A quarterly publication for members of the Medical Outcomes Trust
October 1996

In This Issue . . .
Childhood Asthma

Asthma Outcomes and the Specialist

Initiatives:
Summaries of organizational efforts to evaluate outcomes of asthma care

Asthma Outcomes at Point-of-Service: Dean Clinic's "Asthma Report Card"

Resources:
A description of measures and methods to evaluate asthma and outcomes of asthma care

Research Summary: 1995-1996

Practicing for Patienthood: A Policy Opinion on the Health Care Consumer Agenda

Functional Impact and Utility Information: Decision Support for Treatment Options

Preparing for Risk Adjustment An Interview with Lisa Iezzoni, MD, MSc

Resources for Clinical Performance Measures:
CONQUEST and the National Library of Healthcare Indicators

Childhood Asthma
by David Aaron Bergman, MD, Vice President of Health Care Practices,
Lucile Packard Children's Hospital at Stanford, Palo Alto, California

Childhood asthma in the United States remains a significant public health problem. Asthma is responsible for over 28 million restricted activity days in American children and 2.2 million pediatrician visits annually,1 numbers that continue to increase despite important insights into the pathophysiology of asthma and the availability of improved chemotherapeutic agents. Between 1979 and 1987 asthma mortality rates in children increased 10.1% per year2 and hospitalizations for children with asthma increased 4.5% per year.3 In 1984 asthma was shown to account for 4.7% of all hospitalizations for American children.4 Children with asthma miss3 times as much school as nonasthmatics,5 and the overall cost of severe asthma in childhood has been estimated at $18,000 per year.6

The initiatives described in this issue's coverage of asthma demonstrate some of the key efforts being made to improve the quality of care for children with asthma. Central to these initiatives is the measurement of patient-centered outcomes as a means to assess, monitor, and improve care. Traditionally outcomes in asthma have been measured through pulmonary function tests. Using these tests as outcomes measures can be problematic because it is frequently difficult to establish a relationship between the intensity of the intervention (eg, dose and frequency of medication) and the physiological outcomes measured through pulmonary function tests. Moreover, the physiological outcomes are not stable over time and often don't reflect the long term impact of the illness.

Over the past decade there has been considerable work to document that measures of functional health status surpass traditional clinical measures in their ability to assess disease impact. This has spurred development of generic instruments, and to a lesser extent condition-specific instruments, to measure functional status. Functional status measures are still in their infancy, but as they are further developed and adopted, the special issues involved in measuring the functional status of children will need to receive increased attention. One of the leading efforts in measuring children's functional status has been the 'Health Survey For Asthma.'7 This functional health status instrument is the first measure from the Functional Outcomes Project (1992) component of the American Academy of Pediatrics' Quality Improvement initiative. Drafting, refinement, and pilot testing of the instrument have been completed. The final stage of development, extensive field testing and validation, is in progress.

Many of the initiatives described in this issue have discussed one of the important efforts to improve asthma care: the provision of early access to specialty care. Studies in both the inpatient and the office-based setting have demonstrated a significant gap between what is known to be optimal care for asthma and actual practice patterns.8,9 One effective way to close this gap is by providing early access to knowledgeable asthma specialists.10,11 However, another intervention not addressed in this issue deserves attention: case management. Several studies have shown that case management, when combined with close assessment and feedback, can dramatically reduce emergency department use, decrease hospitalizations, and decrease symptom burden. Close case management allows for a rapid turn around in the improvement cycle allowing for quick changes in the care plan.1,12

There are also two outcomes important to the care of asthma that warrant greater study: access to all types of appropriate specialty care (in addition to asthma specialty care) and patient satisfaction. To date there has been little effort to define measures that assess access to other types of specialty care, such as respiratory therapy, home care or mental health. In this world of managed care and gatekeepers, understanding the value of the spectrum of care becomes a particularly important endeavor. Patient satisfaction is also understudied in the area of asthma and is primarily anecdotal. In a study by Greineder et al. the author states "based on anecdotal reports, patient and family satisfaction was also clearly improved by the intervention (an asthma outreach program) although we did not study this formally."1 One of the few examples of systematic evaluation of patient satisfaction in children with asthma was in a study by Barnett et al. in which a general pediatrics satisfaction instrument was modified for asthma and revealed significantly high overall satisfaction with the asthma management intervention "Living with Asthma" proposed by the National Heart Lung and Blood Institute.13

It is clear from the initiatives described in this issue that much has been accomplished in the design and measurement of effective interventions to improve the care of patients with asthma-both for children and adults. The challenges for the future include the refinement and promulgation of condition-specific functional status instruments for children with asthma and the integration of measures of access to appropriate specialty care and patient satisfaction into a unified, easy-to-use tool.

References
1. Greineder DK, Loane KC, Parks P. Reduction in resource utilization by an asthma outreach program. Arch Pediatr Adolesc Med. 1995;149:415-420.
2. Gergen PJ, Weiss KB. Changing Patterns of asthma hospitalization among children:1979 to 1987. JAMA. 1990;264:1688-1692.
3. Weiss KB, Wagener DK. Changing patterns of asthma mortality. JAMA. 1990;264:1683-1687.
4. National Center for Health Statistics, Public Health Service, Department of Health and Human Services: Utilization for short stay hospitals; annual summary for United States: Washington DC, series 13, no.84, publication 86-1745 (1984).
5. Fowler MG, Davenport MG, Garg R. School functioning of US children with asthma. Pediatrics.1992;90:939-944.
6. O'Brien KP. Managed care and the treatment of asthma. Journal of Asthma. 1995;32:325-334
7. Sullivan SA. Children's Health Survey For Asthma. Personal Communication.
8. Alario AJ, Bergman DA et al. Managment of acute asthma in pediatric practice: Do pediatricians follow national guidelines. Presented at the Pediatric Academic Societies National Meeting, Washington, D.C. May, 1996
9. Bergman DA, Ouren J and Kurtin P. The impact of practice guidelines implementation on the health care outcomes of children admitted with asthma. Presented at the Pediatric Academic Societies National Meeting, Washington, D.C. May, 1996
10. Zeiger RS, Heller S, Mellon MH, et al. Facilitated referral to asthma specialist reduces relapses in asthma emergency room visits. J Allergy Clin Immunol. 1991;87:1160-8.
11. Freund DA, Stein J, Hurley R, et al. Specialty differences in the treatment of asthma. Am J Public Health. 1990;80:36-8.
12. Wissow LS, Warshow M, Box J, et al. Case management and quality assurance to improve care of inner-city children with asthma. AJDC. 1988;142:748-752.
13. Barnett TE, Fatis M, Sonnek D, et al. Treatment satisfaction with an asthma management program: a five year retrospective assessment. Journal of Asthma. 1992;29:109-116.

Asthma Outcomes and the Specialist
by Michael S. Blaiss, MD, Associate Professor of Pediatrics, Assistant Professor of Medicine,
Division of Clinical Immunology, University of Tennessee, Memphis

Diseases that affect a large portion of the population, for which treatment is well-known to be effective, and that have a significant cost on society are perfect candidates for outcomes research. Asthma is a prime example. Over 14 million Americans have asthma, which is estimated to affect 4-5% of adults and 7% of children.1-3 Hospital- ization rates and death from asthma have been increasing in the United States and throughout the world. It is a major cause for hospitalization at children's hospitals, accounting for over 198,000 admissions in 1993.3 Asthma is an expensive disease which accounted for about 1% of all health expenditures in the United States in 1990.1 Asthma costs are significant internationally. In 1990 U.S. dollars, the cost for each asthmatic per year in Sweden is $1,315, in the United Kingdom it is $1,043, and in Australia it is $769.4 A recent report estimated the direct and indirect cost for Canadians receiving inpatient or outpatient care for asthma during 1990 to be between $504 and $648 million Canadian dollars.5 As the cost of asthma continues to rise, as treatment effectiveness improves, and as patients and the payers of health care become more involved in medical decisions, the need for outcomes data in documenting what works best in asthmatic patients and which providers can produce the best overall outcome in these patients becomes imperative.

A growing debate in health care in the United States is over the role of the specialist in management of patients with chronic disease. With the increased emphasis on primary care and the use of gatekeepers, specialty referral for many conditions is decreasing. Will this trend lead to the improved or worsened outcomes in asthmatic patients? Recent outcomes studies have demonstrated that the specialist caring for the moderate to severe asthmatic produces good clinical results which are truly cost-effective and improve the quality of life of the patient.6-10
A retrospective analysis of outcomes in 455 patients with similar asthma severity in an HMO in the Midwest was performed.11 Allergists treated 54% of the asthmatics and 46% were cared for by generalists. There were fewer missed days of school/work and improved activity in allergists' patients. Also the allergists' patients were more satisfied with their physician and asthma care in general and reported better emotional functioning between attacks.

Three hundred and nine asthmatics between the ages of 6-59 were prospectively assigned to an allergist or generalist after presenting to the San Diego Kaiser Emergency Department and evaluated over a six-month period.12 The allergists' patients had a 75% reduction in the number of, and percent of, asthma awakenings per night (p <0.0001) and an almost 50% reduction in asthma emergency department relapses (p <0.017) resulting from a reduction in the frequency of multiple relapses.

Specialists must work with their primary care colleagues and other health professionals to insure that every patient with asthma gets the best care possible. Obviously, the mild asthmatic would not benefit from specialty care. It is important to determine which asthmatic population is better managed with input by the specialist. Hopefully with the development of standardized outcomes measures in asthma, we will have valuable data in determining where value is added if the primary care physician consults with the specialist in the care of the asthmatic.

To this aim, the Joint Council of Allergy, Asthma, and Immunology, the socioeconomic arm of the Board-Certified Allergists/ Immunologists in the United States, has commissioned The Health Institute, New England Medical Center to develop a set of measurement instruments to monitor clinical, humanistic, and economic outcomes in moderate to severe adult and pediatric asthmatics in the offices of Board-Certified Allergists/Immunologists. It is hoped that this program will not only standardize outcomes measures for asthma throughout the United States, but also contribute valuable information on how to coordinate care between the primary care physician and the specialist. This will lead to the ultimate goal of providing the best care for the patient with asthma.

References
1. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. N Engl J Med 1992;326(13):862-6.
2. Sheet DF. Asthma Statistics. National Heart, Lung and Blood Institute, May 1992.
3. Center for Disease Control. Asthma Mortality and Hospitalization Among Children and Young Adults- United States, 1980-1993. MMWR 1996;45:350-353.
4. National Heart L, and Blood Institute, National Institutes of Health. Socioeconomics. In: 94-3276 NPN, ed. Global Strategy for Asthma Management and Prevention: A NHLBI/WHO Workshop Report. Bethesda, MD: U.S. Department of Health and Human Services, 1994.
5. Krahn MD, Berka C, Langlois P. Direct and indirect costs of asthma in Canada. Can Med Assoc J 1996; 54: 821-831.
6. Freund DA, Stein J, Hurley Rea. Specialty differences in the treatment of asthma. J Allergy Clinical Immunol 1989;84:401-406.
7. Mahr TA, Evans Rd. Allergist influence on asthma care. Ann Allergy 1993;71(2):115-20.
8. Bucknall CE, Moran F, Robertson C, Stevenson RD. Differences in Hospital Asthma Management. Lancet 1988;2:748-750.
9. Korenblat PE, Korenblat-Hanin M, Gaioni SJ. An Asthma Center: Outcomes Validation. J Allergy Clin Immunol 1995;95:222.
10. Mayo PH, Richman J, Harris HW. Results of a program to reduce admissions for adult asthma. Ann Intern Med 1990;112:864-871.
11. Fowles J, Schoenwetter W, Pheley Aea. Measuring the severity and outcomes of asthma care by generalists and allergists. J Allergy Clin Immunolo 1990;85:195.
12. Zeiger RS, Heller S, Mellon MH, Wald J, Falkoff R, Schatz M. Facilitated referral to asthma specialist reduces relapses in asthma emergency room visits. J Allergy Clin Immunol 1991;87(6):1160-8.

Initiatives
Summaries of organizational efforts to evaluate outcomes of asthma care

Multi-site Effectiveness Evaluations

AMGA (American Medical Group Association) Asthma Consortium

Project Goal:
Provide study support for medical groups interested in outcomes studies.
Description:
Participating medical groups collect standardized information and submit it to the AMGA. Instruments used include the Health Status Questionnaire, the Asthma TyPE, a Physician Questionnaire, and personal characteristics, including questions about emergency room visits, medications used, and use of peak flow meters. Questionnaires are administered at baseline and at 6 months, in the office or by mail.
Consortium findings include: 31% of patients do not feel they have been told everything they needed to know to manage a severe flare-up and 26% of patients who reported visiting the emergency room in the past six months were classified (by a patient self-assessment questionnaire) as "mild." Further analysis will focus on the unique characteristics of the "mild" ER users.
Status:
Fifteen medical groups currently contribute data to the aggregate data pool. There are 1,987 patients in the asthma data base, many with multiple observations over time.
For more information: Christopher M. Kania, MA, Research Coordinator, American Medical Group Association; Tel: 703-838-8460; Fax: 703-548-1890.

MHCA (Managed Health Care Association)
Outcomes Management System Consortium Asthma Project

Project Goal:
Enable Managed Care Organizations (MCOs) to use outcomes information to improve patient care.
Description:
Fifteen MCOs administered baseline, one- and two-year follow-up surveys (4,895 patients at one-year follow-up); at baseline patients were 18 or over with at least two medical encounters with asthma diagnosis (ICD-9 code 493) in previous 24 months; sample selected to be representative of all asthma patients with sufficient numbers with severe asthma. The study measured symptom severity, functional status, work days lost, and use of medical care services via patient self-report as the primary source, with physician report, claims or other records as secondary sources.
Preliminary findings indicate that conformity to treatment guidelines improves asthma symptoms and receiving care from an asthma specialist seems to reduce the number of work days lost. Preliminary work is being extended to develop a "risk index" that could be computed for all asthma patients using a brief patient questionnaire.
Status:
In process; one year follow-up completed
Project Management:
Health Outcomes Institute, Bloomington, MN
For more information:
Contact Michael Huber or Darcy Frana at Health Outcomes Institute,
Tel: 612-858-9188; Fax: 612-858-9189.

Outcomes Management Systems - Design and Development Efforts

American Academy of Pediatrics
Functional Outcomes Project: Asthma

Project Goal:
Develop valid tools to assess the impact of medical treatment on the lives of children and their families. The Functional Outcomes Project is one of four efforts sponsored by the Academy to enhance quality improvement in children's health care.
Description:
The Academy has developed and pilot tested self-report and telephone-administered questionnaires for parents of children age 5 to 12 with chronic asthma. Domains measured include Activity, Physical Health, Emotional Health, Social Interaction, and Health Care Utilization. A physician-completed medical exam form accompanies the parent survey. The parent self-report instrument has gone through two rounds of pilot testing and revision. The Academy was recently awarded a subcontract on an AHCPR PORT-II grant (with Kevin Weiss, MD, MPH, Principal Investigator) that will support a field test of the instrument. This field test will provide a strong validation of the instrument as an evaluative tool.
Status:
Field test upcoming; development of an adolescent self-report measure in process.
Project Management:
Department of Research, American Academy of Pediatrics, Elk Grove Village, Illinois
For more information:
Contact Linda Asmussen, American Academy of Pediatrics,
Tel: 847-981-7625; Fax: 847-228-9651; E-mail: Lasmussen@aap.org.

Joint Council on Allergy, Asthma and Immunology:
Asthma Outcomes Monitoring System (AOMS)

Project Goal:
Develop a comprehensive, standardized and practical system for monitoring the functional health and well-being (FH&WB), clinical and economic outcomes of adult and pediatric asthma patients.
Description:
All forms have been developed and pilot-tested. Patients of Board-certified allergists nationwide will be enrolled in the AOMS in the fall 1996 and followed for one year or longer. The AOMS includes patient- and clinician-completed encounter forms. The AOMS will yield data on: generic and asthma-specific FH&WB, disease severity, work or school performance and days lost, resource utilization, patient satisfaction, use of medications and other treatments for asthma, and other topics.
Measures Used:
Patient -based assessments will include the SF-36 and SF-12 (adults) and the CHQ-PF28 and CHQ-PF50 (children) for generic FH&WB, Marks AQLQ for symptom-specific QOL, GHAA's Consumer Satisfaction Survey and many other items or modules.
Status:
Patient recruitment will take place during September and October, 1996.
Project Management:
The Health Institute at New England Medical Center, Boston, MA
For more information:
Contact Martha Bayliss, The Health Institute; Tel: 617-636-8648; Fax: 617-636-8077.

Outcome Measures in
Ambulatory Care Study (England)

Project Goal:
To develop practical, condition-specific, patient-centered health outcome measures for routine application in judging the effectiveness of care for groups of patients with asthma (and diabetes) receiving care in ambulatory settings.
Description:
The researchers administered existing generic and disease-specific health status and health-related quality of life measures to 1,300 patients. Core sets of scales drawn from the existing measures were identified that explained most variation in general health perception. The researchers developed their own severity questionnaire. Work in progress aims to test the measures' responsiveness to change and attributability. Another study to assess the test-retest reliability of the core sets over a three-week interval is also in progress. Work has already begun to provide site-specific comparative feedback.
For more information:
Professor Allen Hutchinson, Department of Public Health Medicine, University of Hull,
Hull, England, HU6 7RX, UK.
(See also reference "Developing Outcomes Measures..." by E McColl et al on p. 9.)

External Comparison Efforts

Foundation for Accountability (FACCT)
System Performance Measures: Asthma

Project Goal:
Endorse a set of performance measures for treatment of adult asthma for voluntary use by health care providers.
Description:
Following its established format for endorsing measures, FACCT commissioned a scientific paper and developed, with an advisory group, a draft asthma measure set. FACCT measure sets usually address five areas: quality of life, clinical outcomes, satisfaction, essential processes, and burden. Kevin Weiss, MD, from Rush Presbyterian Hospital is the scientific author of the paper, which provides an overview of quality measures. FACCT is working with an advisory group with representation from the Health Care Financing Administration, American Lung Association, and the Asthma and Allergy Foundation of America.
Status:
In May 1996, FACCT's Board of Trustees recommended more research on methodological issues before release of the performance measures. A clinical meeting is planned for November and additional consumer focus groups will also be scheduled. Expected release date of the asthma measures is January 1997.
Project Management:
Foundation for Accountability, Portland, OR
For more information:
Christina Bethell, PhD, FACCT; Tel: 503-223-2228; Fax: 503-223-4336.

Central Health Outcomes Unit, Department of Health, England:
Asthma Outcome Indicators

Project Goal:
To recommend a set of outcome indicators related to the prevention and treatment of asthma
Description:
Four organizations are contributing to the development of a recommended set of outcome indicators for asthma. Nineteen candidate indicators have been fully specified and allocated a recommendation category in terms of whether they should (1) be collected on a routine basis or by periodic survey; or (2) be implemented now or need further work. The organizations involved are the Unit of Health-Care Epidemiology, University of Oxford; NHS Centre for Reviews and Dissemination, University of York; UK Clearing House on Health Outcomes, University of Leeds; and CASPE Research, London.
Status:
A final report has been submitted to the Department of Health, London.
For more information:
Contact Dr. Alastair Mason, Unit of Health-Care & Epidemiology, Department of Public Health and Primary Care, University of Oxford, Old Road, Headington, Oxford, England OX3 7LF.
Tel: +44 1865 226994 or FAX: +44 1865 226993.

Asthma Evaluation in Health Plans
Harvard Pilgrim Health Care.

This Massachusetts HMO has surveyed yearly adult and pediatric asthmatics, identified through clinical encounters and claims databases, to determine their functional status/quality of life, their levels of satisfaction with their asthma care; and the plan's provision of and their use of anti-inflammatory medications, home peak-flow meters, and written home-treatment plans. The surveys have been administered to adults for the last three years, and to children for the last two years. On a quarterly basis, rates of hospital admissions, ER visits, and outpatient visits for asthma for the preceding two quarters are presented graphically and in tables to clinicians and administrative leaders and staff. These utilization rates are calculated for all asthmatics (both respondents and nonrespondents to the survey). Clinic or health center staff (chiefs, case managers) receive case listings of their asthmatics who were seen in the ER or hospital for asthma during the measurement period.

Asthma Outcomes at Point-of-Service:
Dean Clinic's "Asthma Report Card"
Based on the work of Don Bukstein, MD, Assistant Clinical Professor of Family Practice University of Wisconsin, and Director of Pediatric Pulmonary Disease, Dean Clinic, Madison, Wisconsin and Allan T. Luskin, MD, Associate Professor of Medicine, Associate Professor of Immunology/ Microbiology, Rush Medical College, Chicago, Illinois and Director, Respiratory Center, Dean Medical Center, Madison, Wisconsin.

The Allergy Department of Dean Clinic in Madison, Wisconsin has given a uniquely individual meaning to the healthcare report card. Dean's Asthma Report Card measures the asthma patient's overall condition and helps both patient and physician track progress over time.
Since 1994, the Allergy Department has engaged in point-of-service outcomes measurement. According to Dr. Don Bukstein, the routine use of the "Asthma Report Card" acts as a roadmap for the clinicians, leading to more accurate diagnoses and more valid conclusions about the patient's asthma. The direction provided by the report cards helps physicians avoid problems that can arise from not communicating fully with the patient, misinterpreting cues, missing important patient information, and reacting to the patient inappropriately. "Point-of-service outcomes information," he maintains, "leads to better patient care."

The Asthma Report Card is a 96-item, fill-in-the-dot questionnaire, completed by both patient and physician. Parents complete the forms on behalf of children too young to complete the questionnaire on their own. Contained in the questionnaire are the Health Status Questionnaire (also known as the SF-36 Health Survey) plus an additional three questions screening for depression, the Asthma TyPE (also known as the Bethel 10.1), a set of patient satisfaction questions, and a set of physician-completed questions on diagnosis, medications, and pulmonary function, including FEV-1 and peak flows for all patients able to perform pulmonary function testing.

The Health Status Questionnaire and the Asthma TyPE are printed in one booklet and are usually sent to the patient prior to a visit for return by mail or FAX. The practice receptionist scans the questionnaire and produces a graph chart for the patient and physician to review during the visit. The satisfaction portion of the survey is completed after the visit and, together with the physician's questionnaire, is scanned and added to the patient's electronic file.

Dr. Bukstein and Dr. Luskin have identified several ways the point-of-service outcomes assessment has affected their allergy practice.

Better patient information
Patients tend to be somewhat more truthful in filling out a printed form than they are responding face-to-face to a physician. And in an office visit, both physicians and patients tend to focus on ostensible reasons for the visit, but often take little notice of seemingly unrelated but important symptoms. Also the physician or patient might feel rushed in the face-to-face interview. Sometimes supposedly "irrelevant" information that might be overlooked in a personal interview does not escape attention with a written, standardized questionnaire.

Dr. Luskin provided an example with the following scenario. Jimmy, a 14-year old with a long history of bronchial asthma, arrives with his mother at an office visit requesting a prescription refill for albuterol. Feeling pressured by her absence from work and Jimmy's absence from school, the mother insists her son is much better and just wants the prescription refilled. Jimmy affirms that he's fine. Dr. Luskin points out that since the physician is busy, too, and the physical exam confirms a clear chest, the most expedient action would be to simply refill the prescription. However, based on responses from Jimmy's questionnaire indicating low physical functioning, decreased energy, and decreased general health perception, the physician now has reason to believe there may be inadequate asthma control. Instead of simply refilling the prescription, the physician schedules another appointment where he discovers that Jimmy has missed six days of school due to emergency room visits, has used 15 inhalers in the past year, and that his pulmonary function is low. The physician adjusts the treatment plan, provides a peak flow meter, and educates Jimmy and his mother about the chronic nature of the condition and the use of the various medications. Eight weeks later, at the follow-up visit, the outcomes data shows marked improvement.

"Patients typically set goals for asthma therapy too low and accept a much lower than optimal level of functioning," explains Dr. Luskin. With low expectations, they have no incentive to discuss alternate treatment options with the physician. Capturing more complete patient information can avoid the overutilization of healthcare services and productivity loss that accompany suboptimal asthma treatment.

More efficient clinical encounters
In patient encounters that are of necessity brief, discussions are often short and decisions are made quickly. The physician must move efficiently from one conversation to the next, cognizant of the unique circumstances of each. The Asthma Report Card offers short cuts that help organize and collect relevant information quickly and efficiently.

Point-of-service outcomes data often provides clear, long-term documentation of the patient's disease course. When patients visit the office, they see on their Asthma Report Cards the results of previous questionnaires. With only oral reporting, the patient might have forgotten how he or she felt on a previous visit. Seeing past and current status helps a patient become actively engaged in his or her own treatment.

Improved referral and prescribing practices
One of the first changes noted in connection with the use of questionnaires at the Dean Clinic Allergy practice was the increased number of psychiatric referrals. If a patient responded "My asthma is really bad and my energy is down," but exhibited no physical symptoms or changes in pulmonary function, physicians learned to look for depression. These patients were then referred immediately for psychiatric evaluation where they received relief and avoided unnecessary overutilization of healthcare services.

Use of the questionnaires also helped contribute to a significant change in the use of peak flow meters. Outcomes data revealed that distributing peak flow meters to very mild intermittent asthmatics did not affect the quality of their care. Consequently, the Dean Clinic Allergy practice concentrated efforts and financial resources on getting peak flow meters to patients who had more severe asthma.

Encouragement, with a dose of caution, for point-of-service outcomes assessment
The Allergy Department of the Dean Clinic is the only allergy practice known to Drs. Bukstein and Luskin where outcomes information is used at point-of-service. They encourage other physicians to pursue a computerized system that makes point-of-service information available and outcomes studies manageable. They also advise to proceed with caution in interpreting responses at the individual level. Dr. Bukstein points out that the physicians in their clinic are careful not to make individual judgments based on normative values because "A patient placed in a normative group may not really belong there." Individual patients are compared with their own previous measurements of function. "Outcomes measures of functioning at the individual level are similar to peak flow measurements in that it is always better to compare an individual patient to his own personal data over time, rather than to compare that same individual to a normative group." Dr. Bukstein encourages clinicians using general and specific health status measures to communicate with each other to minimize problems in reliability, validity, and responsiveness of the available outcomes tools.

Additional reading:
Bukstein D, Luskin AT. Successful Treatment of Chronic Asthma Using Point-of-Service Outcomes Analysis. Journal of Clinical Outcomes Management. March/April 1995, 2:2; 45-60.
Bukstein, DA. A Practical Approach to the Use of Outcomes in Asthma. A manuscript prepared for Immunology and Allergy Clinics of North America. Reprint requests to Donald A. Bukstein, MD, Dean Clinic, 1313 Hatchery Road, Madison, WI 53715.

Resources
A description of measures and methods to evaluate asthma and outcomes of asthma care

Quality of Life/Functional Status
There are several questionnaires already developed to measure quality of life and functional status. New instruments are under development for the initiative sponsored by the American Academy of Pediatrics (p. 3). Hutchinson and colleagues (p. 4) are testing the performance of individual items from a selection of these instruments to develop a condensed instrument.

Generic Questionnaires
The generic instruments included below have been evaluated by the Trust's Scientific Advisory Committee and are currently available through the Trust. Citations are included for publications where the instrument's measurement of asthma has been discussed.

Sickness Impact Profile (SIP)

The SIP has been instrumental in the development and validation of asthma-specific questionnaires. For example, the SIP was applied as a standard to evaluate the construct validity of Marks and Juniper's Asthma Quality of Life Questionnaires*. In clinical trial settings, the SIP has demonstrated less responsiveness to treatment effects than asthma-specific measures such as Juniper's AQLQ and the St. George's Respiratory Questionnaire**. Rather, the advantages of using the SIP in asthma clinical trials are related to its potential for comparisons across diseases and for detecting unexpected effects of treatment.

Rowe BH and Oxman AD. Performance of an asthma quality of life questionnaire in an outpatient setting. Am Rev Respir Dis 1993; 148:675-681.
Thomas K, Ruby J, Peter JV, Cherian AM. Comparison of disease-specific and a generic quality of life measure in patients with bronchial asthma. Natl Med J India. 1995 Nov-Dec;8(6):258-60.
Rutten-van Molken MP, Custers F, van Doorslaer EK, Jansen CC, Heurman L, Maesen FP, Smeets JJ, Bommer AM, Raaijmakers JA. Comparison of performance of four instruments in evaluating the effects of salmeterol on asthma quality of life. Eur Respir J. 1995 Jun;8(6):888-98.
*See references for these instruments below under "Condition-Specific Quality of Life."
** Jones PW and the Nedocromil Sodium Quality of Life Study Group. Quality of life, symptoms, and pulmonary function in asthma: long-term treatment with nedocromil sodium examined in a controlled multicentre trial. Eur Respir J 1994; 7:55-62.

SF-36 Health Survey

The SF-36 has been used to measure health-related quality of life in a number of asthma studies. Major studies include:

Bone RC. Health-related quality-of-life assessment in medical care. Dis Mon 1995; 4-71. (Review article that focused on asthma.)
Bousquet J, Knani J, Dhivert H, Richard A, Chichoye A, Ware JE, et al. Quality-of-life in asthma: I. internal consistence and validity of the SF-36 Questionnaire. Am J Respir Crit Care Med, Vol. 149; 1994, 371-5.
Lyons RA, Lo SV, Littlepage BNC. Comparative health status of patients with 11 common illnesses in Wales. J Epid Comm Health 1994; 48 (4):388-90.
Noonan M, Chervinsky P, Busse WW, Weisberg SC, Pinnas J, deBoisblanch BP, et al. Fluticasone propriate reduced oral prednisone while it improves control of asthma and quality of life. Am J Respir Crit Care Med, Vol. 152; 1995, 1467-73.
Okamoto LJ, Noonan M, DeBoisblanc BP, Kellerman DJ. Fluticasone propionate improves quality of life in patients with asthma requiring oral corticosteroids. Ann Allergy Asthma Immunol. 1996 May; 76 (5):455-61.
Viramontes JL, O'Brien B. Relationship between symptoms and health-related quality of life in chronic lung disease. J Gen Intern Med 1994; 9(1):46-8.

Quality of Well Being Scale (QWB)

The QWB has is not prevalent in published studies of asthma. However its performance in a related population, COPD patients, is discussed in the following studies.

Eakin EJ, Kaplan RM, and Ries AL. Measurement of dyspnoea in COPD. Qual Life Res. 1993; 2: 181-191.
Ries AL, Kaplan RM, Limber TM, Prewitt LM. Effects of pulmonary rehabilitation on physiologic and psychosocial outcomes in patients with COPD. Ann Intern Med. 1995; 122: 823-832. (Randomized clinical trial with 6 year follow-up.)
Kaplan RM, Ries AL. Cost-effectiveness of pulmonary rehabilitation (Chapter 16). In Pulmonary Rehabilitation (Eds) Fishman, AP. 1996. (Good review of issues in assessing effectiveness of intervention for COPD and asthma).
Toshima MT., Kaplan RM, Ries AL. Experimental evaluation of rehabilitation in COPD: Short-term effects on exercise endurance and health status. Health Psych. 1990;9: 237-252.

Condition-Specific Quality of Life
To capture the specific impact of asthma on an individual's quality of life or functional status, several instruments have been developed and tested. The instruments described below are representative of condition-specific measures. For large scale, multi-site outcomes measurement such as those described on page 3-4, researchers have selected or adapted these instruments.

Living with Asthma Questionnaire (Hyland)

68-item scale, covering 12 domains: social/leisure, sport, holidays, sleep, work and other activities, colds, mobility, effects on others, medication use, sex, and dysphoric states and attitudes. Patients respond to statements on a 1-4 scale ranging from "untrue of me" to "very true of me."

Hyland ME, Finnis S, Irvine SH. A Scale for Assessing Quality of Life in Adult Asthma sufferers. J of Psychosom Res. 1991; 35(1):99-110.

Asthma TyPE

Developed by Robert A. Bethel, MD, this measure includes patient and clinician forms. The patient form includes 13 questions, including use of health services, night disturbance, physical and emotional functioning, daily activities, social activities, medications and symptoms. The clinician form collects information such as peak flow, medication use, specific environmental triggers.

Available from the Health Outcomes Institute, 2001 Killebrew Drive, Suite 122, Bloomington, MN 55425.

Asthma Quality of Life Questionnaire (Juniper et al)

32 questions covering activity limitations, symptoms, emotional function, and exposure to environmental stimuli. Self or interview administered. Since the patient chooses which activities to rate, patient-specific information is obtained. Domain scores are produced, based on averages of the all items in the domain. Responses are captured on a 7-point scale.

Juniper E, Guyatt GH, Epstein RS, Ferrie PJ, Jaeschke R, Hiller TK. Evaluation of impairment of health related quality of life in asthma: Development of a questionnaire for use in clinical trials. Thorax. 1992;47:76-83.
Juniper EF, Guyatt GH, Feffie PJ, Griffith LE. Measuring Quality of Life in Asthma. Am Rev Respir Dis, 1993;147:832-838.

Asthma Quality of Life Questionnaire (Marks et al)

A 20-item self-administered questionnaire with Likert scale responses to measure quality of life in adult subjects. Yields a total scale score together with subscale scores for breathlessness, mood disturbance, social disruptions, and concern for health.

Marks GB, Dunn SM, Woolcock AJ. A scale for the measurement of quality of life in adults with asthma. J Clin Epidemiol 1992; 45(5):461-472.

Clinical Severity Indicators
Pulmonary Function:
Spirometry testing in the physician's office and monitoring with peak flow meters in the patient's home are the major clinical indicators for severity and frequency of the airway obstruction. For an analysis, see Enright P, Lebowitz MD, Cockroft DW. Physiologic Measures: Pulmonary Function Tests. Am J Respir Crit Care Med, Vol 149, 1994; S9-S1 8.
Medication Use:
A 1994 US paper found no existing severity methods based on medication use for asthma but proposed a weighted scoring approach for further testing. For an analysis, see Busse WW, Maisiak R, Young KR. Treatment Regimen and Side Effects of Treatment Measures. Am J Respir Crit Care Med, Vol 149, 1994; S44-S50.
Symptom Measures:
A variety of self-reported symptom measures have been developed for use in population studies and clinical trials. These measures seek to characterize wheezing, dyspnea, cough, sputum production, and tightness or discomfort in the chest in terms of intensity, duration and frequency, and bothersomeness. For an analysis, see O'Connor GT, Weiss ST. Clinical and Symptom Measures. Am J Respir Crit Care Med, Vol 149, 1994; S21-S28.

Patient Management/Adherence
Questionnaires have been developed to measure self-management behaviors for adults and children. Questions address topics such as adherence to prevention and treatment regimens and environmental control exercised. For an analysis, see Clark NM, Starr-Schneidkraut NJ. Management of Asthma by Patients and Families. Am J Respir Crit Care Med, Vol 149, 1994; S54-S66 and Rand CS, Wise, RA. Measuring Adherence to Asthma Medication Regimes. Am J Respir Crit Care Med, Vol 149, 1994; S69-S76.

Health Care Utilization
Mortality data and hospitalization rates are the traditional outcomes measured. Emergency room use and ambulatory care visits are now being used more frequently. For an analysis, see Vollmer WM, Osborne ML, Buist AS. Uses and Limitations of Mortality and Health Utilization Statistics in Asthma Research. Am J Respir Crit Care Med, Vol 149, 1994; S79-S87.

Patient Satisfaction
Customized measures for patient satisfaction are not available in the literature. For satisfaction measures in current use, see Initiatives on p. 3-4.

Productivity/Opportunity Loss
Many of the outcomes initiatives have designed their own forms to capture this information, either from patient self-report or administrative data. (See Initiatives, p. 3-4)

Research Summary: 1995-1996
The following articles were identified through a Medline search on Asthma and Quality of Life, Outcome Assessment (Health care), Health Status, Risk Assessment or Severity of Illness. Their abstracts have been abbreviated and they are arranged in reverse chronological order within five categories: Practice Evaluation, Population Monitoring, Pharmaceutical Evaluations, Quality of Life/ Functional Status Measurement, Risk and Severity Measurement.

Practice Evaluation
Comprehensive long-term management program for asthma: effect on outcomes in adult African-Americans. Kelso TM, Abou-Shala N, Heilker GM, Arheart KL, Portner TS, Self TH. Am J Med Sci. 1996 Jun;311(6):272-80.
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.

Randomized comparison of guided self management and traditional treatment of asthma over one year. Lahdensuo A, Haahtela T, Herrala J, Kava T, Kiviranta K, Kuusisto P, Peramaki E, Poussa T, Saarelainen S, Svahn T. BMJ. 1996 Mar 23;312(7033):748-52.
Self management reduces incidents caused by asthma and improves quality of life.

Effect of allergy specialist care on the quality of life inpatients with asthma. Storms B, Olden L,Nathan R, Bodman S. Ann Allergy Asthma Immunol. 1995 Dec;75(6 Pt 1):491-4.
Evidence that allergy specialty care improves asthma quality of life over primary care treatment.

Developing outcome measures for ambulatory care -- an application to asthma and diabetes. McColl E, Steen IN, Meadows KA, Hutchinson A, Eccles MP, Hewison J, Fowler P, Blades SMTI. Soc Sci Med. 1995 Nov; 41(10):1339-48.
This paper describes the theory and practice underlying the development of outcome measures for two chronic conditions, asthma and diabetes, for application in ambulatory settings.

Peak flow based asthma self-management: a randomized controlled study in general practice. Jones KP, Mullee MA, Middleton M, Chapman E, Holgate ST. Thorax. 1995 Aug;50(8):851-7.
Rigid adherence to long term daily peak flow measurement in the management of mild asthma in general practice does not appear to produce large changes in outcomes. Self-management and the use of prescribed peak flow meters need to be tailored to individual circumstances.

Evaluation of an asthma self-management education program. Boulet LP, Boutin H, Cote J, Leblanc P, Laviolette M. J Asthma. 1995;32(3):199-206.
The goal of this study was to evaluate a self-management education program for asthmatics. Knowledge of asthma and its management was significantly improved immediately after the program and 1 year later.

Self-management, autonomy, and quality of life in asthma. Gibson PG, Talbot PI, Toneguzzi R. Chest. 1995 Apr;107(4):1003-8.
We conclude that while asthmatics have strong desires to be informed about their illness, they do not wish to be the prime decision makers during an exacerbation. These findings have implications for the success of self-management programs and action plans.

Evaluating the effect of in-patient versus observation unit treatment on quality of life in asthma patients [abstract] Isola ML, Rydman RJ, AHSR FHSR Annu Meet Abstr Book. 1995;12:60.
What the patient identifies as better, worse, or unchanged [is important] because this will lead to practical solutions for improving asthma management. These data provide evidence of patterns and relationships and identified areas to focus on for improving asthma care.

Population Monitoring
A computer-assisted telephone interview technique for assessment of asthma morbidity and drug use in adult asthma. Anie KA, Jones PW, Hilton SR, Anderson HR. J Clin Epidemiol. 1996 Jun;49(6):653-6.
Telephone interviews can provide repeatable and efficient measurements of health and patient-reported drug use in asthma with St. George's Respiratory Questionnaire.

Pharmaceutical Evaluations
Fluticasone propionate improves quality of life in patients with asthma requiring oral corticosteroids. Okamoto LJ, Noonan M, DeBoisblanc BP, Kellerman DJ. Ann Allergy Asthma Immunol. 1996 May; 76(5):455-61.
Health-related quality of life improved in patients with severe asthma following 16 weeks of treatment with fluticasone propionate, 1000 micrograms bid. These improvements were maintained during subsequent fluticasone propionate treatment over a 1-year period.

Influence of inhaled steroids on recovery from occupational asthma after cessation of exposure: an 18-month double-blind crossover study. Malo JL, Cartier A, Cote J, Milot J, Leblanc C, Paquette L, Ghezzo H, Boulet LP. Am J Respir Crit Care Med. 1996 Mar;153(3):953-60.
Inhaled corticosteroids induce a small but significant overall improvement of the asthmatic condition in subjects with occupational asthma caused by high- and low-molecular-weight agents after withdrawal from exposure. The beneficial effect is, however, more pronounced if inhaled steroids are given early after diagnosis.

Effect of treatment with zileuton, a 5-lipoxygenase inhibitor, in patients with asthma. A randomized controlled trial. Zileuton Clinical Trial Group. Israel E, Cohn J, Dube L, Drazen JM. JAMA. 1996 Mar 27;275(12):931-6.
Three months of 5-lipoxygenase inhibition produced a significant improvement in asthma control. These data indicate that 5-lipoxygenase products of arachidonic acid metabolism are mediators of inflammation with an important role in the biology of asthma.

Effects of nedocromil sodium in steroid-resistant asthma: a randomized controlled trial. Marin JM,Carrizo SJ, Garcia R. Ejea MV. J Allergy Clin Immunol. 1996 Feb;97(2):602-10.
Inhaled NS improved pulmonary function and decreased asthma severity in steroid-resistant asthma, but its effectiveness is not homogeneous and cannot be predicted from baseline clinical data.

Fluticasone propionate reduces oral prednisone use while it improves asthma control and quality of life. Noonan M, Chervinsky P, Busse WW, Weisberg SC, Pinnas J, de Boisblanc BP, Boltansky H, Pearlman D, Repsher L, Kellerman D. Am J Respir Crit Care Med. 1995 Nov;152(5 Pt 1):1467-73.
Fluticasone propionate aerosol (750 or 1,000 micrograms twice daily) effectively and safely allowed most asthmatics dependent on oral corticosteroids to reduce or eliminate oral prednisone use while improving pulmonary function and quality of life.

The place of long-acting beta 2 agonists in primary care. Stott PC. Br J Clin Pract Symp Suppl. 1995 Sep;81:20-1.

Salmeterol. An appraisal of its quality-of-life benefits and potential pharmacoeconomic positioning in asthma. Peters DH, Faulds D. Pharmacoeconomics. 1995 Jun;7(6):562-74.
When added to existing asthma therapy, salmeterol improves patient quality of life in the short term (up to 3 months). It may also have some beneficial effects on patient well-being when used to provide a steroid-sparing effect.

Budesonide. An appraisal of the basis of its pharmacoeconomic and quality-of-life benefits in asthma. Davis R, McTavish D. Pharmacoeconomics. 1995 May; 7(5):457-70.
Available data provide an encouraging pharmacoeconomic rationale for budesonide as first-line asthma therapy, and a good basis for future pharmacoeconomic analysis of asthma management.

The influence of an inhaled steroid on quality of life in patients with asthma or COPD. van Schayck CP, Dompeling E, Rutten MP, Folgering H, van den Boom G, van Weel C. Chest. 1995 May;107(5):1199-205.
It was concluded that BDP did not improve the general well-being of patients with asthma or COPD as measured by these generic health instruments. However, BDP significantly improved the course of lung function and temporarily decreased the severity of symptoms.

Effect of a novel potent platelet-activating factor antagonist, modipafant, in clinical asthma. Kuitert LM, Angus RM, Barnes NC, Barnes PJ, Bone MF, Chung KF, Fairfax AJ, Higenbotham TW, O'Connor BJ, Piotrowska B, et al. Am J Respir Crit Care Med. 1995 May;151(5):1331-5.
There was no significant difference between placebo and modipafant in diurnal variation in PEF, morning and evening PEF, clinic FEV1, rescue bronchodilator usage, symptom score, or airway responsiveness. We previously showed that the racemate UK-74,505 had no effect on antigen challenge, and this study shows that the active (+)-enantiomer modipafant has no effect in chronic asthma. This suggests that PAF is not an important mediator in asthma.

Quality of life in asthma clinical trials: comparison of salmeterol and salbutamol. Juniper EF, Johnston PR, Borkhoff CM, Guyatt GH, Boulet LP, Haukioja A. Am J Respir Crit Care Med. 1995 Jan;151(1):66-70.
The aim of this study was to compare the effects of salmeterol (50 micrograms twice daily), salbutamol (200 micrograms four times a day), and placebo on asthma-specific quality of life and to relate the findings to conventional clinical asthma outcomes.

Quality of Life/Functional Status Measurement
Comparison of disease-specific and a generic quality of life measure in patients with bronchial asthma. Thomas K, Ruby J, Peter JV, Cherian AM. Natl Med J India. 1995 Nov-Dec;8(6):258-60.
Though both AQL and SIP were valid measures of quality of life, AQL is likely to be more capable of detecting smaller changes in the health status of patients with bronchial asthma and hence was chosen as the instrument in the proposed clinical trial.

Asthma knowledge, attitudes, and quality of life in adolescents. Gibson PG, Henry RL,Vimpani GV, Halliday J. Arch Dis Child. 1995 Oct;73(4):321-6.
Adolescents with asthma, their peers, and their teachers were studied in order to establish the level of knowledge concerning asthma and its management, their attitudes towards asthma, and the degree quality of life impairment due to asthma. There are opportunities to intervene and improve asthma management among adolescents.

Developing condition-specific measures of functional status and well-being for children. Sullivan SA, Olson LM. Clin Perform Qual Health Care. 1995 Jul-Sep;3(3):132-9.
This paper describes the American Academy of Pediatrics' program to develop several condition-specific measures to assess the functional status and well-being of children with chronic conditions.

Quality-of-life considerations in the treatment of asthma. Juniper EF. Pharmacoeconomics. 1995 Aug;8(2):123-38.
There are now a number of health-related quality-of-life instruments, with strong measurement properties, that may be used in both clinical studies and practice. They provide scientifically sound and valid information about patients' experience that complements conventional clinical measures.

The assessment of dyspnea in bronchial asthma. Subratty AH. Int J Clin Pharmacol Ther. 1995 Jul;33(7):394-6.
A new quality of life measuring instrument for the self-assessment of dyspnea in bronchial asthma patients while performing a daily activities test.

Validation of an asthma quality of life diary in a clinical trial. Hyland ME, Crocker GR. Thorax. 1995 Jul;50(7):724-30.
Structured QOL diaries are valid instruments that appear to be more responsive to longitudinal change in clinical trials than a QOL questionnaire, but QOL questionnaires provide a more sensitive cross sectional measure of disease severity.

Comparison of performance of four instruments in evaluating the effects of salmeterol on asthma quality of life. Rutten-van Molken MP, Custers F, van Doorslaer EK, Jansen CC, Heurman L, Maesen FP, Smeets JJ, Bommer AM, Raaijmakers JA. Eur Respir J. 1995 Jun;8(6):888-98.
Salmeterol showed a greater improvement in quality of life compared to patients given salbutamol. Of the disease-specific questionnaires the Asthma Quality of Life Questionnaire was found to be more responsive to change than the Living With Asthma Questionnaire and showed greater validity. Of the generic instruments, the rating scale utilities were most responsive. The Standard Gamble showed poor correlation with other measures.

[The quality-of-life questionnaire with asthma patients: the Spanish version of the Asthma Quality of Life Questionnaire] (in Spanish). Sanjuas C, Alonso J, Sanchis J, Casan P, Broquetas JM, Ferrie PJ, Juniper EF, Anto JM. Arch Bronconeumol. 1995 May;31(5):219-26.
This paper describes the translation to Castilian and adaptation of a quality of life measurement instrument: the Asthma Quality of Life Questionnaire (AQLQ) by Juniper et al.

[The quality of life in asthma: an evaluation of the AQLQ questionnaire for its use on a Spanish population. Asthma Quality of Life Questionnaire] (in Spanish) . Perpina M, Belloch A, Pascual LM, de Diego A, Compte L. Arch Bronconeumol. 1995 May;31(5):211-8.
We analyzed the reliability, content validity and construct validity (convergence and divergence) of one instrument, the Asthma Quality of Life Questionnaire (AQLQ) designed by Marks and colleagues.

Health-related quality-of-life assessment in medical care. McSweeny AJ, Creer TL. Dis Mon. 1995 Jan; 41(1):1-71.
In this monograph, theoretical issues related to the term and the reasons assessment of quality of life is important are discussed.

Risk and Severity Measurement
Rating asthma severity: a patient-based measure for asthma severity risk adjustment [abstract]. Campbell D, Wu A, Yasui Y, Skinner A, Steinwachs D. AHSR FHSR Annu Meet Abstr Book. 1995;12:84-5.
This paper describes development and validation of an asthma severity measure from patient self-reported symptoms for the Managed Health Care Association's (MHCA) Asthma project's risk adjusted comparisons of asthma care performance among 16 Managed Care Organizations. A valid asthma severity risk adjuster for quality of care comparisons among MCOs can be developed from patient-reported symptoms.

The use of children's medical records to predict the risk of asthma attack. Neville RG, Bryce FP, Clark RA, Crombie IK. Scott Med J. 1995 Oct;40(5):138-40.
Using a large data base of respiratory morbidity in children the opportunity arose to explore the link between what was written in general practice case records and the subsequent risk of a child developing an asthma attack or hospital admission due to asthma. Childhood asthma is an example where the quality of care offered by general practitioners could be improved if a precise estimate of risk could be used to modify clinical management.

The use of objective measures of asthma severity in primary care: a report from ASPN. Fried RA, Miller RS, Green LA, Sherrod P, Nutting PA. J Fam Pract. 1995 Aug;41(2):139-43.
In this study, most primary care clinicians did not have objective data about the severity of their patients' asthma at the time of the encounter. This relative lack of objective data was not explained by lack of access to the relevant technology for determining severity.

Risk factors for recurrent emergency department visits for asthma. Dales RE, Schweitzer I, Kerr P, Gougeon L, Rivington R, Draper J. Thorax. 1995 May;50(5):520-4.
The recommendations of current asthma guidelines are not reaching these patients. The issue of translating guidelines from paper to practice must be addressed before highly effective medications can have an important impact on the frequency of emergency department visits.

Parental asthma knowledge: its association with readmission of children to hospital. Henry RL, Cooper DM, Halliday JA. J Paediatr Child Health. 1995 Apr;31(2):95-8.
To assess whether poor parental knowledge about asthma was a risk factor for readmission of their children to hospital. Knowledge was poor in all groups. Readmission rate was related to the severity of the child's asthma.

Prospective study of hospitalization for asthma. A preliminary risk factor model. Li D, German D, Lulla S, Thomas RG, Wilson SR. Am J Respir Crit Care Med. 1995 Mar;151(3 Pt 1):647-55.
We conducted an exploratory analysis of several prospectively obtained objective measures of disease activity to derive a predictive model of hospitalization for asthma among 310 adults, ages 18 to 50 yr, with moderate to severe asthma. The findings warrant replication and extension in other populations with the goal of developing decision rules for risk stratification and effective interventions for risk reduction.

Assessment of the severity of asthma by an expert system. Description and evaluation. Redier H, Daures JP, Michel C, Proudhon H, Vervloet D, Charpin D, Marsac J, Dusser D, Brambilla C, Wallaert B, et al. Am J Respir Crit Care Med. 1995 Feb;151(2 Pt 1):345-52.
Asthmaexpert, an expert system (ES), was produced at the special request of several clinicians in order to better understand the medical decisions made clinical experts in managing an asthmatic patient. We describe and evaluate this knowledge base, focusing mainly on assessment of the severity of asthma.

Practicing for Patienthood: A Policy Opinion on the Health Care Consumer Agenda
by Sherrie Kaplan, PhD, MPH, Senior Scientist and Co-Director of the Primary Care Outcomes Research Institute at the New England Medical Center in Boston, Massachusetts.

Cost containment is squeezing everybody. Doctors are feeling enormous pressure to do and not do certain things during office visits, such as order specific tests and discuss insurance coverage with patients. Perhaps more importantly doctors are now faced with productivity targets that may not be optimal in terms of patient care. This is especially problematic for patient care involving those with chronic disease where the job is almost as much to persuade the patient to implement the treatment plan as it is to provide good technical care.

If doctors' jobs are being constrained by enforced time limits, what is that doing to the quality of the interpersonal care? Our evidence from the Medical Outcomes Study suggests that care begins to look suboptimal when the office visit lasts 20 minutes or less for patients with chronic disease.1 In shorter visits, there may not be enough time to involve patients effectively in treatment decisions. That involvement appears to have important consequences for patient outcome. There are patients at risk in rapid-fire, cost-conscious environments-older people (75 and over) , young adults (under 30), and men more than women. What is it about those kinds of patients that places them uniquely at risk in the fast-paced, shortened office visit? It could be lack of practice and poor preparation.

How have we prepared people for being a patient? Think about somebody who doesn't have much practice being a patient confronted suddenly with a serious treatment decision. Or somebody who has a chronic disease confronted with a new set of decisions. If you don't have a lot of experience in thinking about what your preferences are or what treatment options might be available other than the one being proposed, you shouldn't be expected to do a very good job in that rapid-fire environment.

Further, the physician-patient interaction takes place under circumstances notorious for impairing recall. You're often without clothes. These days you may easily be without the nice humanistic touches like waiting until you're dressed, sitting in the doctor's office, using a few minutes to chat over the desk before addressing treatment/disease management decisions. Even when you've given some thought to your issues of concern ahead of time, questions you had prepared to ask may be out the window because the doctor found something unexpected on the visit. Or maybe you're suffering from the "white coat syndrome" and are a little nervous in general about seeing the physician. All of this is a set up for impaired recall. Unless you've really planned and prepared ahead of time, the odds that you're going to perform well as a patient are probably random.

More and more autonomy of practice is currently being stripped from physicians. They feel increasingly constrained in their abilities to act on the patient's behalf. What implications does that have for the kind of interpersonal care they give? If physicians are being asked to hand over some aspects of decision-making to patients-while under pressure to conform to guidelines, to limit referrals, to control ordering of tests and performance of procedures, they may well be questioning their role in the New World of health care delivery. Doctors who lose autonomy are unlikely to pass it along to patients.

If we're stuck with a truncated visit to gain the efficiencies that buys us, then the responsible thing to do is to teach doctors and patients to behave in a way that reflects that foreshortened visit and supports activities of both to make the most of the time. We need to help doctors understand that patients don't often come prepared for the visit. Otherwise the physician is spending time, energy and effort that doesn't leave the patient committed to going home and putting the needle in his or her leg. The patient either wasn't convinced that was the right strategy, or there may have been other things he or she wanted to talk over with the doctor before signing on for insulin treatment.

There are ways to achieve such support that don't have huge cost implications and still help physicians and patients to form and implement treatment plans that make medical care work. One of the things we need to do a whole lot smarter is to bring the consumer earlier and more seamlessly into health care delivery and assessment. We are not doing that right now. There is an enormous amount of exciting and developmental work being done primarily as research initiatives that certainly could be implemented today in a more widespread way with immediate and positive consequences.

From the beginning we could be teaching kids what the health care system is all about and how to be a reasonable and cost-effective consumer. We could be teaching people effective disease management-how to do a better job of prevention, how to decide better when to seek medical care, whom to seek it from, how to decide when treatment isn't doing what they wanted it to. We could be teaching people how to maximize a visit with a doctor. We could provide role models of what visits look like: how to comport yourself, how to ask questions, how to negotiate with doctors for the treatment that best fits your lifestyle and circumstances. We could provide 800 numbers, for when the time is up on a visit, to get answers to questions, to get help with disease management at home, where much of it takes place. Eventually we're going to have to get good at this, where we'll log on and ask our question and our question will come back on e-mail.

We must find a more sensible way for bringing the consumer in, from as early as childhood, to making treatment decisions. It is eminently sensible to start preparing people to be efficient and effective consumers of health care resources all the way along. The problem is we're not doing any of the above in any organized or systematized way. Consumers learn the way to be patients the way they learn to be parents, by experience. Some have more than others, some do better than others. It's paternalistic to think that consumer ratings of health care solve the problem for us. What consumers are rating is something that is behind the screen for them. They can't see it and they don't understand quite how it works, what they should get from it, or what they should expect it to accomplish for them. So they rate the food and the parking and whether people were nice to them. But until we bring them closer to the real process of health care and make them vital participants in it, their ratings may tell us less than we think, and we have not made serious and respectful use of the consumer's voice.

If all politics are local, then health care is micro-local: the individual doctor-patient level. People who are trying to manipulate health care at the system level miss the story. The system level offers a top-down approach that misses where the real action is. The patient and doctor operate at the individual level and that's where change is really made. And that's where the clinical and health policy focus needs to be. Preparing those two people to sit down together and talk is where significant resources should be directed. Both sides need to be able to communicate in a comfortable organized way that's supported by the health care system and can help them know and accomplish their agenda in the ten minutes that they will have together.

The lifetime probability that you will be a patient in the US health care system is virtually 100%. If we don't get serious about diverting public and private resources to patient training initiatives, we could be going down a wrong road with dramatic consequences. Consumers need to be brought into this in a sensible way. We've made a good start at it. But we've been at it for twenty odd years now, it's time to get past that good start and think sensibly about where to go next with this.

References
1. Kaplan SH, Gandek B, Greenfield S, Rogers W, Ware JE. Patient and Visit Characteristics Related to Physicians' Participatory Decision-Making Style. Medical Care 1995; 33:12; pp 1176-1187.

Functional Impact and Utility Information:
Decision Support for Treatment Options
Based on an interview with Marc Swiontkowski, MD, Professor and Chief of Orthopedics at the Harborview Medical Center, University of Washington, in Seattle.

As a researcher and an orthopedic trauma surgeon, Marc Swiontkowski, MD, is contributing to the application and development of measures for research and for improving the physician-patient encounter. In his role as Professor and Chief of Orthopedics at an academic medical center, he has gained extensive experience in assessing the functional outcomes of surgery with existing measures, such as the SF-36 Health Survey, the Sickness Impact Profile, and other condition-specific measures such as the Arthritis Impact Measurement Scale (AIMS). Under a recent National Institutes of Health grant, he has also contributed to the development of the Musculoskeletal Function Assessment (MFA) instrument1,2, a 100-item self-reported health status instrument specifically designed for assessing musculoskeletal function in research settings. As a practicing surgeon, Dr. Swiontkowski is looking forward to reaping the benefit that the short-form MFA-now in validation stage-will provide for counseling his patients about treatment options.

The short-form MFA (SMFA) is being developed for use in routine patient monitoring in clinical practice and is being supported in part by the American Academy of Orthopedic Surgeons. "I think that forms like these will prove to be useful in the day-to-day management of patient care, but there is still a real paucity of data along those lines," Dr. Swiontkowski told us. "The community of musculoskeletal practitioners needs outcomes information because we do not know the impact of the majority of our procedures on patient function. We really don't know. I need this information population-based so that I can better inform my patients what to expect as the outcome of a given procedure. For the individual patient, these forms will help me get at information that's difficult to get at on a routine basis."

The SMFA contains items that measure functioning in ten categories (see inset). In the design of the short form, the scaled questions are presented first, such as "Are you tired all the time?" or "Are you making changes in your job?" The last page of the form contains questions evaluating the importance of these categories, such as for the housework category, "How important to you are the problems you are now having doing work around your home?" These questions require a rating response from 1 (not at all important) to 5 (extremely important). Dr. Swiontkowski calls this the utility page and expects this information to be critical in prioritizing the options he discusses with his patients.

"What this chart will show me is the importance to the patient of the functional impact indicated in the categories of functioning. Is the impact extremely important or not really important at all? This is the information that I, a practitioner deciding whether or not to offer a procedure to a patient, want. It will present me with information that would take a much longer time to tease out from an individual patient.

"I can envision myself using this form as a discussion focus after examining the patient. For example, I would say 'Your shoulder stiffness and pain is extremely important to you, why is that?' Then the patient might say something such as 'the stiffness is preventing me from sleeping' or 'I can't reach my demo equipment easily to set up my sales presentations.' With that information, I can explain the options for therapy, exercise, injections or surgical procedures. Using an individual patient's preferences and tolerance for risk, I can educate him about the risks and benefits and efficacy as I understand them and we can reach a decision together. But if that same patient says that shoulder stiffness really doesn't bother him too much, he's started using his other arm as the dominant and found ways around the disability, then I'm going to have a whole different presentation for that individual. I'm not going to spend time educating that patient about the surgical options that might be helpful."

The availability of this particular multipurpose musculoskeletal instrument is still in the future, since the SMFA will not complete its validation phase until the summer of 1997. But Dr. Swiontkowski encourages and regularly consults to orthopedic practices on the use of other generic and condition-specific measures. "What we need a better handle on in our field of orthopedic injury is the functional impacts of 'X' injury-because it helps to counsel patients on what to expect. For example, I want to know what a mean is for an SF-36 score in all eight domains for a patient with a femur fracture so we can counsel the patient early after the injury on what the impact is likely to be. It's not enough to give her the impression that everything's going to be fine after this bone heals up, because that's just not true-even though that's what I was taught. Perfect X-ray equals perfect function-that's really wrong. We need the population-based information that outcomes tools provide to enable practitioners to guide patient expectations."

For clinicians embarking on an outcomes project within an orthopedic setting, Dr. Swiontkowski offers two key pieces of advice. One, start with a question. "Unless you have a plan for how you're going to use the data to answer your question, the routine collection is onerous. It wears out the staff and if you don't have an end-product or a vision in mind for how you're going to use the information, it demoralizes them. It's expensive monetarily and expensive psychologically and it can really poison the waters within an institution or a working group unless you've got a plan for how you're going to do with the data."

Two, choose a common condition, not the entire patient population. "I recommend that practices focus on a clinical entity or procedure that is common enough that there won't be sampling difficulties. Orthopedic surgery constitutes all age groups and it constitutes congenital deformity, arthritis, oncology, injury, metabolic bone disease and more. I try to get practitioners to focus on one of the three top procedures in their practices. If it's a rare clinical entity, it doesn't work. Ensure that there are enough patients coming through your office that there will be enough experience gained and that it will become a part of the staff's everyday thinking. Often working with general orthopedic groups, or orthopedic surgeons involved in many different orthopedic areas, I suggest they analyze the impact of total joint replacement. Total joints work very, very well for outcomes evaluation. The second most common recommendation is probably arthroscopic menisectomy. Those are the two that I recommend that many groups start with."

Although use of functional outcomes measures makes a significant contribution to predicting outcomes for a particular population, Dr. Swiontkowski maintains that the measures will be never be useful in a vacuum. Even when using a specific instrument like the MFA, "it's incumbent on the physician to understand that there are multiple factors that impact a patient. At least on the first visit as you're getting to know the patient, take a careful history. What other things are going on? Or be in good communication with the referring physician as to what the patient's problems are. That's the context in which functional scores and utility information are useful."

This fact should be encouraging. "Many surgeons tend to see a spectre of a clerk in a big insurance company looking at an SF-36 or a short-form MFA score and telling them they can't do a certain procedure because the patient's upper extremity score is thus and such. But I don't ever see these instruments being used to make treatment decisions in a vacuum. If you use the temperature analogy, a chart might show a temperature of 104 degrees, but the patient could be septic from a staph infection or may have just run a marathon. Only the physician will know whether further work-ups and treatment are necessary.

"We must always guard against any potential abuse of these data by educating colleagues, patients, and payors as to how this information should be used. Outcomes instruments help me to treat the individual based on functional impact and utility information and, when data is available in aggregate, how to advise patients about expected outcomes."

References
1. Martin DP, Engelberg R, Agel J, Snapp D, Swiontkowski MF. Development of a Musculoskeletal Extremity Health Status Instrument: The Musculoskeletal Function Assessment Instrument. J of Orthop Res. 1996 14:173-181.
2. Engelberg R, Martin DP, Agel J, Obremsky W, Coronado G, Swiontkowski, MF Musculoskeletal Function Assessment Instrument: Criterion and Construct Validity. J of Orthop Res. 1996 14:182-192.

Preparing for Risk Adjustment
An Interview with Lisa Iezzoni, MD, MSc

Performance comparisons lie inevitably in the future, if not the present, for all health care providers. Whether the comparison will be made among provider organizations or among other clinicians, adjusting for severity of illness is generally essential before any meaningful and credible comparison of outcomes of care can be made. Risk adjustment techniques are designed to account for patient characteristics that might indicate severity of condition before making inferences about the effectiveness of care.

Although there is a growing appreciation for the importance of risk adjustment, there is still minimal consensus on what risk adjustment actually is. In her book Risk Adjustment for Measuring Health Care Outcomes,1 Dr. Iezzoni wrote "...[the] notion of 'risk adjustment'... eludes a single meaning due to its inextricable linkage to the question, risk of what? This query can generate innumerable replies, ranging from resource consumption to imminent death to complications of care to long-term function capabilities to patient satisfaction." She explains that given the diversity of outcomes that can be measured, a uniform definition of what comprises risk is neither possible nor desirable. There are, however, common risk factors that are reasonable to consider (see Figure 1) when measuring an outcome that requires adjustment, although all of these risk factors are rarely practical to take into account. In her book, as in her recent writing, she emphasizes the importance of medical meaningfulness in evaluating the validity of a risk adjustment approach.

The Trust asked Dr. Iezzoni to share her thoughts about how providers can understand the field as it exists today and how to plan for future demands.

Is anything available today that would meet providers' multiple needs for risk or severity adjustment?
Nothing that people can readily take off the shelf. People frequently want to be told what product to buy, but there is no single product that will suit everybody's needs. It is absolutely necessary to identify the major concerns. What is the clinical context and what is the quality or managerial problem? Different clinical areas will require different approaches-cesarean sections will have different risks associated with them than will coronary artery bypass graft (CABG). Once you've defined a clinical area and the type of managerial concern, it becomes easier to see who else has done work in this area. For example, New York state has done significant work in CABG and the New England Cardiovascular Group has a database you can buy into.

Different providers will also consider very different risk factors to be relevant. For instance, a clinician in an inner city public health hospital will not expect the same risk factors to influence length of stay or emergency room use as would a clinician in a suburban, private hospital.

Which severity adjustment systems can providers currently rely on?
Certain clinical areas have been studied in-depth. The severity measures for intensive care units are at a state that they can be used effectively across groups of patients. There are certain types of surgery, such as CABG, that have been so exhaustively investigated people can feel fairly secure. There is still a lot of development that needs to go on to make sure the methods for other groups achieve the same clinical and empirical standards.

This is still a developing field. It's a toddler and it's being asked to run. It's still at its early stages of development.

How then should healthcare organizations begin to use risk adjustment techniques?
The first step is understanding what your questions are and what makes sense for the approach. This is the reason why risk adjustment isn't easy. Clinicians need to begin by sitting down and identifying the concerns.

Providers should begin with a series of questions:

• What is our question about care or performance?

• What are the outcomes that we care about?

• What are the kind of risk factors that are clinically important to us?

• What data do we already have in our information system that would allow us to begin to proxy these risk factors?

• If we don't already have the information, how much are we willing to spend to get it? And if we aren't able or willing to spend very much, are we willing to accept less than perfect risk adjustment?

The healthcare organization as a whole may want to invest in an inexpensive administrative data-based risk adjustment approach. It's a very preliminary screen and there are many, many problems with administrative data. But it makes some people comfortable and it provides relatively inexpensive initial screens.

What advice can you offer for choosing a risk adjustment method?
My colleagues and I have done comparisons of different severity measures and found the comparison process extremely useful. The results of some of these comparisons can be found in several publications.2-5 In the May 1996 Journal of General Internal Medicine we discussed nine severity measures that are widely used to evaluate hospitalized patients across a range of diseases. We outlined some important differences in how severity scores are assigned and provided case studies to illustrate how the same patient is rated using different systems. The major differences among the systems come from these areas: data source, the different types of ratings available from the same data, the definitions of severity, the diagnostic categories used to generate a severity score, scales used for the scoring approach and distribution, and the time intervals used. (See Figure 2)

We found that the most important distinction among severity measures is their source of data. Chart-based measures are preferable for use in quality of care evaluations because they have more clinical content and because code-based measures do not allow for identification of in-house complications. Code-based measures are useful for predicting costs and resource use since they consider all diagnoses treated during hospitalization.

It's important to know the intended use for the system in order to evaluate it. Potential uses could be to address quality of care issues, to stratify costs, to evaluate new treatments, or to conduct research.

Understanding these different purposes is especially important as the organization's needs change. A measure that predicts probability of death may not be able to measure other outcomes well, such as length of stay or resource use.

If a hospital is considering purchasing a system, it is useful to ask each vendor under consideration to score the same sample data set, provided by the hospital. Then assemble a group of clinicians to discuss and compare the ratings. Clinicians should feel comfortable with the severity scores that are being assigned.

What can an organization do to build its capacity to measure outcomes with appropriate risk adjustments?
Pay attention to information systems. Informations systems are the key. Data are expensive. Try to deal with that enormous expense. If data can be electronically transmitted to a central repository where they can be translated into severity ratings, it will allow maximum flexibility with the least expense. An information system that captures the information you will need in the future is critical. This is a challenge for most providers and there are few organizations to use as a benchmark.

Talk to the staff who will use the system and find out what kind of information they need. The technical people designing information systems are essential, but expand out to involve all the staff, especially clinicians. I have been struck by how often the technical designers do not appreciate the fullness of information that clinicians require to make clinical decisions or improve quality of care or services. Take, for instance, nursing information. Functional patterns are recorded on every shift. We know how important this functional information is, but it's not on the information system radar screen. In building information systems, make sure the designers don't focus just on standard information, such as lab reports, but look more broadly at other sources that clinicians know affect the outcomes of care.

You've written over the years about the problem of "black boxes," that is severity systems assigning scores based on formulas not available to the public for comparative purposes. How do you view that problem today?
Products for risk-adjustment are still heavily tilted toward proprietary methods, which makes it difficult to compare across methods. We need some independent standards or ways to evaluate severity measures. But the field is moving so rapidly, it will be hard for an external group to have influence. The state initiatives (such as those in Pennsylvania and Iowa) have provided some context for open comparisons among products, but many vendors are becoming uncomfortable with participating in comparisons. Unless customers band together to demand it, it's not likely we'll see the information we need to compare across severity methods. But if the customer community can talk among themselves, perhaps they can have an influence.

References
1. Iezzoni LI, ed. Risk Adjustment for Measuring Health Care Outcomes. Ann Arbor: Health Administration Press; 1994.
2. Hughes JS, Iezzoni LI, Daley J, Greenberg L. How Severity Measures Rate Hospitalized Patients. J Gen Intern Med, May 1996;11:303-311.
3. Iezzoni LI, Shwartz M, Ash AA, Mackiernan YD. Using Severity Meaures to Predict the Likelihood of Death for Pneumonia Inpatients. J Gen Intern Med, Jan 1996; 11:23-31.
4. Iezzoni LI, Ash AA, Shwartz M, Daley J, Hughes JS, Mackiernan YD. Predicting Who Dies Depends on How Severity is Measured: Implications for Evaluating Patient Outcomes. Ann Intern Med 1995;123:763-770.
5. Iezzoni LI. Editorial: Choosing a Severity Measure. Am J of Med Qual, Fall 1994; 9:3:101-2.

Resources for Clinical Performance Measures:
CONQUEST and the National Library of
Healthcare Indicators

Where do you turn if you're looking for a clinical measure to add to a current or planned outcomes study? In addition to combing the literature, there are two new sources of information designed to help with a search of this kind.

CONQUEST 1.0
Funded by the Agency for Health Care Policy and Research (AHCPR), CONQUEST is a database application, written in Microsoft Access, that enables users to search 53 sets of established clinical performance measures by 52 clinical conditions-and by a variety of other search criteria. Searches can be performed on clinical condition, provider setting, type of health care professional, utilization information and more.
In preparing for this issue of the Monitor, Trust staff searched for all measures relating to Asthma. The report yielded 21 measures used for pediatric and adult asthma care. A sample of the measures reported are:

• PROSPER's1 absence of instructions about signs to contact physician for patients with asthma and lung disease;

• DEMPAQ's2 appropriate prescription of corticosteroids; appropriate prescription of inhaled beta-adrenergic agonist; and appropriate prescription of oral xanthine-derivative; and

• HEDIS3 2.5 Administrative's asthma inpatient admission rate, ages 2-19 and 20-39.

Descriptions of measures include the numerator and the denominator. For instance, in the above example of DEMPAQ's appropriate prescription of oral xanthine-derivative, the numerator is the subset of the denominator who have asthma, emphysema, or chronic bronchitis; the denominator is all patients taking a newly prescribed oral xanthine-derivative.

Other information to evaluate a measure's usefulness in contained in a searchable format. The database also allows users to edit existing measures and to add new ones.

CONQUEST is available at the AHCPR web site:

http://www.ahcpr.gov/research/CONQUEST.htm

or from the AHCPR Clearinghouse at 800-358-9295 (publication number 96-N009). This publication is free and contains the computer disk and user's manual.

NATIONAL LIBRARY OF HEALTHCARE INDICATORS
Sponsored by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), this series of comprehensive indicator catalogues will contain profiles of performance measures. The 1996 publication, the first in a series, describes over 200 measures suitable for health plans and networks.

The measures are described in a standardized format and arranged according to one of three classifications: clinical conditions, functional health status, or satisfaction. For instance descriptive information for clinical condition and functional health status measures follows this format: focus, rationale, type (process, outcome), numerator and denominator statements, domains of performance, delivery settings, testing, stratification, risk adjustment, current development status, submitting organization, and original performance measure source/developer. Domains of performance include ten areas: appropriateness, availability, continuity, effectiveness, efficacy, efficiency, prevention/ early detection, respect and caring, safety, and timeliness.

The 1996 Health Plan and Network Edition will be available for $25 in mid-October. To order the catalogue, call JCAHO's Customer Service Department at 630-792-5800 and ask for order code NHLI-96.

References
1. Patient Reports of System Performance (PROSPER) measure set; patient self-report developed by the Center for Quality of Care Research and Education at Harvard School of Public Health.
2. Develop and Evaluate Methods to Promote Ambulatory Care Quality (DEMPAQ); records-based evaluation developed by the Center for Quality of Care Research and Education at Harvard School of Public Health.
3. Health Employer Data Information Set (HEDIS), a voluntary reporting set sponsored by the National Committee for Quality Assurance (NCQA).

Thank you to October issue contributors
In addition to thanking David Bergman and Michael Blaiss for serving as Editorial Advisors, the Trust also extends its thanks to Linda Asmussen, Martha Bayliss,
Anne Damiano, and Bill Sieber for contributions to the Asthma coverage.

© 1996 Medical Outcomes TRUST