Asthma
Disease Management Program
Colorado
Region
Introduction
The Asthma Disease Management Program of the Kaiser Permanente Colorado
(KPCO) Denver/Boulder Local Market began development in August 1995.
Its pilot project was implemented in February 1997 with full implementation
in August 1998. A very large team of participants contributed to the
initial and ongoing success of this regionwide initiative. Table
1 identifies the complete list of project supporters and contributors.
Asthma
is recognized as a chronic inflammatory disorder of the airways with
symptomatic episodes that range from mild and intermittent to severe
and persistent. Although the reasons are poorly understood, the prevalence
of asthma has increased significantly over the past 20 years in both
children and adults.1 Undertreatment and inappropriate pharmaceutical
therapy have been shown to be major contributors to asthma morbidity
and mortality.2-4 In 1995, the prevalence of asthma within
KPCO was estimated to be 4% to 6% (13,600 to 20,400 asthmatic members).
Because of our large population of asthmatic patients and the potential
for relatively rapid improvement, asthma was one of the diseases chosen
for development of a disease management program.5,6
Background
The development of chronic disease management programs began in 1995
with an assessment of the member population which identified that KPCO's
model of care could be improved to meet the needs of chronically ill
patients. Our traditional ambulatory care model was designed to manage
acute, episodic exacerbations of asthma. The following weaknesses were
identified: 1) inconsistent approaches across the region to manage chronically
ill patients; 2) inconsistent messages to these patients; and 3) no
tools to identify patients receiving inadequate care. Epidemiologic
data, market climate, and utilization and cost data describing KPCO's
asthma population established the need for a redesign process. Employer
groups such as PepsiCo, Coors Brewing Company, and Time/Warner Cable
were asking about our approach to chronic disease management. Likewise,
regulatory and consumer reporting agencies such as the National Committee
for Quality Assurance (NCQA) and the Health Plan Employers Data Information
Set (HEDIS) had established population-based standards for chronic diseases.7
Process
The Asthma Disease Management Program was implemented as a pilot project
in February 1997 focusing on registry development, guideline dissemination,
limited asthma classes, and nursing care management. Early in 1998,
the registry was refined, care managers began limiting their intervention
to high-risk patients under the age of 50 years, and physician/staff
education was instituted. In August, four additional asthma care managers
were hired, and a "real-time" notification process for patient
hospital and emergency department encounters was introduced. Additionally,
the Care Management Institute (CMI) began supporting local disease management
initiatives.
Objectives
The primary objectives of the Asthma Disease Management Program are
to 1) improve quality outcomes of care, 2) equip patients to better
self-manage their chronic illness, 3) reduce costs of care through avoidance
of acute episodes and complications, and 4) increase patient and physician
satisfaction with continuity of care. The purpose of the initial pilot
project evaluation was to compare the impact of the Asthma Disease Management
Program model with "care as usual" regarding quality outcomes,
utilization patterns with related costs, and patient and physician satisfaction.
Subsequent to the pilot program evaluation, additional analyses were
performed to determine if improvements attributable to the Asthma Disease
Management Program could be sustained over time.
Methodology
Scope
In 1995, analysis of KPCO's total population of 340,000 members revealed
that 34.5% had one or more chronic diseases, an amount that accounted
for 60% of primary care physician utilization. By initiation of the
pilot project in February 1997, the baseline number of asthmatic patients
was 19,784 regionwide. Twenty-two percent of our total asthma population
(4708 patients) received their primary care at two of our medical offices,
Westminster and Aurora Centrepoint, which were chosen as the pilot intervention
sites.
Interventions
The Asthma Disease Management Program is a new model of care that addresses
the needs of asthma patients at a population level. The key program
elements described represent the primary changes from "care as
usual." Of particular note is the use of care managers--registered
nurses--who are responsible for monitoring the population of asthma
patients in a specified geographic area, targeting high-risk patients
for intervention, and providing one-to-one and group care. The primary
care physician retains accountability for directing and managing the
patient's care. Asthma care managers extend physicians' ability to monitor
and educate their patients. Key program elements include processes for
registry and reporting, "real-time" notification, care manager
intervention, and KP Regionwide education.
Registry
and Report Process
The Asthma Disease Identification, Registry, and Reporting System is
a set of software programs developed by clinicians, programmers, and
analysts at KPCO.8 This system extracts and compiles data
from clinical and administrative sources to produce a continuously updated
registry of patients with asthma. This system also generates monthly
and quarterly reports of patient levels in electronic and printed formats.
These reports contain clinically relevant data of interest to physicians
and care managers. Patients are risk-stratified by pharmaceutical dispensing
and by Emergency Department (ED), observation/clinical decision unit
(OBS/CDU), and inpatient utilization.
"Real-Time"
Notification Process
Information about hospitalizations and ED encounters for asthma is received
and entered into an Access database every day. The appropriate care
manager is notified to contact patients shortly after discharge, when
they are most likely to make behavior changes to better manage their
asthma.
Care
Manager Intervention
The care managers identify patients for proactive outreach using a combination
of the panel report, real-time feedback from hospitals and EDs, and
physician referral. The goal of intervention is to empower patients
to learn self-management skills and to make lifestyle changes to decrease
asthma morbidity. The care manager follows patients enrolled in the
program by in-person and telephone contact during a two- to three-month
period. When appropriate, patients are referred to classes for additional
asthma education. Upon discharge from active care management, patients
resume usual care with their primary care physician but may reenter
the program if they again meet high-risk criteria. Population monitoring
continues for all patients in the Asthma Registry. Major functions and
accountabilities of care managers include:
- Assessment,
treatment modification, and patient education based on clinical guidelines;
- Addressing
patients' psychological responses to living with a chronic disease
using behavioral change strategies;
- Providing
telephone follow-up, trouble-shooting, monitoring, and coaching to
ensure success of the patient's self-management plan;
- Coordinating
complex care of patients and planning their transition back to primary
care as their condition permits;
- Reviewing
and interpreting panel reports and communicating with physicians about
the status of their asthma patients.
Regionwide
Patient and Staff Education
Asthma management classes for patients (and family members) are offered
monthly at all facilities. Asthma skills training sessions are offered
to the staff to enhance their ability to conduct pulmonary function
tests and to provide education on medication and equipment use, peak
flow monitoring, environmental controls, and asthma pathophysiology.
In addition, guidelines for asthma care, which are based on the National
Institutes of Health (NIH) recommended criteria for diagnosis and treatment,
were widely disseminated to physicians and mid-level providers across
the KPCO region.9
Subjects
and Setting
In the initial pilot study, conducted from February 1 to July 31, 1997,
the Asthma Disease Management Program model was compared with "care
as usual." Three hundred seventy-eight patients from a population
of 19,784 members were enrolled in care management during the pilot
period. These patients comprised the pilot intervention group. Patients
were eligible for enrollment if they received primary care services
at one of the two pilot sites and either were identified via registry
report as being at high risk or were referred to the program by their
primary care physician. High-risk asthma was defined as: 1) overuse
of beta2-agonists, 2) high-dose beta2-agonist use with no dispensed
inhaled steroid, or 3) a recent asthma-related hospital or ED admission.
The pilot project was conducted at two medical offices: Westminster
and Aurora Centrepoint. An additional 976 intervention patients were
enrolled into the program between August 1997 and January 1999. Patients
not in the intervention groups received care as usual. Two asthma care
managers (1.6 FTE) were hired to participate in the pilot project. Four
care managers and 13 additional offices were added in August 1999.
Study
Design
A quasi-experimental pretest-posttest study design was used to examine
changes in process of care and clinical outcomes measures for asthma
intervention patients. For some outcomes, multiple pre- or postintervention
measurements were performed. Changes within a comparison group (composed
of Asthma Registry members who had not been seen by the care managers)
were also assessed on similar outcome measures. A pretest-posttest design
was also used to evaluate changes in patient and physician satisfaction.
Although planned for the pilot evaluation, use of a more rigorous randomized
case-control design was not feasible because of pilot implementation
issues. For example, during the pilot project, a large number (48%)
of all intervention patients were referred to asthma care managers by
their primary care physicians. Given that provider acceptance of the
asthma care managers was an important key to successful implementation
of the Asthma Care Management Program, referred patients were accepted
into the Program at the expense of a more rigorous study design. Also,
a great amount of refinement of the computer programs used for patient
identification and risk stratification took place during the pilot period.
Statistical methods used in analyses of pilot and postpilot data included
parametric and nonparametric within-group tests for significance (eg,
chi-square, t, and Wilcoxon signed rank tests) as appropriate.
Patient
Intervention Procedure
The asthma care managers contacted patients by telephone and screened
them for eligibility. Patients were enrolled if they met entry criteria
and were willing to participate. A one-to-one initial visit was scheduled.
At the initial visit, the care manager recorded a detailed history,
performed pulmonary function tests, and provided education about asthma
and its treatment. Small steps toward behavior change were negotiated
with the patient according to the patient's readiness to change, lifestyle,
and areas (s)he was willing to address. Patients were given a written
home care plan if their asthma was stable. Patients experiencing an
asthma flare were appropriately treated and were scheduled for a second
visit. Subsequent patient contacts occurred by telephone a mean of three
times during a three-month period. All face-to-face and phone contacts
were recorded on coding sheets designed for the pilot project.
Measures
The project addressed the following outcomes, utilization, and process
of care measures: patient and physician satisfaction, use of ambulatory
and hospital-based health care services, overuse of beta2-agonist medication,
usage of prescribed anti-inflammatory medication, dispensing of peak
flow meters, prevalence of spirometry testing, and the provision of
a home care plan.
Satisfaction
Data
Pre- and postpilot patient satisfaction surveys were administered to
258 intervention and nonintervention asthma patients. Pre- and postpilot
satisfaction surveys were administered to 71 physicians at pilot and
nonpilot medical offices. Survey questions used a five-point Likert
scale response, where "1" defined the negative pole and "5"
defined the positive pole. Patient satisfaction data were collected
from a random sample of nonintervention patients and intervention patients
via telephone survey. Survey results were analyzed for patients who
responded to both the baseline and postpilot surveys. Questions addressed
patient satisfaction with education, home care planning, asthma management,
and continuity of care. A written survey was administered in December
1996 and again in August 1997 to a random sample of physicians in primary
care departments at pilot and nonpilot sites. Questions addressed satisfaction
related to provision and monitoring of asthma patient's care, meeting
patient expectations, and availability of resources to manage asthma
patients. Within-group and between-group changes in satisfaction were
measured using Wilcoxon signed rank tests and Wilcoxon-Mann-Whitney
U tests.
In March
of 1998 and 1999, the CMI-sponsored administration of a survey to adults
who were identified as having asthma. Eighty percent of patients who
responded to the 1998 survey also responded to the follow-up 1999 survey,
for a total of 1225 patients. Data from a subsample of these respondents
were published in the 1998 CMI Asthma Outcomes Report.10
Three of the 64 survey questions address issues of patient satisfaction.
Paired t tests and the Wilcoxon signed rank test were used to assess
changes between the two time periods.
Utilization
Data
Administrative databases were used to collect pharmacy, outpatient encounter,
and hospital (inpatient, ED, and observation/clinical decision unit
(OBS/CDU)) data for all patients identified with asthma. To control
for seasonality and the effect of secular trends in both clinical and
data systems, utilization data were extracted for several time periods
before and after the pilot.
Outpatient
utilization data were obtained from claims and encounter databases.
Using paired t tests, the rate of asthma-related ambulatory care visits
to primary care departments was measured for several periods before
and after the initial encounter with a care manager for intervention
patients enrolled during the pilot and for a similar time period for
nonintervention patients. During the postpilot phase, changes in ambulatory
care utilization was measured at the population level for patients identified
in the registry in both March 1998 and March 1999. Patients were stratified
into three age groups: 0-18 years (pediatric members), 19-49 years (adult
members), and 50+ years (older adult members). Paired t tests were used
to evaluate the change in mean number of visits per patient per year
to Primary Care Departments for asthma-related conditions and to the
Pulmonology and Allergy Departments.
Changes
in hospital-based utilization rates (inpatient, ED, and OBS/CDU) were
measured for a pilot-specific cohort of intervention patients and for
all asthma patients identified via the registry for several time periods.
Paired t tests and repeated measures ANOVA were used to evaluate differences
in mean number of inpatient admissions, emergency department visits,
and OBS/CDU visits per year for the initial pilot cohort of intervention
patients. Paired t tests were used to evaluate the change in mean number
of inpatient admissions, emergency department visits, and OBS/CDU visits
per patient per year. No statistical tests were performed on the likelihood
of a hospital event occurring.
Parametric
and nonparametric statistical tests were used to assess changes over
time for four aspects of pharmacy dispensing patterns for all asthma
patients: 1) number of beta2-agonists dispensed per patient in a six-month
period; 2) percentage of patients with beta2-agonist overuse; 3) percentage
of patients taking high-dose beta2-agonists with a dispensed inhaled
anti-inflammatory; and 4) percentage of asthma patients who had been
dispensed a peak flow meter. Many of these measures were stratified
by intervention status, by age category, or by both. In order to control
for unmeasured seasonal variation, the following four time periods were
examined for the pilot evaluation: one-year prepilot, baseline
(or prepilot), end-of-pilot, and six-month postpilot periods. Overuse
is defined here as 12 or more metered dose inhalers (MDI), or 180 or
more milliliters of nebulizer solution, or 2160 or more milliliters
of nebulizer premix solution of a beta2-agonist product dispensed in
a six-month period. A patient who is a high-dose beta2-agonist user
with an inhaled anti-inflammatory product is one who is dispensed six
or more MDIs or the equivalent of a beta2-agonist product in a six-month
period and also receives an inhaled anti-inflammatory product (eg, beclomethasone
dipropionate, budesonide, fluticasone, cromolyn sodium, etc.).
Process
of Care Measures
Dispensing of peak flow meters, administration of spirometry tests,
and development and documentation of patient home self-care plans are
all considered to be process-of-care measures in this evaluation. In
September 1997, registered nurses performed retrospective chart audits
on all pilot intervention patients and on a random sample of nonintervention
patients to document receipt of spirometry testing and existence of
a home care plan. Chi-square and Wilcoxon signed rank tests were used
to evaluate within-group and between-group changes in these two measures.
Data from the CMI-sponsored Survey of Adults with Asthma was used to
describe changes in population-based measures of peak flow meter dispensing
and receipt of home self-care instructions.7 The Wilcoxon
signed rank test was used to examine changes in the distribution of
responses between March 1998 and March 1999 for a sample of 1222 documented
asthma patients who responded in both time periods.
Results
Satisfaction
Seventeen physicians at pilot sites and 54 physicians at nonpilot sites
completed baseline and postpilot surveys. At baseline, no statistical
difference for any of the ten questions was noted between pilot and
nonpilot physicians responding. Using the Wilcoxon signed rank test,
significant improvement (p < 0.05) in three of the satisfaction scores
was found for the pilot site physicians. For the nonpilot site physicians,
one measure improved significantly (p < 0.05), and five measures
declined significantly. The magnitude of improvement was greater for
the pilot group of physicians and staff for all questions. The magnitude
of improvement was greater for the pilot group of providers for all
questions. Wilcoxon-Mann-Whitney U tests showed that the mean differences
for six of the ten questions were significantly greater (p < 0.05)
for the pilot group.
Eighteen
intervention patients and 240 nonintervention patients responded to
a patient satisfaction survey. No differences in satisfaction were found
pre-pilot between the two groups. Postpilot satisfaction increased for
both groups. Although the magnitude of improvement was greater among
the intervention group for three of the five measures, no differences
were statistically significant. Two of the three surrogate measures
for patient satisfaction contained in the CMI-sponsored Adults with
Asthma Survey showed no significant change. The Likert scale response
to the question addressing ease of getting medical care for asthma when
needed did show a significant change (p < 0.05).
Ambulatory
and Hospital Utilization
Early pilot results found that for the six months before and after the
initial encounter with asthma care managers, ambulatory visits to primary
care departments decreased significantly (p < 0.05) for the pilot
intervention patients. Visits for nonintervention patients increased
slightly during the same time period, although the increase was not
statistically significant. The visit rate to the Allergy and Pulmonology
Departments increased slightly for pilot intervention patients for the
same time period.
Ambulatory
care utilization measures were assessed for a sample of 17,298 asthma
patients who were identified in the registry in both March 1998 and
March 1999. The rate of asthma-related visits per 1000 patients per
year to Primary Care Departments showed significant reductions (p <
0.001) for all age groups (Figure
1). For the age categories of patients targeted by the care managers,
Figure
2 shows that the visit rate to the Allergy Department decreased
significantly in the 0- to 18-year age group (p < 0.001) and in the
19- to 49-year age group (p < 0.05). Although a slight decrease in
visits for the > 50-year age group was observed, the difference was
not significant. The visit rate to the Pulmonology Department showed
no statistically significant change.
Figure
3 demonstrates the finding that both the annual hospital inpatient
and ED admission rate for 252 continuously enrolled pilot intervention
patients declined significantly from March 1997 to March 1999 (p <
0.05). OBS/CDU admission rates did not change. For the sample of 17,298
asthma patients on the registry in both March 1998 and March 1999, significant
reductions were found in the number of annual inpatient admissions in
the pediatric (p < 0.001) and adult (p < 0.05) age groups (Figure
4). OBS utilization showed no significant changes between the two
time periods. ED utilization increased significantly for pediatric patients
(p < 0.05) and for adults and older adults (p < 0.001). Although
reasons for the increase are unclear, the reduction in hospital utilization
may have resulted in a shift to ED visits.
Pharmacy
Utilization
Measures of beta2-agonist pharmacy dispensing patterns suggest that
the pilot intervention patients may have been "higher risk"
than the average asthma patient identified from the registry. Although
during the pilot project, no significant change occurred in the mean
number of beta2-agonist MDIs dispensed per patient, the baseline rate
of dispensing was much higher (7.7 versus 4.5) for the pilot intervention
patients compared with the nonintervention patients. Figure
5 shows a statistically significant (p < 0.05) decline in beta2-agonist
overuse among pilot intervention patients from the year preceding to
the year subsequent to the pilot. Figure
6 demonstrates that during this same time period, the percentage
of pilot intervention patients receiving high-dose beta2-agonist medications
who were also dispensed an inhaled steroid significantly increased (p
< 0.05). Follow-up analyses of these same measures (Figures
7 and 8),
evaluated for the entire sample of asthma registry patients and stratified
by age category, were performed using data collected two years subsequent
to the pilot. Statistically significant improvement in most age categories
for both measures was demonstrated (p < 0.05).
Process
of Care
The percentage of intervention patients with a pharmacy-dispensed peak
flow meter increased significantly from 42% before the pilot to 75%
after the pilot project (p < .05). No change occurred for the nonintervention
patients. Analysis of pharmacy dispensing of peak flow meters by product
code was also undertaken for a cohort of 17,298 asthma registry patients
in 1998 and 1999. The percentage of patients with a dispensed peak flow
meter increased significantly (p < 0.05) in the 0- to 19-year age
group (45.9% to 49.1%), but relatively no change was found for the other
age categories. Results from data derived from the CMI-sponsored Survey
of Adults with Asthma found a 4.5% increase (73% to 77.5%) in number
of asthma patients responding to the survey who reported owning a peak
flow meter (n = 1222). These data demonstrate a discrepancy between
patient reports of having a peak flow meter and pharmacy dispensing.
Reasons for the differences are unknown; however, pharmacy dispensing
data were available only from 1995 to the present, so we presume that
part of the variation is due to some patients having received a peak
flow meter prior to 1995 or having received it somewhere else besides
a KPCO pharmacy. Because the patients responding to the CMI survey were
not drawn from exactly the same sample as the patients for whom we collected
pharmacy dispensing records, additional statistical calculations could
not be made to further examine the source of the discrepancies.
Analyses
using chart audit data for the pilot intervention patients and for a
random sample of nonintervention patients found a statistically significant
increase (from 16% to 60.0%) in percentage of pilot intervention patients
with a documented home care plan (p < 0.05). For the sample of nonintervention
patients, the prevalence of home care plans increased from 21% to 26%,
but this change was not statistically significant. The percentage of
pilot intervention patients who had received spirometry testing also
experienced a statistically significant increase (4.6% to 58%). No change
was found for the sample of nonintervention patients. Although not statistically
significant, results also derived from the CMI-sponsored Survey of Adults
with Asthma found that more than 4% (62% to 66.7%) of the 1222 asthma
patients responding to the survey had received written directions about
how to take their asthma medication and what to do if they had a severe
attack.
Comment
Effect of Program on Direct Patient Care
The Asthma Disease Management Program's effect has been to extend the
physician's ability to 1) educate patients about their disease, 2) provide
appropriate treatment, and 3) monitor patient care over time. The registered
nurse's role is expanded to include monitoring and management of asthma
patients. Improved collaboration and communication between subspecialty
and primary care departments has resulted in improved asthma control
in this population.
What
Makes the Project Innovative?
The Asthma Disease Management Program is unique because it provides
a mechanism to identify patients who are at high risk for complications
and exacerbation and to implement proactive interventions to prevent
these occurrences. The patient's care plan is negotiated according to
the patient's readiness to change and lifestyle choices. The care managers
assure communication between key departments such as Allergy, Pulmonology,
Pharmacy, and Primary Care.
Has
the Program Led to Development of New or Improved Processes That Can
Be Considered "Best Practices"?
KPCO's Primary Care Quality Council targeted asthma as one of its primary
areas of focus for 1999. The Council adopted asthma quality measures
recommended by the Asthma Disease Management Program and is piloting
an asthma initiative in the Pediatric Departments. In 1998, all KP Divisions
collaborated to create the CMI to disseminate knowledge and assist with
local implementation of evidence-based clinical best practices throughout
all regions. Our established Asthma Disease Management Program has been
highlighted in interregional discussions and is being evaluated as a
best practice.
Has
the Project Resulted in Excellent Performance Compared with Other Programs
or Relevant Benchmarks?
In October 1998, the CMI published data comparing the results for care
processes and outcomes for all KP.10 For the period 8/1/96
through 7/31/97, which included our six-month pilot period, the KPCO
compared favorably with other KP Regions.
Benefits
of Multidisciplinary Team Involvement
The volume and focus of the Program's development required the skills
of clinical and nonclinical staff. The Program is not viewed as replacing
any part of patient care but is well integrated into the primary care
milieu. The Program's successful implementation occurred because the
viewpoints, concerns, and interests of the staff who would be affected
were represented. These individuals continue to be on-site champions
of the program.
Implications
and Conclusion
The pilot project resulted in implementation of asthma care management
for all clinics in the Denver/Boulder Local Market and the addition
of four asthma care managers. KP Kansas City recently insti
tuted an asthma care management program based on the successful results
of our Denver/Boulder Local Market project. The KPCO Colorado Springs
Local Market is working collaboratively with the Denver/Boulder Disease
Management Team to implement an asthma disease management program.
In conclusion,
evaluation of the Asthma Disease Management Program demonstrated improved
patient and physician satisfaction, improved treatment/care planning,
and improved medication utilization with gains sustained over a two-year
period. The evaluation thus demonstrated that systematic, population-based
asthma disease management can improve quality, satisfaction, and utilization
outcomes and can sustain these gains over time.
References
1. National Heart, Lung and Blood Institute, National Asthma Education
and Prevention Program. Global Initiative for Asthma: Global Strategy
for Asthma Management and Prevention NHLBI/WHO Workshop Report National
Institutes of Health Pub No .95-3659. Bethesda, MD, 1997.
2. Pappas G, Hadden WC, Kozak LJ, Fisher GF. Potentially avoidable hospitalizations:
inequalities in rates between US socioeconomic groups. Am J Public Health
1997;87:811-6.
3. Friday GA Jr, Khine H, Lin MS, Caliguiri LAA. Profile of children requiring
emergency treatment for asthma. Ann Allergy Asthma Immunol 1997;78:221-4.
4. Hartert TV, Windom HH, Peebles RS Jr, Freidhoff LR, Togias A. Inadequate
outpatient medical therapy for patients with asthma admitted to two urban
hospitals. Am J Med 1996;100:386-94.
5. van Essen-Zandvliet EE, Hughes MD, Waalkens HJ, Duiverman EJ, Pocock
SJ, Kerrebijn KF. Effects of 22 months of treatment with inhaled corticosteroids
and/or beta2-agonist on lung function, airway responsiveness, and symptoms
in children with asthma. The Dutch Chronic Non-Specific Lung Disease Study
Group. Am Rev Respir Dis 1992;146:547-54.
6. Solomon N, Steinbruegge J, Juhn P. White Paper: The Opportunity of
Disease Management at Kaiser Permanente, December 2, 1996, unpublished
position statement.
7. Bodily M. Disease Management System Specifications and Documentation.
Kaiser Permanente, 1999.
8. HEDIS 2000 Technical Specifications. National Committee for Quality
Assurance, Washington, DC, 1999, p. 98-100.
9. National Heart, Lung and Blood Institute, National Asthma Education
and Prevention Program. Expert Panel Report 2: Guidelines for the Diagnosis
and Management of Asthma. National Institutes of Health Pub No .97-4051.
Bethesda, MD, 1997.
10. Care Management Institute. 1998 Kaiser Permanente Asthma Outcomes
Report. October 1998.
