy, 26 S . Dunlap, University of Tennessee, M emphis, TN 38163.
272
tervention period (2.6 ± 0.2 pre-intervention versus 2.0 ± 0.2 post-intervention; P = 0.11). Both groups had significant reductions in hospitalizations, but the study group had a greater reduction. Sixty-two percent of study patients had complete elimination of ED visits and hospitalizations, whereas no control patients had total elimination of the need for institutional acute care. QOL in the study patients revealed significant improvements for most parameters. A comprehensive long-term management program emphasizing inhaled corticosteroids combined with other state-of-the-art management, including intensive patient education, improves outcomes in adult African-American asthmatics. KEY INDEXING TERMS: Asthma; Outcomes; Minorities; Inhaled corticosteroids; Patient education. [Am J Med Sci 1996;311(6):272-280.]
D
espite the avai.l ability of effective methods of long-term asthma management, morbidity and mortality rates that reflect this common condition are rising. 1 •2 In the United States, morbidity and mortality due to asthma are more common in African-Americans than Caucasians. 1 Asthma is estimated to result in costs of $6.2 billion in the United States each year, primarily due to inadequate long-term management that results in excessive acute care costs. 3 As a response to the problem of human suffering and economic costs associated with this condition that affects at least 10 million Americans, the National Institutes of Health (NIH) launched the National Asthma Education Program in 1991. The NIH publications Guidelines for the Diagnosis and Management of Asthma1 and International Consensus Report on Diagnosis and Management of Asthma2 are landmark documents that helped redefine the approach to optimal asthma management. Using the principles of stressing inhaled antiinflammatory therapy and in tenJune 1996 Volume 311 Number 6
Kelso et al
sive patient education, Mayo et al 4 reported reduced hospitalization rates in a group of adult asthmatics in New York City. Mayo and associates 4 studied primarily Hispanics, but also some African-Americans and Caucasians. The goal of our study was to further validate that a comprehensive asthma management program results in improved outcomes. Our program emphasized the basic principles in the NIH Guidelines. 1·2 In addition, our goal was to improve outcomes exclusively in adult African-Americans in a different geographic area of the United States as compared to Mayo et al. 4 Primary outcome measures for this study were emergency department (ED) visits and hospitalizations. The study was approved by the university Institutional Review Board. Methods
Beginning in August 1992, patients were recruited from local teaching hospitals, referrals, and through local public service announcements, including television, radio, and newspapers. Patient recruitment continued through April 1993. Inclusion criteria included being an African-American adult (~ 18 years of age), meeting NIH 1 criteria for moderate to severe asthma, and frequent ED visits (~ 5 in the previous 2 years or 3 in the previous 12 months) or hospitalizations (~ 2 during the previous 2 years), or at least 1 intensive care unit (ICU) admission in the previous 2 years. Exclusion criteria included chronic bronchitis/emphysema (chronic obstructive pulmonary disease [COPD]), pregnancy, inability to perform correct use of peak flow meter or metered dose inhaler plus Aerochamber (Monaghan Medical, Plattsburg, NY) despite adequate instruction, clinically significant cardiac disease, psychosis, or substance abuse. The university-based, board-certified pulmonologist investigator was careful to rule out COPD in any patient by using NIH criteria for the differential diagnosis of asthma. 1 After obtaining written informed consent, we obtained routine history, physical examination, blood chemistry, electrocardiogram, and spirometry for each patient. Special attention was given to rule out other medical conditions that can mimic asthma and to identify and treat endogenous triggers of asthma such as gastroesophageal reflux and sinusitis. Because we could not in good conscience deny stateof-the-art1·2 treatment and education for patients with whom we came in contact, our control group was identified via chart review, primarily at local hospitals. The medical record department from each hospital generated lists of all adult African-American asthmatics at least 18 years of age with the same criteria for numbers of ED visits or hospitalizations in the same time period as the study group. Once a diagnosis of asthma was verified via chart review by an investigator (eg, COPD ruled out), each ED visit and hospitalization was verified as being due to asthma. Exclusion criteria for the study group were applied to the control group. In reTHE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
viewing charts of potential control patients, if there was any significant smoking history (eg, smoking from teenage years to mid 30s), patients were not included in the control group. Furthermore, if there were any physician notes even questioning possible COPD, or any lung disease other than asthma, these patients were not included in the control group. In addition, the control group was selected to assure it was comparable to the study group for current employment, gender, and onset of asthma. Severity of asthma was obviously more difficult to assess, but was likely quite similar based on ED visits and hospitalizations. Because five of our patients were recruited from the surrounding region, we included five control patients from another hospital outside the Memphis area in West Tennessee. In Table 1, we summarize characteristics of the study and control groups. Educational Interventions. On enrollment in the study, each patient received 1 hour of asthma education from one of the doctor of pharmacy investigators. Education was provided to individual patients versus group educational sessions. Educational content was consistent with the recommendations of the NIH, 1·2 stressing that asthma is primarily an inflammatory condition and that inhaled antiinflammatory medication is absolutely critical to successful long-term management in moderate to severe disease. The four major components of asthma management were covered with each patient, including (1) environmental control, (2) objective monitoring at home with peak flow meters, (3) asthma education and partnership 1 between the patient and the health-care professionals in our clinic, and (4) medications. Throughout the initial encounter and subsequent visits, the concept of asthma prevention was stressed. Patients were told with enthusiasm, "Your asthma can be controlled: expect nothing less!" and the NIH National Asthma Education Program booklet with this title was given to each patient. Asthma triggers for each patient were reviewed, with techniques to eliminate them or minimize exposure. An Assess peak flow meter (HealthScan, Cedar Grove, NJ) was given to each patient with green (80-100% of personal best), Table 1. Demographics of Study and Control Patients
Age (mean) Gender( % female) Onset of asthma Childhood Adult Employedt (%) ICS* prescribed( %)
Study Group (n = 21)
Control Group (n = 18)
41 ± 2
35 ± 3 p = 0.12 72% p = 0.43
86% 48%
52% 95% 48%
72% p 28%
=
0.12
94% NS 39% p = 0.54
* ICS =Inhaled corticosteriod (prior to study initiation).
t Employed or in college or homemaker with spouse employed.
273
Management Program for Asthma
yellow (50-80%), and red (<50%) zones established, 1 including colored stickers placed directly on the peak flow meter. The investigator would always demonstrate use of the peak flow meter and have the patient demonstrate correct use on the initial and each subsequent visit. With each visit and telephone contact, the investigators made a concerted effort to maintain a partnership with the patient. As part of showing a genuine concern for each patient, active listening was an important part of our clinic. On each visit, we would ask the following question: "What is bothering you the most about your asthma right now?" Patients were involved in decision making with regard to therapy and how they felt about changes (eg, dosage adjustments). As part of self-management, patients were taught how to deal with increased symptoms early. In addition, each patient was given written instructions about crisis management (eg, procedure to follow if peak expiratory flow rate (PEFR) is in the "red zone"). Education with regard to medications received great emphasis in our program. Special care was given to stress the absolute importance of inhaled corticosteroids. Patients were told that their inhaled beclomethasone was the key to optimal long-term control of asthma, and that it must be used every day at the dose prescribed without ever missing a dose. An albuterol metered dose inhaler was given to each patient to prevent exercise-induced asthma and to treat symptoms "as needed." If two puffs of albuterol were needed more than three times daily, patients were instructed to contact the investigators for a dosage adjustment of their inhaled beclomethasone. Other medications that were required were explained to each patient verbally and in writing. Inhalation technique was carefully taught on the initial visit and checked on each subsequent visit. Teaching included discussion of each step, demonstrations by the investigators, and observation of each patient's technique on each visit. 5 •6 Patients were told that correct technique was vitally important to successful therapy. The frequency of clinic visits in the study group was initially every month, but as patients improved, we saw most patients every 2-3 months. Obviously, a few patients needed more frequent visits, whereas others did so well that visits were less frequent. Based on a review of hospital medical records in the control group, we could not determine the frequency of office visits. Therapeutic Interventions. Inhaled beclomethasone (Beclovent, Allen & Hanburys, Research Triangle Park, NC) was the only oral inhalation corticosteroid used in this study and was always administered via an Aerochamber. Individualized dosing of the inhaled corticosteroid was viewed by the investigators as being of major importance in optimizing therapy for these patients. Each patient's dose was tailored based on symptom control as well as peak flow monitoring and
274
inhaled albuterol requirements. The maximum dose used in our clinic was 40 puffs daily (42 ~tg/puff). Albuterol metered dose inhaler was used "as needed" and before exercise. An emergency supply of prednisone (40 mg daily for 3 days was based on the Mayo et al4 study) was given to each patient. If the PEFR was in the "red zone" and two puffs of albuterol did not bring the PEFR back to at least the "yellow zone" in 15 minutes along with some subjective relief, the patient was to call the investigators immediately and start the emergency supply of prednisone. Early use of systemic corticosteroids was clearly shown to prevent further dangerous escalation of asthma symptoms in many patients,4·7 and is recommended by the NIH Guidelines. 1•2 For routine clinic visits, patients were seen by one of the doctor of pharmacy investigators. Medication adjustments were done by NIH Guidelines. 1•2 Objective assessment of asthma was done at each clinic visit as well as review of the PEFR patient diary, symptom frequency, and "as needed" albuterol requirement. For unusual presentations or if patients were not clinically stable, the pulmonologist investigator was always available to see patients. Other medications required by some patients included occasional antibiotics for sinusitis, histamine (H 2 ) receptor antagonists for gastroesophageal reflux, and intranasal beclomethasone for allergic rhinitis. A low dose of sustained release theophylline (Uniphyl, Purdue-Frederick, Norwalk, CT) in the evening only was used as adjunctive therapy for nocturnal asthma only if tailored doses of inhaled beclomethasone did not eliminate nighttime or early morning symptoms. Oral beta agonists were not used in our patients. When the study was initiated, salmeterol was not available in the United States, and was not allowed until completion of the 2-year intervention. Outcomes Assessment. In both the intervention and control groups, numbers of ED visits and hospitalizations were the primary outcome indicators. For the intervention group only, three quality of life measures were also used, including the SF-36,8 •9 the Asthma Bother Profile, 10 and the Asthma Sleep Scale. 11 Each questionnaire was self-administered by the patient at baseline and every 6 months for the 2-year study period. Statistical Analysis. The data were extremely skewed because many patients had zero ED visits and zero hospitalizations for years 1 and 2 after intervention. However, conclusions were similar for parametric and non parametric analyses. Because mean numbers of ED visits and hospitalizations are directly interpretable, parametric results are reported, but levels of significance are taken from nonparametric tests. P values of 0.05 or less were considered statistically significant. For each group, the ED and hospitalization data were analyzed using nonparametric univariate analysis of variance for repeated measures (Friedman's test). For each group, preplanned comparisons were made within the group between the pre-intervention and post-inJune 1996 Volume 311 Number 6
Kelso et al
3.0
2.5 ~
·v;
> !il '0 .8e z"
2.0
1.5
1.0
0.5
0.0 Baseline
lst Year
2nd Year
Figure 1. Number of ED visits at baseline and after one and two years in clinic for both study and control patients. Baseline values were taken an average of two years before initiation of study (or if all visits in one year, then that number used). The values are mean ± standard error. *Significant at P < 0.05 in favor of the study group at first and second years.
tervention. In addition, preplanned comparisons were made between groups for the pre-intervention and postintervention periods. At each time, groups were compared by the Wilcoxon rank sum test. The Statistical Analysis System (SAS Institute, Cary, NC) version 6.08 software was used for these analyses. Mean scores from the SF -36, Asthma Bother Profile, and Asthma Sleep Scale were analyzed by repeated measures multivariate analysis of variance. Post hoc paired t-tests were performed to identify where specific differences existed. The Statistical Package for the Social Sciences was used for this analysis. Results
The study and control groups were comparable for all demographic variables assessed (Table 1). Both the study and control groups had similar numbers of ED visits and hospitalizations in the 2 years before the study interventions. Twenty-five study patients were initially enrolled in our clinic. Three were excluded due to diagnoses of co-existing sarcoidosis, vocal cord dysfunction, and a significant restrictive component. One patient was excluded after repeatedly seeing other
health providers who prescribed medications not included in our Institutional Review Board-approved protocol. Fourteen patients had been seeing a primary care physician and six patients had been seeing a specialist (ie, pulmonologist, allergist) before enrollment in our clinic. One patient had been relying solely on ED care. In the control group, 14 patients were under the care of a primary care physician and 4 patients were seeing a pulmonologist or allergist. A highly significant reduction in ED visits was found for the study patients after both the first and second years of the clinic (Figure 1). There was a 91% reduction of ED visits in the first year and an 83% reduction in the second year. Study patients had an 83% reduction in hospitalizations for the first and second years of our clinic (Table 2). In addition, 62% of patients in our clinic had total elimination of the need for institutional acute care management. In comparison, none of the patients in the control group had complete elimination of ED visits or hospitalizations. There was no reduction in ED visits in the control group after the first year of the intervention period (Figure 1). During the second year, there was a significant reduction in ED visits in the control group, but the reduction in the study group was greater (P < 0.05). Although the average number of hospitalizations for control patients was reduced after 1 year (Table 2), a comparison with the study group revealed a significant difference in the amount of reduction in favor of the clinic patients (P < 0.05)). Similarly, during the second year, the number of hospitalizations for both groups decreased, but the study group had a significantly greater reduction (P < 0.05). In Table 3, we summarize a comparison of 2-year mean ED visits and hospitalizations before and after intervention for both groups. Four of the patients in the intervention group had a history of ICU admission with intubation before admission to the study. Three of these patients had an excellent response and completed the 2-year trial. After having a good response to therapy for 18 months in the study, the other patient (age 55 at the time of this event with a history of ICU admission and intubation before enrollment) experienced sudden, severe asthma at home. Despite promptly calling 911, she died later that day in the ED. Two patients in the control group
Table 2. Hospitalizations Before and After Asthma Clinic* Before* Study group (n = 21) Control group (n = 18)
0.6 ± 0.1 [1 (0, 1)) 1.3 ± 0.2 [1 (0, 2))
Year 1 0 .1 ± O.lt 0.8 ± 0.2t
*t
[0 (0, 0)) [1 (0, 1))
Year 2 0 .1 ± O.lt 0.4 ± 0.2t
*t
[0, (0, 0)) (0 (0, 1))
* Median number of visits during 2 years before clinic (if all visits in one year, then that value used) [median (25th percentile, 75th percentile)]. Data are means ± standard error. t Significant at P < 0.05 compared with before intervention period. :j: Significant at P < 0.05 between groups during Year 1 and Year 2. THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
275
Management Program for Asthma
Table 3. Emergency Department Visits and Hospitalizations 2 Years Before and 2 Years After Asthma Clinic* Before Clinic Study group (n = 21) Emergency department visits Hospitalizations Control group (n = 18) Emergency department visits Hospitalizations
P Value
After Clinic
2.3 ± 0.2 [3 (1, 3)) 0.6 ± 0.1 [1 (0, 1))
0.6 ± 0.2t [0 (0, O)) 0.1 ± O.lt [0 (0, 0))
0.0001 0.002
2.6 ± 0.2 [3 (2, 3)) 1.3 ± 0.2 [1 (0, 2))
2.0 ± 0.2t [2 (1.5, 2)) 0.6 ± O.lt [0.5 (0, 1))
0.14 0.004
• Data are mean ± standard error. Brackets indicate median (25th percentile, 75th percentile). t Significant at P < 0.05 between groups.
had a history of ICU admission before the study intervention period, and both of these patients had an ICU admission during the intervention period. Table 4 is a description of the medical treatment and knowledge of asthma for the intervention group before and after enrollment in the clinic. Because our education was reinforced at each clinic visit and because the primary outcome was change in ED visits and hospitalizations, we did not formally evaluate patient knowledge with extensive testing. Although we did not calculate exact rates of clinic appointments kept in the study group, the rate of appointments kept was very high. Likewise, we did not use sophisticated methods to assess medication compliance, but the best measure of this parameter was the dramatic improvement in outcomes, which strongly suggests excellent medication compliance. Patients always brought their medications to the clinic so we could see them, and they would call us consistently for refills if they were getting low in supply of any medication. During the study period, 78% of control patients had inhaled corticosteroids prescribed (frequently in low doses), whereas all study patients were given individ-
Table 4. Study Group Patient Characteristics• Before Clinic Medical treatment Using daily ICS Average number puffs ICS Using spacer device Daily oral corticosteroids Using theophylline bid Using theophylline PM only Using home peak flow meter Seeing clinician regularly Seeing a specialist Knowledge of asthma Correct inhalation technique Difference in ICS and beta agonist Self-management, crisis prednisone
• % of patients (n
=
21).
ICS = inhaled corticosteroids.
276
After Clinic
48% 5.4 14% 14% 24 % 10% 0% 67 % 24 %
100% 25 100% 0% 0% 14% 100% 100% 100%
52% 0% 0%
100% 100% 100%
ualized doses of inhaled corticosteroids. In study patients, the mean inhaled beclomethasone dose was 25 puffs daily or 1,050 J,Lg daily (doses ranged from 4 puffs twice daily to 10 puffs four times daily). The mean number of times that the 3-day emergency supply of prednisone was required was 4.4. Fifteen of21 patients required 0-3 uses of emergency prednisone in the 2year study period. Although three patients began the study taking oral corticosteroids, all three were successfully started on inhaled beclomethasone and tapered from oral agents. Theophylline products were discontinued in five patients but were continued in three others because of persistent nocturnal symptoms. Eleven patients required intranasal corticosteroids for allergic rhinitis, and 6 patients required a 3-week course of appropriate antibiotics to treat sinusitis. In patients suspected of having gastroesophageal reflux, confirmation was obtained by 24-hour gastric pH monitoring. Subsequently, five patients were placed on Hz antagonists. Some patients required addition of metoclopramide or substitution with omeprazole. Ipratropium was added in two patients who had a significant psychogenic triggerY Two patients required premenstrual increase in inhaled corticosteroid dose. Peak expiratory flow rates improved significantly after a few months of clinic involvement. Baseline mean PEFR was 58% of eventual personal best versus 83% after 9-12 months in the clinic and 86% at the end of the trial. Each of our patients expressed appreciation for having an objective measure of their condition. Quality of life measures also showed significant improvements for most parameters for both years of intervention (Tables 5-7). Improvements generally occurred within the first 6 months and were maintained at roughly the same scores throughout the 2 years for most patients. Reductions in nocturnal asthma and the ability to exercise again are excellent examples of improved quality of life for our patients. No side effects of inhaled corticosteroids were reported by our patients, despite high doses. One patient reported gastrointestinal intolerance of pulse prednisone requiring treatment with an Hz antagonist, and June 1996 Volume 311 Number 6
Kelso et al
Table 5. SF -36 Scores at Baseline, 6, 12, 18, and 24 Months
SF -36 Dimension Health perception (F(4,52) = 2.64, P = .04) Physical functioning (F(4,44) = 1.74, P = .16) Limitations - physical (F(4,48) = 10.41, P = .00) Limitations - emotional (F(4,56) = 4.22, P = .01) Social functioning (F(4,56) = 6.81, P = .00) Mental health (F(4,52) = 2.74, P = .04) Bodily pain (F(4,48) = 1.50, P = .22) Energy /fatigue (F(4,44) = 3.12, P = .02)
Baseline (n = 19)
6 Months (n = 19)
44.86 (25.03)"·b.c
52.29 (25.74) 8 *
57.50 (28.23 )b *
50.27 (32.05)
57.99 (30.68)C *
52.92 (31.58)
65.42 (36.40)
71.25 (29.71)
61.58 (29.57)
70.42 (27.00)
11.53 (24.19)"•b,c,d
80.77 (37.02)" *
59.62 (43.94)b *
60.31 (42.99)C *
63.81 (42.52)d *
42.22 (49.55 )"·b
86.67 (35.19) 8 *•C
73.33 (38.21)
58.37 (43.70)c,d
80.00 (37.37)b *·d
55.83 (24.49) 'b,c
90.00 (22.76)" *
84.17 (23.84 )b *·d
68.93 (32.10)d.e
82.50 (23.53)c *••
62.57 (23.51)"·b,c
75.43 (20.27)" *
75.42 (19.46)b *
72.07 (25.35)
78.29 (20.74)C *
71.15 (25.28)
82.31 (24.93)
77.89 (21.23)
65.50 (33.11)
81.54 (25.20)
38.75 (29.40)"•b,c
63.75 (16.39)" *
62.08 (26.24)b *
57.00 (33.85)
61.85 (33.85)c *
8
12 Months (n = 19)
18 Months (n = 18)
24 Months (n = 18)
Data are means ± SD; scored such that the higher score represents the positive health attribute. Data analyzed using repeated measures analysis of variance. Means with the same superscript letter (a, b, c, d, or e) are significantly different at P .:5 0.05. * Significantly different from baseline.
one patient had glucose intolerance secondary to pulse prednisone. One patient reported anorexia with theophylline, which resolved after discontinuation of the drug. Discussion
This clinic was offered to poorly controlled minority asthmatics to determine whether a comprehensive long-term asthma management program that stressed the principles of the NIH Guidelines 1 •2 would improve outcomes. Because African-Americans have nearly three times the mortality rate from asthma than Caucasians, as well as a greater morbidity rate, 1 this population was targeted in our study. Lang and Polansky 13 recently reported an analysis of asthma deaths in Philadelphia, and found that black women were at increased risk of mortality. Other minorities were also at greater risk of asthma mortality, and poverty was another risk factor. This study was conducted in Memphis where African-Americans constitute approximately 50% of the population. Our patients were primarily working, middle-class individuals. Mayo and associates4 demonstrated that aggressive, state-of-the-art therapy combined with a concerted effort at patient education and easy access to clinicians reduced hospital readmission rates and duration of hospital stay in a group made up primarily of Hispanics. Our results reinforce the findings of Mayo et al 4 and several other groups of investigators 14- 18 that state-ofthe-art drug therapy and patient education improve outcomes in minority asthmatics. The effectiveness of our interventions was assessed in two ways-patients as their own control subjects and comparison with a THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
contemporary and comparable nonintervention control group. Despite very careful state-of-the-art management, including application of the principles of the NIH Guidelines, one of our patients with a history of severe asthma died from a sudden attack 18 months into the study. Asthmatics with a history ofiCU admission and intubation have a high mortality rate. 19- 22 Marquette et al 19 reported a posthospitalization mortality rate of 10.1% at 1 year and 14.4% at 3 years after near fatal asthma and mechanical ventilation. Unfortunately, these deaths are often preceded only by rapid, severe airflow obstruction. 23 •24 In an analysis of 80 asthma deaths from 1988-91, Campbell et al 21 found a mean age of 52 years, with 61% female and 32% with sudden collapse. Mayo et al 4 refer to "blitz" asthma characterized by severe respiratory failure within Yz hour of the onset of attack. Such attacks occurred in 4 of their special clinic patients during 32 months of follow-up. Ea(!h of these patients required ICU admission with intubation. Our patient apparently suffered such an attack, which occurred 2 years after her ICU admission with intubation. She successfully used her emergency supply of prednisone on several prior occasions, in conjunction with peak flow monitoring and use of inhaled albuterol while contacting the investigators. We were unable to determine which aspect of the clinic was primarily responsible for the improvement in outcome in the majority of our patients. One important aspect was the "partnership" described in the NIH Guidelines. 1 We placed great emphasis on listening actively, showing real concern for the patients, and involving the patients in therapeutic decisions (eg,
277
Management Program for Asthma
Table 6. Asthma Bother Profile Scores at Baseline, 6, 12, 18, and 24 Months
Item How much does your asthma bother . you at your paid work? (F(4,52) = 2.55, P = .05) you when you do jobs around the house? (F(4,56) = 3.69, P = .01) your social life? (F(4,56) = 3.45, P = .01) your personal life? (F(4,56) = 4.92, P = .00) your involvement in leisure activities? (F(4,60) = 4.80, P = .00) you when you sleep? (F(4,60) = 5.45, P = .00) How much . . . does the cost of your asthma medicines bother you? (F(4,48) = 6.70, P = .00) does the inconvenience or embarrassment of taking your asthma medicines bother you? (F(4,56) = 1.94, P = .11) How much . . . bother is the worry that you will have an asthma attack when visiting a new place? (F(4,56) = 2.79, P = .04) bother is the worry that you will catch a cold? (F(4,56) = 3.18, P = .02) bother is the worry that you will let others down? (F(4,56) = 4.08, P = .01) bother is the worry that your health may get worse in the future? (F(4,56) = 4.03, P = .01) bother is the worry that you won't be able to cope with an asthma attack? (F(4,56) = 4.92, P = .00) do coughs and colds bother you? (F(4,56) = 3.75, P = .01) are you bothered by feeling anxious, depressed, tired, or helpless? (F(4,52) = 2.76, P = .04)
Baseline (n = 20)
6 Months (n = 20)
12 Months (n = 19)
18 Months (n = 18)
24 Months (n = 18)
4.21 (1.25)"•b,c
2.57 (1.34) 8
*
2. 71 (1.38)b
*
3.21 (2 .12)
3.14 (2 .14)' *
3.67 (l.ll) 8 'b,c
2.67 (1.23)"
*
2.67 (1.29)b
*
3.07 (1.34)
2.67 (1.29)'
*
3.60 (1.68)"•b,c
2.20 (1.08) 8
*
2.60 (1.24)b
*
2.80 (1.27)
2.53 (1.64)'
*
3.53 (1.81) 8 'b,c,d
1.87 (0.92) 8
*
2.48 (1.36)b
*
2.40 ( 1.45 )'
*
2.40 (1.64)d
*
4.00 (1.55)"•b,c,d
2 .56 (1.46) 8
*
2. 75 (1.39)b
*
3.00 (1.41)'
*
2.69 (1.62)d
*
3.75 (1.57)"·b,c,d
2.19 (1.17)
8
*
2.63 (1.46)b
*
2.44 (1.15)'
*
2.50 (1.46)d.
5.00 (1.08)"•b,c,d
2.85 (1.82)"
*
2.85 (2.04)b
*
2.85 ( 1.99 )'
*
2.92 (2.10)d.
2.73 (1.62)
1.93 (1.49)
4.07 (1.83) 8
2.47 (1.13) 8
1.67 (1.18)
1.73 (1.39)
2.27 (2.31)
*
3.07 (1.62)
3.07 (1.49)
2.93 (1.83)
4.20 (1.94) 8
2.80 ( 1.4 7)" •
3.27 (1.67)
3.27 (1.49)
3.07 (1.53)
3.93 ( 1.67) 8 'b,c,d
2.80 (1. 70)"
*
3.00 (1.77)b
4.60 (1.45) 8 'b,c,d
3.27 (1.53) 8
*
3.60 ( 1. 72)b *·'
4.60 (1.35)"·b,c,d
3.13 (1.51) 8
*
3.33 (1.50)b
4.33 (1.50)"·b,c,d
3.40 (1.68) 8
*
4.36 (1.45)"•b,c,d
3.00 (1.71) 8
*
*
2.53 (1.60)'
*
2.67 (1.50)d *
3.13 (1. 73)'
*
3.00 (1.89)d
*·'
*
3.20 (1.32)'
*
2.87 (1.73)d
*
3.13 (1.51)b
*
3.53 (1.25)'
*
3.33 (1.50)d
*
3.29 (2.05)b
*
3.21 (1.67)'
*·'
2.93 (2.02)d *·'
Data are means ± SD; rated on a six point scale where a lower number indicates less bother. Data analyzed using repeated measures analysis of variance. Means with the same superscript letter (a, b, c, d, or e) are significantly different at P :5 .05.
* Significantly different from baseline.
change in doses or initiation of prednisone). Patients had short waiting times for clinic visits, and were given written instructions at each visit and easy telephone access to us for follow-up questions. Mellins et al 25 state that these aspects of care along with meeting the patient's expectations and patient's satisfaction are associated with improved compliance. Virtually all of our patients commented that "no one has ever taken the time to explain asthma and its management like you have." Several patients were excited to tell us that they
278
could exercise (eg, brisk walking and softball) again, and the difference it made in their quality of life. This asthma clinic and study was initiated and managed by doctors of pharmacy with a pulmonologist. In other studies, it was shown that pharmacists working in clinics as a team with physicians improve outcomes in asthmatics. 17 •18•26 Pauley et al 17 applied the NIH Guidelines in minority patients and dramatically reduced ED visits by 87%. Whereas our current report involved primarily minorities who were employed, we June 1996 Volume 311 Number6
Kelso et al
Table 7. Asthma Sleep Scale Scores at Baseline, 6, 12, 18 and 24 Months
Item Sleep loss (F(4,60) = 7.49, P = .00) Work/school interference (F(4,60) = 7.44, P = .00) Nights awakened (F(4,60) = 5.70, P = .01)
Baseline (n = 20)
6 Months (n = 20)
12 Months (n = 20)
18 Months (n = 18)
3.50 (1.37)"•b,c,d
1. 75 (1.07)" *
1.88 (1.26)b *
2.13 (1.26)• *
2.13 (1.36)d *
3.60 (1.45)"•b,c
1.94 (1.48)" *
1.94 (1. 33)b *
2.09 (1.20)• *
2.41 (1.70)
4.38 (1.75)"·b,c,d
1.75 (1.07)" •.•
2.50 (2.07)b *
2.69 (1.82)• • .•
2.44 ( 1.87)d •
24 Months (n = 18)
Data are means ± SD; a higher rating indicates a more negative effect on sleep. Data analyzed using repeated measures analysis of variance. M eans with the same superscript letter (a, b, c, d, or e) are significantly different at P s .05. • Significantly different from baseline.
previously reported a 41% reduction in ED visits in a group of indigent minorities.18 An asthma clinic in Kaiser Permanente in California reduced ED visits by 66% and hospitalizations by 48%.26 An obvious benefit of close teamwork between clinical pharmacy and medicine is that once there is agreement to apply the NIH Guidelines, the pharmacist can take adequate time to educate the patient and help in self-management. The busy physician has more time for other patients, yet asthma patients have enhanced outcomes. Obviously, for difficult patients or when there are concerns regarding concomitant medical problems, the protocol is set to have the physician spend more time with these patients. For routine visits when patients are doing well objectively and subjectively, the physician is freed to spend more time with other patients. As in our study and others previously cited, other groups of investigators also found that concerted efforts at patient education help improve outcomes in adult asthmatics. 27-29 Some of these reports involve highly structured educational programs. 27·28 The study group had significantly greater reductions in hospitalizations than the control group. Still, the control group had a reduction, and one potential explanation for this decrease in hospitalizations is that the prescription of inhaled corticosteroids increased from 39% (prestudy period) to 78% during the intervention period. Certainly, this is only a potential explanation, because we could not assess inhalation technique or level of patient education given the control subjects. It is reasonable to speculate that the increased use of inhaled corticosteroids in the control group resulted from gradual incorporation of the NIH Guidelines into the practice of these clinicians. A limitation of our study is lack of randomization. There is a possible selection bias in that our study patients responded to public service announcements and were perhaps more motivated and receptive to education and treatment compliance. In addition, via the informed consent process, study patients were aware of the purpose of the study. Therefore, one could argue that participation in the study itself might have been THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
responsible for some reduction in ED use. Our view is that we absolutely urged patients to seek emergency care based on protocol previously described, and we doubt that being in the study affected patient behavior of possibly not going to the ED. Another potential limitation would have been if the control group did not have easy access to medical care. From our review of medical records of the control group, there was every indication that each patient had easy access to primary care physicians or subspecialists. Another variable to consider is that whereas we believe the two groups are comparable (Table 1), we were not able to assess salary ranges, level of education, and living conditions in the control group. Although a full discussion of our quality of life data is beyond the scope of this article, a brief comment is appropriate. For each of the three validated instruments,8-11 a majority of quality of life parameters showed significant improvements for both years of the intervention period. For example, note in Tables 5-7 improved health perception, less physical limitations, fewer nocturnal awakenings, and less bother with asthma at home, work, and in social life. These types of improvements along with decreased ED visits obviously have a highly significant impact on quality of life. Although we did not assess lost work or school days before the study intervention, some quality of life measures (eg, asthma bother at work) imply an improvement in these parameters. In addition, patients with dramatic reductions in need for acute care would be expected to have reduction in lost work or school days. Cost implications of our results are noteworthy in that $1.86 billion (in 1990 dollars) are estimated to be spent in the United States for ED visits and hospitalizations. An initial investment in time educating the patient, planning therapy, and establishing a partnership with the patient can pay very large dividends in outcomes. Our results combined with similar results of other investigators suggest the cost of institutional acute care for asthma can be reduced dramatically when state-of-the-art, long-term management is applied.
279
Management Program for Asthma
Since the initiation of our study, recent doubleblinded trials demonstrated that addition of salmeterol to low-to-moderate doses of inhaled corticosteroids provides better outcomes than further increases in inhaled antiinflammatory therapy. 30•31 If other long-term trials validate this research, that is an appealing strategy to pursue. The International Consensus Report 2 already includes this approach as an option (before in~ creasing inhaled corticosteroid dose beyond 1,000 ,.,g daily). This study extends the findings of other investigators that a comprehensive long-term asthma management program that emphasizes the principles of the NIH Guidelines can improve outcomes in adult minorities. Clearly, specific program elements in state-of-the-art, long-term asthma management are interdependent. For example, individualized inhaled corticosteroid therapy must be accompanied by repetitive patient education regarding the purpose of the medication and correct inhalation technique. Further studies are recommended in larger populations. Acknowledgments We thank Elizabeth Tolley, Ph.D. for her assistance in data analysis. References 1. Expert Panel Report. Guidelines for the diagnosis and management of asthma. National Institutes of Health. Publication No. 91-3042, 1991. 2. International Report. International consensus report on diagnosis and treatment of asthma. National Institutes of Health. Publication No. 92-3091 , 1992. 3. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. N Eng! J Med. 1992;326:862-6. 4. Mayo PH, Richman J, Harris HW. Results of a program to reduce admissions for adult asthma. Ann Intern Med. 1990;112: 864-71. 5. Newman SP, Pavia D, Clarke SW. How should a pressurized beta adrenergic bronchodilator be inhaled? Eur J Respir Dis. 1981;62:3-21. 6. Self TH, Rumbak MJ, Kelso TM. Correct use of metered dose inhalers and spacers. Postgrad Med. 1992;92:95-106. 7. Brunette MG, Lands L, Thibodeau L. Childhood asthma: Prevention of attacks with short-term corticosteroid treatment of upper respiratory tract infection. Pediatrics. 1988;81:624-9. 8. Bousquet J, Knanl J, Dhlvert H, Richard A, Chicove A, Ware JE, Jr, et al. Quality of life in asthma: internal consistency and validity of the SF -36 questionnaire. Am J Respir Crit Care Med. 1994;149:371- 5. 9. Ware JE, Sherbourne CD. The MOS 36-item short form health surve1r (SF-36) . Med Care. 1992;30:473-83. 10. Hyland ME, Ley A, Fisher DW, Woodward V. Measurement of psychological distress in asthma and asthma management programs. Br J Clin Psycho!. 1995;34:601-11. 11. Boyer G, Aaronson D, Meltzer E, Laforce C, Grossman J, Yancey S. Enhancement of a general quality of life scale: Validation of a sleep scale (abstract). J Allergy Clin Immunol. 1992;89:186.
280
12. Rumbak MJ, Kelso TM, Arheart KL, Self TH. Perception of anxiety as a contributing factor of asthma: Indigent versus nonindigent. J Asthma. 1993;30:165-9. 13. Lang DM, Polansky M. Patterns of asthma mortality in Philadelphia from 1969 to 1991. N Eng! J Med. 1994;31:1542-6. 14. Bolton MD, Tilley BA, Kuder J, Reeves T, Schultz LR. The cost and effectiveness of an education program for adults who have asthma. J Gen Intern Med. 1991;6:401-7. 15. Clark NM, Feldman CH, Evans, Levinson MJ, Wasilewski Y, Mellins RB. The impact of health education on frequency and cost of health use by low income children with asthma. J Allergy Clin Immunol. 1986;78:108-15. 16. Perla EN. A program to reduce emergency room utilization and hospitalizations for inner-city asthmatics. First National Conference on Asthma Management. Arlington, VA., 1992 (abstract). 17. Pauley TR, Magee MJ, Cury JD. Results of a pharmacy managed, physician directed program to reduce emergency department visits in a group of inner city adult asthma pat ients. Ann Pharmacother. 1995;29:5-9. 18. Kelso TM, Self TH, Rumbak MJ, Stephens MA, Garrett W, Arheart KL. Educational and long term therapeutic intervention in the emergency department: Effect on outcomes in adult indigent minority asthmatics. Am J Emerg Med. 1995;13: 632-7. 19. Marquette CH, Saulvier F, Leroy 0, Wallbert B, Chopin C, Demarq JM, et al. Long term prognosis of near fatal asthma. Am Rev Respir Dis. 1992;146:76-81. 20. Richards GN, Kolbe J, Fenwick J, Rea HH. Demographic characteristics of patients with severe life threatening asthma: Comparison with asthma deaths. Thorax. 1993;48:1105-9. 21. Campbell DA, McLennan G, Coates JR, Frith PA, Gluyas P A, Latimer KM, et al. A comparison of asthma deaths and near fatal asthma attacks in South Australia. Eur Respir J . 1994;7:490-7. 22. Westerman DE, Benatar SR, Potgieter PD, Ferguson AD. Identification of the high-risk asthmatic patient. Am J Med. 1979;66:565- 72. 23. Mackonald JB , Seaton A, Williams DA. Asthma deaths in Cardiff, 1963-74:90 deaths outside hospital. BMJ. 1976;2:7213. 24. Bateman JRM, Clarke SW. Sudden death in asthma. Thorax. 1979;34:40-4. 25. Mellins RB, Evans D, Zimmerman B, Clark NM. Patient compliance: Are we wasting our time and don't know it? Am Rev Respir Dis. 1992;146;1376-7. 26. lm J. Evaluation of the effectiveness of an asthma clinic managed by an ambulatory care pharmacist. California Journal of Hospital Pharmacy. 1993;5:5-6. 27. Wilson SR, Scamagas P, German DF, Hughes GH, Lulla S, Coss S, et al. A controlled trial of two forms of self-management education for adults with asthma. Am J Med. 1993;94: 564-75. 28. Muhlhauser Richter B, Kraut D, Weske G, Worth H, Berger M. Evaluation of a structured treatment and teaching program on asthma. J Intern Med. 1991;230:157- 64. 29. Zieger RS, Heller, Mellon S, Wald J, Falkoff R, Schatz M. Facilitated referral to asthma specialist reduces relapses in asthma emergency room visits. J Allergy Clin lmmunol. 1991;87: 1160- 8. 30. Greening AP, Ind PW, Northfield M Shaw G. Added salmeterol versus higher dose corticosteroid in asthma patients with symptoms on existing inhaled corticosteroid. Lancet. 1994;344: 219-24. 31. Woolcock A, Lundback B, Ringdal OLN, Jacques LA. Comparison of the effect of addition of salmeterol with doubling the inhaled steroid dose in asthmatic patients (abstract). Am J Resp Crit Care Med. 1994;149:A280.
June 1996 Volume 311 Number 6