 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a Department of Pediatrics, Mount Sinai School of Medicine, New York, New York
b Department of Pediatrics (Emergency Medicine), Albert Einstein College of Medicine/Jacobi Medical Center, Bronx, New York
c Boston University School of Medicine, Boston, Massachusetts
d Rho, Inc, Chapel Hill, North Carolina
e Department of Pediatrics, University of Washington School of Medicine, Seattle, Washington
f Departments of Pediatrics and Medicine, Northwestern University Medical School, Chicago, Illinois
g National Institute of Allergy and Infectious Diseases, Bethesda, Maryland
h Departments of Medicine and Pediatrics, University of Texas Southwestern Medical Center, Dallas, Texas
i Respiratory Sciences Center, University of Arizona College of Medicine, Tucson, Arizona
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS. The purpose of this study was to test the effectiveness of timely patient feedback in the form of a letter providing recent patient-specific symptoms, medication, and health service use combined with guideline-based recommendations for changes in therapy on improving the quality of asthma care by inner-city primary care providers and on resultant asthma morbidity. This was a randomized, controlled clinical trial in 5- to 11-year-old children (n = 937) with moderate to severe asthma receiving health care in hospital- and community-based clinics and private practices in 7 inner-city urban areas. The caretaker of each child received a bimonthly telephone call to collect clinical information about the child's asthma. For a full year, the providers of intervention group children received bimonthly computer-generated letters based on these calls summarizing the child's asthma symptoms, health service use, and medication use with a corresponding recommendation to step up or step down medications. We measured the number and proportion of scheduled visits resulting in stepping up of medications, asthma symptoms (2-week recall), and health care use (2-month recall).
RESULTS. In this population, only a modest proportion of children whose symptoms warranted a medication increase actually had a scheduled visit to reevaluate their asthma treatment. However, in the 2-month interval after receipt of a step-up letter, 17.1% of the letters were followed by scheduled visits in the intervention group compared with scheduled visits 12.3% of the time by the control children with comparable clinical symptoms. Asthma medications were stepped up when indicated after 46.0% of these visits in the intervention group compared with 35.6% in the control group, and when asthma symptoms warranted a step up in therapy, medication changes occurred earlier among the intervention children. Among children whose medications were stepped up at any time during the 12-month study period, those in the intervention group experienced 22.1% fewer symptom days and 37.9% fewer school days missed. The intention-to-treat analysis showed no difference over the intervention year in the number of symptom days, yet there was a trend toward fewer days of limited activity and a significant decrease in emergency department visits by the intervention group compared with controls. This 24% drop in emergency department visits resulted in an intervention that was cost saving in its first year.
CONCLUSIONS. Patient-specific feedback to inner-city providers increased scheduled asthma visits, increased asthma visits in which medications were stepped up when clinically indicated, and reduced emergency department visits.
Key Words: asthma • clinical trial • provider feedback • access to care • inner city
Abbreviations: PCP—primary care provider • ED—emergency department • ICAS—Inner-City Asthma Study • CATI—computer-assisted telephone interview • NAEPP—National Asthma Education and Prevention Program
Despite the development and dissemination of guidelines for the diagnosis and management of asthma, a gap remains between current recommendations and actual practice.1,2 Undertreatment contributes to the disproportionate morbidity experienced by inner-city children with asthma.3,4 Factors associated with undertreatment include the presence of barriers to timely and continuous follow-up care,5,6 underreporting of symptoms,7 and suboptimal adherence to guideline-based asthma care by inner-city primary care providers (PCPs).3,8 Because of the high rate of emergency department (ED) use for acute asthma care and the suboptimal communication from the ED, there is a lack of timely information about the patient's asthma status.5
Most strategies developed to reduce barriers to asthma care are aimed at the patient and not the physician. Interventions, such as advising patients to return for follow-up care,9 asthma education,10 or provision of asthma coaching,11 encourage or facilitate the efforts of the patient or family to obtain appropriate preventive and follow-up care. These programs have had limited impact and are frequently expensive and time consuming.
Strategies to change physician performance, such as academic detailing, treatment reminders, printed educational materials, and didactic educational meetings, have also met with varied success, often require considerable time, and usually affect relatively few providers.12,13 Combined strategies have been more successful than single-faceted interventions.14
There are few interventions designed to change physician behavior that specifically target asthma.15–18 In particular, little information is available on the impact of education interventions on physicians caring for patients with asthma in poor urban areas where resources are suboptimal and barriers to delivering quality care are especially challenging.19
The Inner-City Asthma Study (ICAS) feedback intervention was based on the premise that the lack of timely information regarding patient symptoms and medication use and poor adherence to asthma guidelines contribute to delays in follow-up care, increased morbidity, and increased frequency of exacerbations and ED use. Unlike most interventions to reduce barriers to care, this program addressed the health care provider rather than the patient or family. In contrast to other interventions to change physician behavior, it required little effort or time on the part of the provider. We conducted a randomized, controlled trial in children with asthma to test whether timely feedback mailed to the PCPs regarding their patient's clinical status, accompanied by guideline-based treatment recommendations, influenced clinician-based scheduling of preventive care, ED visits, and asthma morbidity.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Potential participants were identified for recruitment from inpatient units of hospitals, EDs, and community pediatric clinics. ICAS recruited 937 children with moderate-to-severe asthma from centers in Boston, Massachusetts; Bronx, New York; Chicago, Illinois; Dallas, Texas; New York, New York; Seattle/Tacoma, Washington; and Tucson, Arizona. Eligibility was limited to residents of census tracts in which 
20% of households had incomes below the federal poverty level except in Seattle, where participants could be enrolled if they met Medicaid eligibility. Other inclusion criteria included a history of 1 hospitalization or 2 unscheduled visits for asthma in the previous 6 months and a positive allergy skin test to 1 of 11 indoor allergens. Children were excluded if they made 2 visits to an asthma specialist or asthma clinic in the previous 6 months or if they had any other serious chronic illness. The institutional review board at each site approved the study. Written informed consent was obtained from the caretakers of the participants.
A baseline clinical evaluation included questionnaires on asthma morbidity and the home environment. Allergy skin testing was performed using the percutaneous MultiTest method (MultiTest II, Lincoln Diagnostics, Decatur, IL). Approximately 3 weeks after the baseline clinical examination, a baseline home evaluation was performed involving both direct visual inspection and dust collection from the child's bedroom.
Intervention
Group assignments were randomly preassigned to study identification numbers by the coordinating center using a random number generator with a uniform distribution and blocks of size 8 and 12 within the site. Group assignments were supplied to sites in opaque envelopes and labeled with sequential study identification numbers, which were opened by the site interviewers on determination of the child's eligibility. Neither study staff nor participants were blinded to group assignment. At study enrollment, families were asked to identify the child's PCP. If they did not have a PCP, they were referred to one from lists of providers in the hospital service area before study initiation. If the child had not seen a PCP in the last 6 months, the study staff offered to make an appointment regardless of group assignment.
Every 2 months for 1 year from the date of enrollment, each child's caretaker underwent a standardized computer-assisted telephone interview (CATI) to determine the child's asthma symptoms and use of controller and reliever medications in the past 2 weeks, and health service use (scheduled visits, ED visits, and hospitalizations) over the previous 2-month interval. These interviews were conducted by a centralized service for all of the study sites, and the interviewers were blinded to study group assignment.
For children in the intervention group, information from each CATI call was used to generate a feedback letter that was mailed directly to the child's PCP. Although study staff and participants were aware of group assignments, they were not aware of the content of the letter. The computer-generated letter, designed with input from provider focus groups, was a single page that displayed a color photograph of the child, identifying information, and a current telephone number. A summary box detailed the child's symptoms and use of controller and quick-relief medications over the previous 2 weeks and the number of ED visits and hospitalizations over the previous 2 months, as reported by the caretaker. Based on reported symptoms, health care use and medication use, a computer algorithm developed from the severity classification of the National Asthma Education and Prevention Program (NAEPP) guidelines generated a 1-sentence recommendation for treating the child.1 The possible recommended actions based on the NAEPP guidelines were to step up, step down, or to make no changes in medications. To facilitate prescribing, each letter was accompanied by a single-page enclosure summarizing the NAEPP asthma severity classification and therapy guidelines on one side and recommended medication doses on the other.
A study investigator met with the PCP of the child in the intervention group before any letters were mailed to explain the nature of the intervention, review a sample of the bimonthly letter, and provide a copy of the NAEPP guidelines. For a few providers, the study information was given over the telephone and the printed educational information mailed. A letter about the study and educational materials were mailed to any PCP that could not be contacted. Letters were not sent to the providers of children in the control group. For this group, the information from the CATI calls was used to determine what recommendation would have been generated had the child been in the intervention group.
Outcome Measures
The quality of care outcomes were scheduled visits and changes in medications. We counted scheduled visits that occurred after CATI calls that revealed asthma symptoms warranting a step-up letter. Changes in medications were determined from the CATI call after a scheduled visit. Step up in medications was defined as an increase from no antiinflammatory use to any antiinflammatory use or from occasional to daily antiinflammatory use. We surveyed PCPs in a telephone interview at the end of the intervention to determine actions taken based on the letters and obstacles to implementation of recommendations.
Patient outcomes were the maximum number of symptom days, ED visits and hospitalizations for asthma, and school days missed because of asthma. The sample size was determined on the basis of being able to detect a minimum difference of 0.5 symptom days per 2 weeks with 80% power. Maximum symptom days were defined as the largest value among the following 3 variables: days with wheezing, tightness in the chest, or cough; nights with disturbed sleep because of asthma; and days when the child had to slow down or discontinue play activities because of asthma.
The direct medical costs of the intervention were calculated for scheduled and unscheduled visits, ED visits, hospitalizations, and medication use. The monetary value was determined by applying the mean Medicaid reimbursement for a specific service or medication. The costs of the intervention were estimated by using a simple cost allocation method that summed costs for program implementation, personnel, and materials. We estimated that the attempts to reach the family by telephone, administering the questionnaire, and entering the information into the computer would take 40 minutes per patient. We included cost for supplies and informational materials for the PCP. Intervention costs were included in the direct medical costs. The probability of cost saving was based on a Monte Carlo simulation, which takes repeated draws from the distribution of the estimated mean effects and SEs to determine the distribution of the incremental cost-effectiveness ratio estimates.21
Statistical Analysis
The intention-to-treat analysis on symptom outcomes used a mixed linear model, adjusting for site, baseline morbidity, and repeated observations per participant. Use of health care services was calculated over the entire intervention year and analyzed using analysis of covariance. Because the intervention did not start until after the first 2 months of morbidity data were available for the generation of the first letter, the 1-year outcome was based on calls 2 to 7 (months 4–14). Baseline morbidity was calculated as the average of the baseline visit and the first call. The intention-to-treat analysis included all of the randomly assigned study participants. For subsequent analyses, those children in the control group whose PCP also cared for intervention children were considered to be "contaminated" and were removed.
Analyses were performed to examine the effect of physician feedback on the quality of care, changes in the child's asthma morbidity, and health service use. The effect of the feedback letters on quality of care outcomes (scheduled visits and medication changes) was analyzed using a logistic model with a generalized estimating equation approach to account for the correlation across 2-month reporting intervals on the same children.22 The PCPs of 114 control children also cared for intervention children and received feedback letters. These children were considered "contaminated," because their PCP was exposed to the intervention. The 379 call intervals for these children and an additional 60 intervals during which children had changed their PCP were excluded from the quality of care analysis.
Children in both intervention and control groups with persistent symptoms would likely receive asthma care during the course of the intervention year. However, we expected that children in the intervention group would receive this care earlier, resulting in fewer symptom days over the year. We, therefore, performed a subgroup analysis, limiting the comparison to children whose symptoms resulted in a scheduled visit and whose medications were increased in order to test the impact of the intervention in treated patients.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
3. The characteristics of the clinicians receiving the feedback letters are shown in Table 2. The initial contact with the clinician by the study investigator was through a face-to-face visit for 309 clinicians (71%), by telephone for 58 (13%), and by letter for 68 (16%). The median duration of the face-to-face visits was 15 minutes.
|
Feedback to PCP
The PCP of a child in the intervention group received 
6 letters during the intervention year. With 471 children in the intervention group, a maximum of 2826 letters could have been sent. Caretakers were reached in 93% of the CATI calls; 97% of families were reached for 4 of the 6 calls, and only 4 families were not reached for any follow-up calls. These CATI calls resulted in 2621 letters, of which 2489 were sent, and 132 could not be sent because of a change in the PCP that could not be integrated into the intervention process in a timely manner. The average time between the CATI interview and mailing the letter to the PCP was 12.5 (±6.3) days.
Of the letters sent, 58.7% recommended stepping up medications (Table 3). Although PCPs of children in the control group received no letters, the distribution of the type of letter that was warranted based on the caretaker's report of current symptoms was similar to those sent to the PCPs of the intervention group. Of the entire cohort, 92% required a step up in medication at least once during the intervention year.
|
Intervention group physicians received 1332 letters recommending a step-up of medications. In the subsequent 2-month interval after receipt of the letter, 17.1% of the letters were followed by scheduled medical visits. In contrast, after telephone interviews in the control group that clinically warranted stepped-up therapy (1117), families scheduled follow-up visits 12.3% of the time (P = .005). Among intervention group children, 46.0% of such visits led to a medication step up in the immediate 2-month interval compared with only 35.6% for the control group children (P = .03).
Impact of Provider Feedback on Asthma Outcomes
During the intervention year, children in the feedback group had significantly fewer ED visits compared with controls (0.87 vs 1.14 per year; P = .013; Table 4). However, the maximum number of symptom days and the number of school days missed were similar in the 2 groups. The number of days with limitation of activity for more than half a day tended to be lower in the feedback group (P < .1).
|
|
40 minutes per child to reach the caretaker and make the assessment call, enter the data, and mail the letter. In calculating the costs, we used an hourly wage of $15 for a clerical employee. There were 6 calls per child per year resulting in a cost of $60. We estimated $10 for supplies and informational materials for the PCP. Because some PCPs had >1 child in the study, the cost for these materials on a per child basis was $9.20. The intervention was estimated to cost $69.20 per child over the year. When this cost was added to the cost of health services use for the year by intervention children and compared with the cost of health service use by control children, there was a savings of $337.00 per child in the intervention group. The Monte Carlo simulations, using the observed distributions of symptom days and resource use, showed that the intervention had a 97% chance of being cost saving. |
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite the fact that the patients in the study had moderate-to-severe asthma, at baseline, antiinflammatory medications were prescribed less often than ß-agonist therapy. Furthermore, a large majority (92%) of this cohort required a step-up in medication at least once during the intervention year. The finding that inner-city children with asthma are generally undertreated confirms previous reports of lack of adherence to current guidelines and underscores the need for improvement in provider practice.4,23,24
Previous studies have documented that feedback interventions can improve medical practice, although the effects are typically small.25,26 Providing feedback to the PCP regarding a specific patient outcome, such as blood pressure, has been more successful in changing physician performance than general feedback regarding the care that the provider gives.27 The ICAS intervention, a unique application of patient-specific feedback, was designed specifically to incorporate current behavioral change principles. Local opinion leaders, the study investigators, introduced and provided a copy of the NAEPP guidelines to the PCP and discussed general issues of asthma management. The bimonthly feedback letter contained information to encourage and facilitate contact with the patient. It provided the PCP with key elements of the asthma history as recommended in the guidelines for a follow-up visit. It also gave treatment recommendations tailored to the patients' symptoms and a list of medications and doses to enhance provider self-efficacy. Green et al28 stressed the concepts of predisposing, enabling, and reinforcing in efforts to change physician behavior. The support and reinforcement provided by the feedback letters may have been, in part, responsible for the more timely increase in medications of the intervention group. Providing educational materials and including face-to-face encounters with opinion leaders to increase physician knowledge were predisposing factors toward behavior change. The availability of the opinion leader and the guidelines addressed the physician's self-confidence in treating asthma. The repetitive, computerized feedback letters and educational materials were practice-enabling and reinforcing strategies.
This intervention provided feedback to the PCP in the form of patient morbidity data. Undertreatment was addressed by linking individual symptoms to recommendations for treatment in published guidelines. Given the short time available for office visits, the feedback letter allowed for more efficient doctor-patient encounters. In inner cities, care is often fragmented, and children typically receive care for asthma exacerbations at sites other than their PCP's office, frequently in EDs.29 The letter facilitated the timely transfer of information to the PCP. By reducing barriers to obtaining information necessary for asthma care, it enhanced the likelihood that the PCP would engage in best practice. Unlike systems that require complicated computer interaction and are seldom used, the simple, timely feedback letter used in this study is straightforward and requires little effort on the part of the PCP.30
Of note is that the ICAS feedback intervention targeted all of the clinicians who had patients enrolled in the study but required neither lengthy training nor changes in practice organization. They practiced in hospital-based and community clinics or private offices in cities across the United States and are representative of the spectrum of inner-city providers. The clinicians were engaged only because patients identified them as their PCP and not because they volunteered to be part of an education program. This differs from the interactive seminars reported by Brown et al31 that had an overall impact on physician prescribing and communication behavior but actually engaged only a limited proportion of contacted physicians.16 The ICAS feedback intervention also contrasts with those interventions in pediatric clinics that require considerable staff time and for which the impacts rely on major organizational changes in the clinics.32 By efficiently reaching more providers, the ICAS intervention resulted in cost savings despite the fact that only a modest proportion of patients that required a step-up of medications actually had their medications changed. This intervention may be particularly cost-effective in a managed care setting, where there is more supervision of provider activities, and responses to the letters can be more easily incorporated into daily practice.
A limitation of the study is that we may have underestimated stepping-up of medications. We were only able to assess the addition of a medication or the change from occasional to daily use. If the patient was already taking antiinflammatory medications, we were unable to determine changes in the daily dose. Dose changes would more likely have occurred in the intervention group rather than the control group, because the letter made the physician aware of the patients' increasing symptoms. Another limitation is in our ability to determine the cost of this intervention. We were unable to estimate the cost for the opinion leader to introduce the study to health care providers, because this cost likely depends on the nature of the health care delivery system involved. In a managed care organization or in pediatric outpatient clinics, which are part of voluntary or public hospitals, there would likely be no cost for the specialist to introduce the program, whereas in a public health clinic system, there would be a charge, because the opinion leader would not likely be a member of the staff of the organization.
In addition to underestimating step-ups that did occur, we may also have underestimated the impact of the letters. When providers were contacted at the end of the study, one reason they mentioned for not stepping up medications was that the patient's symptoms did not warrant stepping up medications by the time the patient came in for the visit.
The source of information regarding the care provided was the patient's caregiver, not clinical records or direct observations of the encounter. The caregiver's memory of the encounter may have been faulty, resulting in misclassification of the quality of care. However, we based this approach on a pilot study in which we used audiotapes of asthma treatment as the gold standard and revealed that parental recall of the physician's asthma care was nearly twice as accurate as the documentation found in clinical records.33
Many barriers exist along the path from the identification of new therapeutic modalities to the delivery of quality care for asthma. They include financial, system, and provider barriers; familial psychosocial burdens; cultural issues; language discordance; and poor continuity of care.34–37 Despite these barriers, which disproportionately impact asthma care for inner-city children, this intervention was successful in changing provider behavior and reducing ED asthma visits and associated health care costs.
|
ACKNOWLEDGMENTS |
|---|
The Inner-City Asthma Study was a collaboration of the following institutions and investigators: Boston University School of Medicine, Boston, MA: G. O'Connor (principal investigator), S. Steinbach, A. Zapata, J. Casagrande, and L. Schneider (Children's Hospital, Boston, MA); Albert Einstein College of Medicine/Jacobi Medical Center, Bronx, NY: E. Crain (principal investigator), L. Bauman, Y. Senturia, and D. Rosenstreich; Children's Memorial Hospital, Chicago, IL: R. Evans III (principal investigator), J. Pongracic, A. Sawyer, and K. Koridek; University of Texas Southwestern Medical Center, Dallas, TX: R. S. Gruchalla (principal investigator), V. Gan, Y. Coyle, and N. F. Gorham; Mount Sinai School of Medicine, New York, NY: M. Kattan (principal investigator), C. Lamm, M. Lippmann, E. Luder, M. Chassin, and G. Xanthos; University of Washington School of Medicine and Public Health, Seattle, WA: J. Stout (principal investigator), G. Shapiro, L. Liu, J. Koenig, M. Lasley, S. Randels, and H. Powell; University of Arizona College of Medicine, Tucson, AZ: W. Morgan (principal investigator), P. Enright, J. Goodwin, and T. Garcia, El Rio Health Clinic (Tucson, AZ); Data Coordinating Center, Rho, Inc, Chapel Hill, NC: H. Mitchell (principal investigator), M. Walter, H. Lynn, S. Hart, W. Tolbert, and E. Nuebler; Allergen Assay Laboratories, Harvard School of Public Health, Boston, MA: H. Burge, M. Muilenberg, and D. Gold; Johns Hopkins Dermatology, Allergy and Clinical Immunology (DACI) Reference Laboratory, Johns Hopkins University School of Medicine, Baltimore, MD: R. Hamilton; National Institute of Allergy and Infectious Diseases, Bethesda, MD: M. Plaut, E. Smartt, and K. Adams; National Institute of Environmental Health Sciences, Research Triangle Park, NC: G. Malindzak and P. Mastin.
We extend particular thanks to Dr Dennis Wallace for statistical consultation, Dr Sally Stearns for economic analyses, and William Tolbert and Sheri Hart for data management.
|
FOOTNOTES |
|---|
Address correspondence to Meyer Kattan, MD, Department of Pediatrics, Mount Sinai School of Medicine, Box 1202B, One Gustave L. Levy Place, New York, NY 10029. E-mail: meyer.kattan{at}mssm.edu
Financial Disclosure: Dr Steinbach has received lecture fees from GlaxoSmithKline and consulting fees from Aventis; Dr Gruchalla is a member of the GlaxoSmithKline Allergy Fellowship Grant review committee; Dr Morgan has received consulting fees from Genentech; and Dr O'Connor is GlaxoSmithKline-Data Safety and Monitoring Board chair and Astellas Pharma-Data Safety and Monitoring Board chair.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
Deceased. 