Published ahead of print on June 13, 2003, doi:10.1164/rccm.200208-877OC
© 2003 American Thoracic Society Asthma Intervention Program Prevents Readmissions in High Healthcare UsersDepartments of Medicine and Biostatistics, Washington University School of Medicine and Barnes-Jewish Hospital, St. Louis, Missouri Correspondence and requests for reprints should be addressed to Mario Castro, M.D., M.P.H., Washington University School of Medicine, Box 8052, 660 South Euclid Avenue, St. Louis, MO 63110–1093. E-mail: castrom{at}msnotes.wustl.edu
The largest portion of the cost for asthma healthcare is due to hospitalizations. Improved methods of healthcare delivery for patients with asthma are needed to prevent readmissions. From 1996 to 1999, 96 adult subjects (predominantly young African American women) hospitalized with an asthma exacerbation, who had a history of frequent healthcare use, were randomized to an asthma nurse specialist intervention (n = 50) or a usual care group (n = 46) for 6 months. Our aim was to decrease rates of readmissions within 6 months of hospital discharge, to reduce cost, and to improve health-related quality of life. Our results demonstrate a 60% reduction in total hospitalizations (31 readmissions in the intervention group and 71 in the control group, p = 0.04), with no significant change in emergency department visits. Readmissions for asthma were reduced by 54% (21 vs. 42 in the control group; p = 0.04). We found a marked reduction in lost work or school days: 246 versus 1,040 days in the control group (p = 0.02). The intervention resulted in a substantial reduction in direct and indirect healthcare costs, saving $6,462 per patient (p = 0.03). A brief intervention program focusing on high healthcare users with asthma can result in improved asthma control and reduced hospital use with substantial cost savings.
Key Words: asthma education • healthcare costs • hospitalizations Asthma is a common chronic disease that affects 15 million people in the United States (1, 2). The economic impact of this disease has been estimated to be 11.3 billion dollars in 1998 (2). Hospitalizations account for almost half of annual direct healthcare expenditures for asthma (3). Hospitalizations for asthma disproportionately affect African Americans, who are more than three times as likely as whites to be hospitalized (1, 4). More than 80% of the resources used for asthma are consumed by 20% of the population, who are frequent users of the healthcare system (3). Interventions directed at reducing costs for this group are needed. We hypothesized that patients with a history of frequent healthcare use would benefit from a limited nurse-focused intervention. We proposed to study the effect of this intervention on the rate of readmission, lost work/school days, asthma-specific quality of life, and overall cost of healthcare.
Patients The extended version of the methods is available in the online data supplement. All patients admitted to Barnes-Jewish Hospital with the primary admitting diagnosis of asthma were screened for participation from September 1996 to July 1999. From a total of 828 hospitalized patients, 96 were eligible for randomization based on the following criteria: physician diagnosis of asthma of at least 12 months duration, 18–65 years of age, FEV1 to FVC ratio of less than 80%, and had a history of one or more hospitalizations in the previous 12 months. The study was approved by the Human Studies Committee of Washington University School of Medicine. All patients gave informed consent before entry into the study.
Study Design and Intervention The study intervention consisted of the following multifaceted approach: (1) suggestions by the nurse to the primary physician regarding potential simplification or consolidation of current regimen in accordance with the National Asthma Education and Prevention Program II (5); (2) completion of a daily "Asthma Care" flow sheet while in the hospital; (3) provision of asthma education appropriate to the patient's education, motivation, and cultural beliefs; (4) provision of psychosocial support and screening patients for professional counseling; (5) establishment of an individualized asthma self-management plan; (6) consultation with social service professionals to facilitate discharge planning; and (7) the provision of outpatient follow-up through telephone contact, home visits, and follow-up appointments with the primary physician, as necessary. Three consecutive nurses provided the intervention and collected the data for the study. All analysis was conducted with the intention-to-treat principle with the primary endpoint being readmission due to asthma within 360 days. Secondary endpoints included total readmissions, emergency department visits, quality of life, direct and indirect healthcare costs, lost school or work days, and cumulative number of days of hospitalization. Costs were estimated using cost logs that were collected contemporaneously by all patients during the 180-day enrollment period. Quality of life was measured using a disease-specific measure, the Asthma Quality of Life Questionnaire, at entry into the study and at 6 months (9, 10).
Statistical Analysis
Baseline Characteristics Baseline characteristics of study patients, including demographics, asthma history, healthcare use, and pulmonary functions, are shown in Table 1 . In general, the patients were young, African American women with a high school education or less with a history of frequent healthcare use. There were slightly more patients in the intervention group who had Medicaid and slightly more in the control group who had private insurance. In general, patients had moderate to severe airflow obstruction at entry into the study. Both groups were well balanced with respect to all baseline characteristics, and there were no significant differences between the groups.
Hospital Readmissions As shown in Table 2 , there was a 60% reduction in total readmissions in the intervention group compared with the control group over 12 months (31 vs. 71 readmissions, respectively, p = 0.04). The majority of readmissions were due to asthma (21 vs. 42 [intervention vs. control group], p = 0.04). There was a 68% reduction in hospitalizations not due to asthma (10 vs. 29 [intervention versus control group], p = 0.19). Multiple readmissions (two or more) were more frequent in the control group (15 vs. 7 in the intervention group, p = 0.03). Time to the first readmission was not significantly different between groups (Figure 1) . There was a 69% reduction in total hospital days in the intervention group compared with the control group over 12 months (p = 0.04), mostly driven by the reduction in hospital days for asthma (129 vs. 53 days, respectively, p = 0.04). There were no significant differences in the number of healthcare provider visits. There was a slightly higher number of emergency department (ED) visits in the intervention group (n = 93) than the control group (n = 64), although it was not statistically significant. However, we found one individual in the intervention group with 30 ED visits who was a clear outlier. When we recomputed the average ED visits without this one outlier in the intervention group, we found, on average, 1.3 ± 1.5 ED visits in the intervention group compared with 1.4 ± 1.5 in the control group.
To determine whether readmission rates differed during the first 6 months as compared with the second 6 months (when no intervention was being provided by the nurse), we evaluated when each of the subsequent 98 readmissions occurred. A total of 46% of readmissions in the control group and 48% of the intervention group occurred during the second 6 months (p = 0.85). This suggests that the effect of the intervention on readmissions was sustained for at least an additional 6 months. We then analyzed whether the assignment to the intervention group was associated with a decreased number of readmissions after adjusting for baseline demographic factors, lung function, and quality of life. As Table 3 shows, the strongest independent predictors of readmission were the number of previous hospitalizations and the duration of asthma. After adjusting for these variables and other univariate predictors of readmission, assignment to the control group remained a significant independent predictor of readmission. Those patients in the control group were 3.6 times more likely to be readmitted two or more times in 12 months after the index hospitalization than those patients receiving the intervention.
Healthcare Costs The healthcare costs, including direct and indirect costs, for the patients in the control and intervention groups for the 180 days after the initial hospitalization are shown in Table 4 . The average total healthcare costs per patient in the control group was $12,188 ± $19,352 compared with $5,726 ± $5,679 in the intervention group (p = 0.03), resulting in an overall savings of $6,462 per patient attributable to the intervention. The majority of savings came from a decrease in the direct healthcare costs (69% or $4,430 per patient, p = 0.07). These were primarily due to lower costs for hospitalizations for asthma in the intervention group, accounting for a cost savings of $2,955 per patient for hospitalizations (p = 0.01).
The savings in direct healthcare costs in the intervention group was not due to differences in medication refills. The total number of medication refills that occurred during the 180 days after the initial hospitalization included (control vs. intervention) inhaled corticosteroids (174 vs. 214), leukotriene modifier (39 vs. 10), long-acting ß-agonist (69 vs. 134), and theophylline (46 vs. 64) (all p > 0.05). Thirty-seven percent of all patients were on an inhaled corticosteroid on admission, and this increased to 83% at discharge. Furthermore, we found that the mean number of refills for antiinflammatory medications was 5.1 ± 4.5 in those with one or more readmission compared with 4.2 ± 4.5 in those with no readmissions (p = 0.34). Therefore, there were no significant differences between the groups in the provision of antiinflammatory treatment at discharge or at the 180-day follow-up. Indirect healthcare costs, including lost workdays and nonprofessional/caregiver help, were lower in the intervention group resulting in a cost savings of $2,220 per patient (p = 0.03). The lower indirect healthcare costs were primarily due to a 76% less lost work days. The next largest portion of indirect healthcare cost savings was due to less unpaid caregiver cost in the intervention group.
Quality of Life
Our study demonstrates that a brief nurse-directed intervention in hospitalized asthma patients with a frequent history of healthcare use results in a significant reduction in subsequent readmissions, lost work or school days, and direct and indirect healthcare costs. Previous studies have had relatively small sample sizes, selected cohorts, use of historical control subjects, or ignored the impact of such a program on healthcare costs (8–16). Some of these studies, but not all, have demonstrated a beneficial effect on healthcare use and asthma knowledge. Few studies have performed a prospective, randomized controlled trial of an asthma education program in a hospital setting (13, 17). George and colleagues studied 77 patients with asthma randomized to a nurse-directed intervention, including education and outpatient follow-up in an asthma program (13). In those patients for which information was available, they demonstrated a significant reduction in hospitalizations and emergency department visits. In a similar study, 201 children admitted to the hospital with acute asthma were provided a nurse-directed teaching program focusing on asthma home management. Subsequent follow-up demonstrated a significant reduction in readmissions (17). Finally, a study in 160 children admitted to the hospital with acute asthma provided discharge planning, which included a focused education program and self-management plan (22). Six-month follow-up demonstrated a significant reduction in readmissions and emergency department visits. Healthcare costs, the impact on school or work days, and quality of life were not available in these studies. An interesting finding in our study was a trend toward a reduction in hospitalizations not due to asthma (68% reduction) similar to that seen due to asthma (54%), although not statistically significant. This potential effect of the intervention on hospitalizations for causes other than asthma is similar to that seen in a program conducted in patients with congestive heart failure (18). This suggests that access to nursing and psychosocial support may provide benefits to healthcare problems other than those intended by the intervention. Furthermore, findings from this study suggest that a short-term intervention targeted at high healthcare users may have longer term effects. The study nurse provided asthma education and support during the initial hospitalization and only for the subsequent 6 months. However, we found no significant difference in readmission rates between the groups in the subsequent 6 months when no intervention was being provided. This effect may be due to greater asthma knowledge and self-management skills, the positive impact of the nurse–patient relationship on accessing future healthcare needs, and improving sense of control over one's health care needs (19, 20). A potential limitation of our study is the lack of generalizability of the results. Our study focused on high healthcare users in an urban academic medical center. The majority of these patients were young, African American women with asthma from lower socioeconomic strata. Only 12% of those patients admitted to our hospital with a primary diagnosis of asthma were enrolled. However, we did show a significant impact on healthcare use and days lost from work or school without compromising the patient's health or quality of life. Therefore, the potential application of this intervention to other patients with asthma should have a similar positive benefit. A second limitation of our study is the inability to identify which specific component of the intervention is most effective (21). Multiple components were delivered by the nurse during the initial hospitalization and then subsequently reinforced. We often found that the social barriers that these patients faced outweighed their ability to initially change their health behavior. Interestingly, we found no significant differences in subsequent visits to healthcare providers or in the number or type of asthma medications refilled between the groups. For example, the intervention group received 214 refills for inhaled corticosteroids compared with 174 in the "usual care" group during the 6 months of the intervention (p > 0.05). This suggest that the enhanced patient education, provision of follow-up care and medications by the nurse, and addressing social barriers appeared to be key to the success of our program as seen in other studies (13, 17, 23). A third limitation of our study is that all of the data from our participants were collected by the asthma nurse specialist conducting the intervention. This is a potential source of bias as the interviewer may exert a qualitative influence on the participant's response. We took several measures to minimize this source of bias: (1) Each nurse specialist was trained and supervised in appropriate interviewing techniques. (2) All questions regarding the study endpoints were structured and administered in exactly the same manner among participants, and (3) responses to questions were confirmed with an external source of data when possible (e.g., hospitalization or office visit record). In summary, this study demonstrates that a brief, nurse-directed intervention in asthmatic subjects with a history of frequent healthcare use will significantly reduce subsequent readmissions and decrease lost work or school days in a cost-effective manner. Recent efforts by the National Asthma Educator Certification Board (www.naecb.org) to improve the delivery of asthma education should increase the availability of certified asthma educators to deliver a similar program. We believe that widespread implementation of this approach in patients hospitalized with asthma with a history of high healthcare use will result in substantially improved patient satisfaction and reduce healthcare costs due to this common disease.
The authors acknowledge the contributions of Bev Long for her expert secretarial assistance, Michael W. Rich, M.D., and Robert Strunk, M.D., for their insightful advice and comments, and the following persons and agencies without whose support this study could not have been successfully completed: Mary C. Spencer, R.N., M.S.N., Barb Gubser, R.N., M.S.N., Betty Leventhal, ED.D., L.C.W., M.N., R.N., Lee Ann Blue, R.N., M.S.N., James Crane, M.D., Philip Korenblat, M.D., Daniel L. Hamilos, M.D., Robert Senior, M.D., Michael J. Holtzman, M.D., and Barnes-Jewish Hospital and Foundation.
Supported by the Barnes-Jewish Hospital Foundation. This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest: M.C. has no declared conflict of interest; N.A.Z. has no declared conflict of interest; S.C. has no declared conflict of interest; J.B. has no declared conflict of interest; C.L. has no declared conflict of interest; K.B.S. has no declared conflict of interest. Received in original form August 15, 2002; accepted in final form May 30, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
1.18, p < 0.001) and also improved significantly in the intervention group from 2.70 at baseline to 4.02 (
