- Email: [email protected]
Assessment of a University-Based Outpatient Asthma Education Program for Children
ARTICLE
ARTICLE IN PRESS
Assessment of a UniversityBased Outpatient Asthma Education Program for Children Ling Wang, PhD, Suzanne Timmer, MD, & Kenneth Rosenman, MD ABSTRACT Introduction: To assess the effect of a pediatric asthma intervention program on reducing asthma morbidity. Methods: Study eligibility criteria included aged less than 18 years and at least two office visits for asthma in the previous year. Patients were randomly assigned to either the control or intent to intervene group. The intervention included home visits and education on the basic pathophysiology of asthma, self-management techniques, modification of asthma triggers, and proper use of asthma medications by a certified nurse educator. Results: Using simple randomization, 901 eligible pediatric patients with asthma were assigned; 458 to the control and 443 to the intent to intervene group. Of the 443 patients randomized to the intent to intervene group, 271 received the asthma education intervention. Most of the remaining 172 patients in the intent to intervene group did not receive the intervention owing to not having an appointment during the study period. Only 27 families allowed a home visit. After controlling for the difference in sex, children in the intent to intervene group had significantly less total clinic visits (incidence rate ratio Ling Wang, Assistant Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Suzanne Timmer, Assistant Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Kenneth Rosenman, Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Conflicts of interest: None to report. Correspondence: Ling Wang, PhD, Michigan State University, 909 Wilson Road, West Fee Hall, Room 120, East Lansing, MI 48864; e-mail: [email protected]. J Pediatr Health Care. (2019) 00, 1−8 0891-5245/$36.00 Copyright © 2019 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.pedhc.2019.09.004
www.jpedhc.org
[IRR] = 0.53, p < .01), and steroid bursts (IRR = 0.47, p < .01) than controls. Discussion: The implementation of a pediatric asthma education program decreased both the total clinic visits and the need for steroid bursts consistent with better asthma control. We demonstrated the benefit of a dedicated asthma educator in university-based community practice and recommend this intervention be considered a standard of care for children with asthma in all health-care settings. J Pediatr Health Care. (2019) XX, 1−8
KEY WORDS Asthma, child health, health education
INTRODUCTION Asthma is a chronic disease of the respiratory system, which affects 22 million Americans annually (National Heart, Lung, and Blood Institute, 2007), including 6 million children. It is estimated that 10.9% of children and 11.5% of adults in Michigan have asthma (Health Risk Behaviors within the State of Michigan, 2015). Sixty percent of both children and adults with asthma in Michigan have poorly controlled asthma leading to reduced school or work attendance, poor quality of life, and increased medical costs from emergency room visits and hospitalizations (Callo, Dinh, Fussman, & Wahl, 2008). The National Asthma Education and Prevention Program has issued guidelines on asthma care that include assessment, monitoring, patient education, and control of factors affecting asthma severity (National Heart, Lung, and Blood Institute, 2007). There are at least 19 studies that have assessed the effectiveness of asthma education and/or self-management and nine that have assessed reduction of triggers to improve quality of life and reduce medical costs (Labre, Herman, Dumitru, Valenzuela, & Cechman, 2012). The evidence for the effectiveness of a home-based multicomponent intervention is most compelling for children (Dumitru, 2011). A review of nine outpatient asthma self-education management and 17 home-based 000 2019
1
ARTICLE IN PRESS intervention studies found the majority reported a positive return on investment (Hsu, Wilhelm, Lewis, & Herman, 2016). Home-based and school-based environmental assessments and modifications for asthma triggers are encouraged as important elements of asthma management programs by the U.S. Environmental Protection Agency (2019). National standards for asthma self-management education have been developed to meet the requirements of Current Procedural Terminology codes needed for reimbursement (Gardner et al., 2015). This quality-care improvement project was designed as a randomized case-control intervention to assess the effectiveness of a multicomponent asthma intervention in a population of children in reference to the standard care received. The health-care organization had not instituted a formal education/asthma management program across its multiple clinics that cared for children with asthma. A randomized methodology was selected as being the strongest approach to assess the possible benefits. METHODS The methodology was a randomized intervention/control quality improvement project with comparison made before and after intervention and with the control group (Ogrinc et al., 2008). The aims of the proposal were to compare the number of clinic visits, emergency department (ED) visits, urgent care visits, hospitalizations, and spirometry results before and after intervention. The short-term goal of this project was to demonstrate the capability to set up and initiate a quality improvement program for the management of patients with asthma. The long-term goal was to develop a program for self-education management as a reimbursable Current Procedural Terminology code 98960-98962 (Gardner et al., 2015). Cohort Selection Between April 1, 2014 and March 31, 2015, patients with asthma were identified in the Michigan State University (MSU) clinic system located in the Lansing, Michigan metropolitan area. There are five clinics in the system that care for children with asthma: three pediatric, one family practice, and one pediatric pulmonary clinic. Patients were included in the asthma quality-care improvement project if they had at least two office visits in the billing database during the previous year with a diagnosis of asthma (ICD -9 code 493.0-493.9) listed as one of the diagnostic codes and aged less than 18 years. Although asthma was listed as a diagnostic code, it was not necessarily the reason for the clinic visit. There were no exclusion criteria; all patients meeting the criteria were included (Figure 1). A random number generator was applied to a listing of eligible patients to assign them either to the control or intent to intervene group. Reviewing the medical charts after randomization, it was noted that 29 patients with asthma diagnosis codes were described in their medical record as having reactive airway disease. These patients were considered to have asthma and were not excluded. The nurse educator checked the clinic schedule for the upcoming week before approaching the parent/guardian at 2
Volume 000 Number 000
the planned visit. The nurse educator notified the patient’s health-care provider via the electronic medical record (EMR) that their patient would be approached at their next clinic visit and offered asthma education. The patient’s parent/guardian was not approached to participate if the patient’s health-care provider notified the asthma educator that they did not want the patient to participate. The patient’s parent/guardian provided verbal agreement for their child to receive the intervention. Intervention As part of the intervention, patients and their families were offered a home visit to assess the presence of environmental asthma triggers. Each participant’s family received a 66-page handout. The intervention potentially included multiple home visits to evaluate possible environmental triggers and education on the basic pathophysiology of asthma, self-management techniques, modification of asthma triggers, and proper use of asthma medications. The education was provided by a registered nurse, who was certified by the National Asthma Education Board as an asthma educator. The nurse tailored her conversation with the family/patient based on her review of the medical record, questions asked by the family, and age of the child. Content varied by willingness of the family to talk and ask questions. The encounter could be brief in the clinic or more extensive as part of a home visit. The nurse educator provided feedback to the patient’s health-care provider by writing a note in the EMR on all encounters she had with the patient/patient’s guardian. A board-certified occupational, environmental, and internal medicine physician met with the asthma education nurse on a regular basis to review patient cases. Interventions took place over 20 months from March 1, 2016 to December 31, 2017. Data Collection In the eleventh month of the project, charts were abstracted by two nursing students who were not involved in the intervention. The nursing students abstracted medical data from both the control and intent to intervene group charts for one year before the date of intervention. For those in the intent to intervene group, this date was the date the nurse educator met with the patient and parent/guardian. For control patients and patients who were randomized to the intent to intervene group but never received intervention, April 1, 2016 was arbitrarily used as the cutoff date for evaluating the before and after medical data. The nursing students abstracted the following information from the EMR for the year before and after the intervention date or April 1, 2016 as appropriate: asthma severity/control, the total number of general clinic visits, unscheduled visits for asthma, ED visits, urgent care visits, intensive care unit (ICU) admissions, hospitalizations, use and type of respiratory medication, number of steroid bursts, and predicted percent of forced expiratory volume in one second (FEV1%). The records from the two local hospitals were also accessed for information regarding emergency visits
Journal of Pediatric Health Care
ARTICLE IN PRESS FIGURE 1. Study FLOW diagram.
(This figure appears in color online at www.jpedhc.org) and hospitalizations. Asthma severity/control was assessed using the Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma Summary Report 2007 (National Heart, Lung, and Blood Institute, 2007). Based on whether the first clinic visit reviewed was the patient’s initial asthma or a follow-up visit, we used the age-appropriate Figures 3−4a, b, or c to classify severity and Figures 3−5a, b, or c to classify control depending on whether the first clinic visit reviewed was the patient’s initial asthma or a follow-up visit. All outcomes were based on information abstracted from medical records. Statistical Analysis Based on randomization, patients were divided into two groups: Group 1−controls, patients receiving usual care; and Group 2−intent to intervene group, patients randomized to receive intervention. Based on participation in the educational intervention, Group 2 was divided into two groups: Group 3−intervention group, patients in the intent to intervene group that actually received intervention; and
www.jpedhc.org
Group 4−not intervened among intent to intervene group, patients in the intent to intervene group who did not receive intervention. Continuous variables were reported as mean and standard deviation and categorical variables as count and percentage. Each variable was compared between the control and intent to intervene groups using t test for continuous variables and chi-square or Fisher’s exact test for categorical variables. The demographics of the control and intent to intervene groups were compared. The main outcome was the comparison of the control (Group 1) and intent to intervene group (Group 2). However, we also evaluated the outcome among patients who actually received the asthma education, labeled the intervention group (Group 3). For comparison, we also examined those patients randomized for intervention who did not receive intervention (Group 4). Because there were many zeros in clinical visits, a Poisson regression analysis was used to compare the number of visits expressed as either zero, one, or greater than one between the intent to intervene and control groups before and after
000 2019
3
ARTICLE IN PRESS intervention. We controlled for severity/control of asthma, sex, and patient age in this regression. The incidence rate ratio (IRR) of the interaction term of group times the indicator of before/after intervention captured the difference of visits between the groups before and after intervention. All statistical analysis was carried out by SAS version 9.4. This project was approved for funding by the administrators of the MSU clinic system and approved by the MSU Institutional Review Board as a quality improvement project that did not require a signed consent form. RESULTS A total of 901 patients were randomized with 458 in the control group (Group 1) and 443 in the intent to intervene group (Group 2) based on simple randomization (each patient had an equal probability to be assigned to the control or treatment group). Of the study group patients, 172 did not receive the intervention because of one of the following reasons: 126 (73%) patients did not have a follow-up clinic visit in the allotted time frame; 22 (13%) patients noted in the clinic records to have left the practice; 13 (8%) patients’ guardians refused; 10 (6%) providers refused; and one (0.1%) patient died where the cause of death was asthma (Figure 1). Of the 271 patients who received the educational intervention (Group 3), 100% received the educational material and met with asthma educator in clinic at least twice (13 met with asthma educator in clinic three times), and 28 of the 271 (10.3%) allowed a home visit (three had two home visits). Table 1 shows the demographics of the 458 control (Group 1) and the 443 intent to intervene (Group 2) patients. The groups were similar in mean age, type of clinic, and asthma severity. However, there were significantly more females in the intent to intervene group. The majority of the patients received their asthma care in a pediatric pulmonary office. Table 1 also displays the demographics of control patients (Group 1) compared with patients who actually received the asthma education (Group 3). The two groups were similar in age, sex, and type of clinic that managed their asthma. However, there were significantly more patients with moderate and severe asthma who received asthma education (Group 3) compared with the control patients (Group 1), which is consistent with the fact that those patients who were sicker presented for care more often than those who were well controlled, and therefore had a greater chance of being contacted by the nurse educator. The individuals randomized to receive intervention but who did not receive intervention were similar to those who received intervention except they were classified at lower severity/better control (p < .01). Table 2 shows asthma morbidity in the control (Group 1) and intent to intervene patients (Group 2) before and after intervention. The intent to intervene patients (Group 2) had significantly more total clinic visits before the intervention (p < .01) compared with the control patients (Group 1). The groups were similar for the number of unscheduled asthma visits, urgent care visits, total hospitalizations, and ICU admissions before the intervention. The median values for 4
Volume 000 Number 000
each of these variables were not statistically different between the groups either before or after intervention. All medians were one, except five that were two and one that was 2.5. The control patients (Group 1) had significantly more emergency department visits before the intervention, whereas the intent to intervene patients (Group 2) had significantly more patients who required steroid bursts. In addition, the intent to intervene patients (Group 2) had an average percent predicted FEV1 of 93.9%, which was significantly lower than the 95.5% of the control patients (Group 1; p = .016). The finding of more steroid bursts and decreased FEV1% suggest that the asthma of the intent to intervene patients (Group 2) was not as well controlled as the control patients (Group 1). After the intervention, the groups had similar numbers of total clinic visits, unscheduled asthma visits, urgent care visits, ED visits, hospitalizations, ICU admissions, intubations, and steroid bursts (Table 2). Although, the magnitude of difference in predicted FEV1% after the intervention was the same as before intervention, there were fewer patients (387) with these parameters in both groups after the study compared with the 788 before the intervention and thus, the difference in the predicted FEV1% between the two groups was no longer statistically significant (Table 2). Table 3 compares similar measures of morbidity as in Table 2 except that the control patients (Group 1) are compared with patients who actually received the asthma education intervention (Group 3). Although the patients who actually received the asthma education (Group 3) had more severe asthma than control patients (Group 1) after controlling for age, sex, and discrepancy in the severity of asthma, there were significantly less total clinic visits (IRR = 0.49, p < .01), urgent care visits (IRR = 0.33, p < .01), and steroid bursts (IRR = 0.44, p < .01) in patients who received asthma education (Group 3) compared with control patients (Group 1). Table 4 reports the estimates of interaction term of group times the indicator of before/after intervention for the intent to intervene patients (Group 2) versus the control patients (Group 1). Controlling for patient sex, there were significantly less total clinic visits (IRR = 0.53, p < .01) and steroid bursts (IRR = 0.47, p < .01) after the intervention in the intent to intervene patients (Group 2) in comparison with the control patients (Group 1). Table 4 also reports before/after intervention results for patients who actually received intervention (Group 3) versus control patients (Group 1). The magnitude of changes were greater than when we compared the intent to intervene patients (Group 2) with the control patients (Group 1; Table 4). A comparison among those patients who received intervention (Group 3) and those patients in the intent to treat group who did not receive intervention (Group 4) showed similar results as the Group 3 versus Group 1 comparisons. Only 28 (10%) of those patients who received the asthma education group had a home visit by the asthma nurse educator. The other parent/guardians of those who received the asthma education group declined a home visit.
Journal of Pediatric Health Care
ARTICLE IN PRESS TABLE 1. Comparing demographics in control (Group 1), intent to intervene (Group 2), and intervened patients (Group 3) before intervention Characteristic
Sex Male Female Mean ages, years (SD) Type of clinic Pediatrics Family medicine Pediatric pulmonary Missing Severity/Control Unknown Mild intermittent Mild persistent Moderate persistent Severe
Control (Group 1), n = 458
Intent to intervene (Group 2), n = 443
Intervened (Group 3), n = 271
Not intervened among intent to intervene (Group 4), n = 172
288 (62.9) 170 (37.1) 10.8 (3.6)
244 (55.1)* 199 (44.9) 11.1 (3.7)
151 (55.7) 120 (44.3) 11.1 (3.5)
93 (54.1) 79 (45.9) 11.1 (3.9)
154 (33.6) 3 (0.6) 294 (64.2) 7 (1.5)
148 (33.4) 2 (0.5) 288 (65) 5 (1.1)
106 (39.1) 1 (0.4) 164 (60.5) 0 (0)
42 (27.0) 1 (0.6) 124 (72.1) 5 (3.0)
67 (14.6) 82 (17.9) 176 (38.5) 110 (24.0) 23 (5.0)
64 (14.4) 71 (16.0) 168 (25.3) 112 (25.3) 28 (6.3)
14 (5.2)** 36 (13.3) 112 (41.3) 84 (31) 25 (9.2)
50 (29.1) 35 (20.4) 56 (32.6) 28 (16.3) 3 (1.7)
Note. Values are n (%), unless otherwise specified. *p = .02; Sex distribution is significantly different between Group 1 and 2 patients. **p < .01; Severity/Control distribution is significantly different between Group 1 and 3 patients.
DISCUSSION There are many published studies in pediatric patients with asthma which have demonstrated decreased morbidity after an education-based intervention (Dor et al., 2018; Karnick
et al., 2007; Kelly et al., 2000; Kercsmar et al., 2017; Largo, Borgialli, Wisinski, Wahl, & Priem, 2011; Rapp et al., 2018; Shani et al., 2015). The majority of these involved urban, low income, predominantly Medicaid populations. Our
TABLE 2. Measures of asthma morbidity in the control (Group 1) and intent to intervene (Group 2) patients before and after intervention Control (Group 1), n = 458
Measure
Before intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1a (SD) After intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1 (SD)
Intent to intervene (Group 2), n = 443
0
1
>1
0
1
>1
299 (65.3) 318 (69.4)
114 (24.9) 63 (13.8)
45 (9.8) 77 (16.8)
239 (54.0) 295 (66.6)
154 (34.8) 68 (15.4)
50 (11.3) 80 (18.1)
.01 .65
404 (88.2) 388 (84.7) 440 (96.1) 451 (98.5) 389 (84.9)
36 (7.9) 45 (9.8) 14 (3.1) 5 (1.1) 44 (9.6) 95.5 (9.2), n = 393
18 (3.3) 25 (5.5) 4 (1.4) 2 (0.4) 25 (5.5)
386 (87.1) 389 (87.8) 421 (95.0) 436 (98.4) 345 (77.9)
42 (9.5) 45 (10.2) 16 (3.6) 6 (1.4) 65 (14.8) 93.9 (9.7) n = 395
15 (3.4) 9 (2.0) 6 (1.35) 1 (0.2) 33 (7.5)
.64 .03 .70 .81 .02 .02
300 (65.5) 321 (70.1)
102 (22.3) 78 (17.0)
56 (12.2) 59 (12.9)
322 (72.7) 336 (75.9)
39 (8.8) 56 (12.6)
.06 .05
37 (8.1) 19 (4.2) 26 (5.7) 14 (3.1) 5 (1.1) 2 (0.4) 1 (0.22) 1 (0.22) 67 (14.7) 31 (6.8) 95.2 (16.1), n = 211
393 (88.7) 412 (93) 435 (98.2) 440 (99.3) 364 (83.1)
11 (2.5) 9 (2.0) 0 (0) 0 (0) 24 (5.5)
.36 .55 .26 .36 .22 .24
402 (87.8) 418 (91.3) 451 (98.5) 456 (99.6) 359 (78.6)
82 (18.5) 51 (11.5) 39 (8.8) 22 (5.0) 8 (1.8) 3 (0.68) 50 (11.4) 93.4 (13.8), n = 176
p value
Note. Values are n (%), unless otherwise specified. a % predicted FEV1 are reported as mean values in Groups 1 and 2 that had FEV1 values tested. ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit.
www.jpedhc.org
000 2019
5
ARTICLE IN PRESS TABLE 3. Measures of asthma morbidity in the control (Group 1) and intervened (Group 3) patients before and after intervention Control (Group 1), n = 458
Measure
Before intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1a (SD) After intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1 (SD)
Intervened (Group 3), n = 271
0
1
>1
0
1
>1
299 (65.3) 318 (69.4)
114 (24.9) 63 (13.8)
45 (9.8) 77 (16.8)
115 (42.4) 166 (61.3)
112 (41.3) 47 (17.3)
44 (16.2) 58 (21.4)
< .01 .08
404 (88.2) 388 (84.7) 440 (96.1) 451 (98.5) 389 (84.9)
36 (7.9) 45 (9.8) 14 (3.1) 5 (1.1) 44 (9.6) 95.5 (9.2), n = 393
18 (3.3) 25 (5.5) 4 (1.4) 2 (0.4) 25 (5.5)
232 (85.6) 234 (86.4) 257 (95.6) 269 (99.3) 182 (67.2)
27 (10.0) 12 (4.4) 31 (11.4) 6 (2.2) 10 (3.7) 4 (1.5) 2 (0.7) 0 (0) 59 (21.8) 30 (11.1) 93.7 (10.1), n = 255
.58 .10 .67 .49 < .01 .02
300 (65.5) 321 (70.1)
102 (22.3) 78 (17.0)
56 (12.2) 59 (12.9)
185 (66.5) 192 (70.9)
37 (8.1) 19 (4.2) 26 (5.7) 14 (3.1) 5 (1.1) 2 (0.4) 1 (0.22) 1 (0.22) 67 (14.7) 31 (6.8) 95.2 (16.1), n = 211
250 (92.3) 249 (91.9) 265 (97.8) 269 (99.3) 210 (77.8)
402 (87.8) 418 (91.3) 451 (98.5) 456 (99.6) 359 (78.6)
55 (20.3) 37 (13.7)
31 (11.4) 42 (15.5)
17 (6.3) 4 (1.5) 14 (5.2) 8 (3.0) 6 (2.2) 0 (0) 2 (0.7) 0 (0) 38 (14.1) 22 (8.2) 93.6 (13.1), n = 156
P value
.74 .35 .08 .95 .27 .42 .78 .29
Note. Values are n (%), unless otherwise specified. a % predicted FEV1 are reported as mean values in the Groups 1 and 3 that had FEV1 values tested. ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit.
study is one of the few in a university-based community practice, which demonstrated a decrease in acute care services for pediatric patients with asthma after an education program. We were able to show this decrease in acute asthma care services despite the fact that only 10% of the intervention group allowed a home visit to assess possible environmental triggers. Although approximately 64% of the children with asthma in our system are treated by specialists in a pediatric pulmonology clinic, they still benefited from the educational intervention. Examples of Previously Published Studies Five examples of the studies with largely Medicaid-covered patients are summarized here. Rapp et al. (2018) found a decrease in ED visits for asthma from 41% to 21% and fewer asthma symptoms after an education intervention. The education was provided by a certified asthma educator in a Louisiana clinic setting at both baseline and 12-month follow-up visits and included an environmental evaluation by a community health worker (Rapp et al., 2018). The study did not provide data regarding the number of environmental evaluations or what asthma triggers were identified. The majority of the 187 children had less severe disease than those in our study with 70% having mild intermittent asthma. Twenty percent of the participants’ primary insurance was private or self-pay with the remainder being Medicaid. The Community Healthcare for Asthma Management and Prevention of Symptoms study in Arizona, Michigan, and Puerto Rico analyzed the cost-effectiveness of an 6
Volume 000 Number 000
evidence-based childhood asthma intervention (Dor et al., 2018). No information about the content of the intervention was provided. These patients were in Medicaid managed care organizations. They noted that the Community Healthcare for Asthma Management and Prevention of Symptoms study reduced asthma symptoms in pediatric patients by 29.75 days per child year, which equated to a cost savings of $28.76 per symptom free day. Unlike our results, a decrease in ED visits after a pediatric asthma intervention has been reported (Kercsmar et al., 2017). This intervention consisted of hospital-based care, outpatient care enhancements, and community-based support and resulted in a decrease in ED visits from 21.5 to 12.4, which was sustained over the subsequent 12 months. The inpatient asthma management for these patients focused on identifying and mitigating triggers for asthma exacerbations. It included aftercare with five in-home visits. The outpatient enhancements included case management for all patients with one or more asthma related hospitalizations or two or more ED visits in the past year. The community support focused on building relationships with local school nurses to increase completion of an asthma control test and to ensure documentation of an asthma control plan. The population for this study was similar to our study with only about a third being insured by Medicaid. Karnick et al. (2007) described a decrease in hospitalizations, ED visits, and clinic visits in a study of children with asthma from an inner-city area of Chicago. They compared outcomes before and after a single 20−30-minute asthma
Journal of Pediatric Health Care
www.jpedhc.org
b
% predicted FEV1
a
Regression analysis controls for sex as a covariate. Regression analysis controls for severity/control as a covariate. CI, confidence interval; ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit; IRR, incidence rate ratio.
.14 .29 < .01 .25 .37 .11 .10 p value .48
p value IRR (95% CI)
0.72 (0.46−1.12) 1.23 (0.84−1.82) 0.16 (0.08−0.31) 1.63 (0.70−3.81) 2.21 (0.38−12.6) 10.0 (0.58−170.7) 0.52 (0.24−1.15) Estimates (95% CI) 2.03 ( 3.53−7.59) < .01 .34 < .01 .66 .47 .49 < .01 p value .93 < .01 .14 .27 .92 .20 .48 < .01 p value .87
p value IRR (95% CI)
0.49 (0.37−0.65) 0.88 (0.69−1.14) 0.33 (0.19−0.57) 1.13 (0.66−1.95) 0.66 (0.21−2.07) 2.20 (0.24−20.4) 0.44 (0.31−0.63) Estimates (95% CI) 0.13 ( 2.92−3.18)
p value IRR (95% CI)
0.53 (0.41−0.68) 0.84 (0.66−1.06) 0.79 (0.51−1.20) 0.98 (0.60−1.59) 0.51 (0.18−1.42) 0.55 (0.11−2.86) 0.47 (0.33−0.66) Estimates (95% CI) 0.24 ( 3.07−2.60) Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts
Intervened (Group 3) versus control (Group 1) patientsb Intent to intervene (Group 2) versus control (Group 1) patientsa Measure
TABLE 4. Regression results of measures of asthma morbidity before and after intervention
Intervened (Group 3) versus not intervened among intent to intervene (Group 4) patientsb
ARTICLE IN PRESS education program, a reinforced asthma education program, and the reinforced asthma education program plus case management. The estimated cost savings from the intervention were » $4,000 per child. Kelly et al. (2000) had reported similar results among pediatric patients with asthma enrolled in Medicaid in Virginia. Shani et al. (2015) assessed the effectiveness of an innercity home-based asthma education program of 132 children in an environmental justice community and also found a significant reduction in ED visits. The actual number of homes visited were 41 of the 179 because of families dropping out of the study and incomplete surveys. We also found a reluctance of parents/guardians to allow a home visit, which may have limited our ability to decrease a patient’s environmental triggers. Our asthma nurse educator generally provided comprehensive asthma education in two office visits (85.2%) with only 13 patients having more than two encounters (4.8%). Despite the limited number of encounters our asthma educator had with guardians/patients, we still found that the intervention was effective. Of the seven published outpatient pediatric asthma education programs, ours is one of two whose patients had predominantly private insurance. One of the studies did not specify the type of insurance, but the population was identified as low income (Shani et al., 2015). The other studies involved patients whose primary payer was Medicaid. Because of the fee structure of the pediatric patients in the MSU clinics, there was no direct financial benefit to the MSU clinics to employ a health educator. To make hiring an asthma educator viable, it is essential to be able to bill insurance companies for this service. The cost benefits accrue to the insurance company, and studies such as ours can be used to convince the insurance company to reimburse for this service. Limitations One limitation of our study was the number of patients who were randomized to the intent to intervene group but who did not receive the asthma education intervention owing to the lack of an appointment during the study period. The group who attended clinic had more severe asthma but also were presumably more adherent to their treatment regimen because they came more frequently to clinic and might be more adherent to the educational intervention. However, Table 4 demonstrates that after controlling for asthma severity, the asthma education intervention was effective. A second limitation was that if the health-care provider in clinic did not record an urgent care visit, or the patient with asthma was not treated for an exacerbation at either of the two local hospitals, we would not know about these events. However, there is no reason to believe that events among intervention patients were less likely to be recorded or that they were more likely to seek care at other than the two local hospitals. The Health Homes University home-based environmental intervention and education program for families with pediatric asthma in Michigan also had 34% attrition rate (Reddy, Gomez, & Dixon, 2017). This study population was in the same city as ours but used three home visits over a
000 2019
7
ARTICLE IN PRESS six-month period. Medicaid was the primary payer for 81% of the study patients, whereas in our study, it only accounted for 29% of the patients in the control group and those who received the asthma education. Insurance information was not abstracted in our project on the patients who were randomized to the study group but did not receive the asthma education intervention. The Health Homes University found a 47% decrease in unscheduled health-care visits for asthma (Shani et al., 2015). One reported factor, which affected dropout attrition in a pediatric asthma intervention study, was younger age of caregiver (Zebracki et al., 2003). Other published pediatric asthma intervention studies have noted large study dropout rates ranging up to 75% (Wood, Tumiel-Berhalter, Owen, Taylor, & Kattan, 2006). CONCLUSION Asthma remains a significant contributor to childhood morbidity and use of health care in this country. There was a discrepancy in sex between the control and intent to intervene groups. However, after controlling for sex, there was a decrease in total clinic visits and the need for steroid bursts after the intervention. The effect was even greater in those patients who actually received asthma education, as there was also a decrease in urgent care visits. These results show that patients in the intent to intervene group benefited from an asthma education program. Although most families did not allow for a home evaluation, education in the clinic alone still demonstrated a measurable improvement in their asthma. Based on our results, even a modest intervention can improve clinically important outcomes. The benefit of this intervention was not limited to clinics predominantly caring for children with Medicaid insurance. Our results showed that a dedicated asthma educator also reduced morbidity from asthma in a predominantly pulmonary pediatric university-based community practice. We acknowledge the hard work and dedication of the asthma nurse educator, Amy Transue, RN. The Michigan State University Health team funded this quality improvement project. SUPPLEMENTARY MATERIALS Supplementary material associated with this article can be found in the online version at https://doi.org/10.1016/j. pedhc.2019.09.004. REFERENCES Health risk behaviors within the State of Michigan. (2015) Retrieved from http://www.michigan.gov/mdhhs/0,5885,7339-71550_5104_5279_39424-134600−,00.htm Callo, S. L., Dinh, P., Fussman, C., & Wahl, R. (2008). Asthma control and clinical management in children and adults. Retrieved from http://www.michigan.gov/brfswww.michigan.gov/asthma Dor, A., Luo, Q., Gerstein, M. T., Malveaux, F., Mitchell, H., & Markus, A. R. (2018). Cost-effectiveness of an evidencebased childhood asthma intervention in real-world primary care settings. Journal of Ambulatory Care Management, 41, 213–224. Dumitru, G. G. (2011). Recommendations from the task force on community preventive services to decrease asthma
8
Volume 000 Number 000
morbidity through home-based, multi-trigger, multicomponent interventions. American Journal of Preventive Medicine, 41 (Suppl. 2S1), s1–s4. Gardner, A., Kaplan, B., Brown, W., Krier-Morrow, D., Rappaport, S., & Marcus, L. (2015). National standards for asthma self-management education. Annals of Allergy, Asthma and Immunology, 114, 178–186.e1. Hsu, J., Wilhelm, N., Lewis, L., & Herman, E. (2016). Economic evidence for US asthma self-management education and homebased interventions. Journal of Allergy and Clinical Immunology. in Practice, 4, 1123–1134.e27. Karnick, P., Margellos-Anast, H., Seals, G., Whitman, S., Aljadeff, G., & Johnson, D. (2007). The pediatric asthma intervention: A comprehensive cost-effective approach to asthma management in a disadvantaged inner-city community. Journal of Asthma, 44, 39–44. Kelly, C. S., Morrow, A. L., Shults, J., Nakas, N., Strope, G. L., & Adelman, R. D. (2000). Outcomes evaluation of a comprehensive intervention program for asthmatic children enrolled in Medicaid. Pediatrics, 105, 1029–1035. Kercsmar, C. M., Beck, A. F., Sauers-Ford, H., Simmons, J., Wiener, B., Crosby, L., . . . Mansour, M. (2017). Association of an asthma improvement collaborative with health care utilization in Medicaid-insured pediatric patients in an urban community. JAMA Pediatrics, 171, 1072–1080. Labre, M. P., Herman, E. J., Dumitru, G. G., Valenzuela, K. A., & Cechman, C. L. (2012). Public health interventions for asthma: An umbrella review, 1990−2010. American Journal of Preventive Medicine, 42, 403–410. Largo, T. W., Borgialli, M., Wisinski, C. L., Wahl, R. L., & Priem, W. F. (2011). Healthy homes university: A home-based environmental intervention and Education Program for families with pediatric asthma in Michigan. Public Health Reports, 126 (Suppl. 1), S14–S26. National Heart, Lung, and Blood Institute. (2007). National Heart, Lung, and Blood Institute National Asthma Education and Prevention Program Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma Full Report 2007. Retrieved from https://www.nhlbi.nih.gov/sites/default/files/ media/docs/asthgdln_1.pdf Ogrinc, G., Mooney, S. E., Estrada, C., Foster, T., Goldmann, D., Hall, L. W., . . . Watts, B. (2008). The Squire (Standards for QUality Improvement Reporting Excellence) guidelines for quality improvement reporting: Explanation and elaboration. Quality and Safety in Health Care, 17(Suppl. 1), i13–i32. Rapp, K. I., Jack, L., Wilson, C., Hayes, S. C., Post, R., McKnight, E., & Malveaux, F. (2018). Improving asthma-related outcomes among children participating in the head-off environmental asthma in Louisiana (HEAL), Phase II study. Health Promotion Practice, 19, 233–239. Reddy, A. L., Gomez, M., & Dixon, S. L. (2017). An evaluation of a state-funded healthy homes intervention on asthma outcomes in adults and children. Journal of Public Health Management and Practice, 23, 219–228. Shani, Z., Scott, R. G., Schofield, L. S., Johnson, J. H., Williams, E. R., Hampton, J., & Ramprasad, V. (2015). Effect of a home intervention program on pediatric asthma in an environmental justice community. Health Promotion Practice, 16, 291–298. U. S. Environmental Protection Agency. (2019). Indoor Air Quality (IAQ). Retrieved from https://www.epa.gov/indoor-air-quality-iaq Wood, P., Tumiel-Berhalter, L., Owen, S., Taylor, K., & Kattan, M. (2006). Implementation of an asthma intervention in the inner city. Annals of Allergy, Asthma and Immunology, 97, S20–S24. Zebracki, K., Drotar, D., Kirchner, H. L., Schluchter, M., Redline, S., Kercsmar, C., & Walders, N. (2003). Predicting attrition in a pediatric asthma intervention study. Journal of Pediatric Psychology, 28, 519–528.
Journal of Pediatric Health Care
ARTICLE IN PRESS
Assessment of a UniversityBased Outpatient Asthma Education Program for Children Ling Wang, PhD, Suzanne Timmer, MD, & Kenneth Rosenman, MD ABSTRACT Introduction: To assess the effect of a pediatric asthma intervention program on reducing asthma morbidity. Methods: Study eligibility criteria included aged less than 18 years and at least two office visits for asthma in the previous year. Patients were randomly assigned to either the control or intent to intervene group. The intervention included home visits and education on the basic pathophysiology of asthma, self-management techniques, modification of asthma triggers, and proper use of asthma medications by a certified nurse educator. Results: Using simple randomization, 901 eligible pediatric patients with asthma were assigned; 458 to the control and 443 to the intent to intervene group. Of the 443 patients randomized to the intent to intervene group, 271 received the asthma education intervention. Most of the remaining 172 patients in the intent to intervene group did not receive the intervention owing to not having an appointment during the study period. Only 27 families allowed a home visit. After controlling for the difference in sex, children in the intent to intervene group had significantly less total clinic visits (incidence rate ratio Ling Wang, Assistant Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Suzanne Timmer, Assistant Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Kenneth Rosenman, Professor of Medicine, Department of Medicine, College of Human Medicine, Michigan State University, East Lansing, MI. Conflicts of interest: None to report. Correspondence: Ling Wang, PhD, Michigan State University, 909 Wilson Road, West Fee Hall, Room 120, East Lansing, MI 48864; e-mail: [email protected]. J Pediatr Health Care. (2019) 00, 1−8 0891-5245/$36.00 Copyright © 2019 by the National Association of Pediatric Nurse Practitioners. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.pedhc.2019.09.004
www.jpedhc.org
[IRR] = 0.53, p < .01), and steroid bursts (IRR = 0.47, p < .01) than controls. Discussion: The implementation of a pediatric asthma education program decreased both the total clinic visits and the need for steroid bursts consistent with better asthma control. We demonstrated the benefit of a dedicated asthma educator in university-based community practice and recommend this intervention be considered a standard of care for children with asthma in all health-care settings. J Pediatr Health Care. (2019) XX, 1−8
KEY WORDS Asthma, child health, health education
INTRODUCTION Asthma is a chronic disease of the respiratory system, which affects 22 million Americans annually (National Heart, Lung, and Blood Institute, 2007), including 6 million children. It is estimated that 10.9% of children and 11.5% of adults in Michigan have asthma (Health Risk Behaviors within the State of Michigan, 2015). Sixty percent of both children and adults with asthma in Michigan have poorly controlled asthma leading to reduced school or work attendance, poor quality of life, and increased medical costs from emergency room visits and hospitalizations (Callo, Dinh, Fussman, & Wahl, 2008). The National Asthma Education and Prevention Program has issued guidelines on asthma care that include assessment, monitoring, patient education, and control of factors affecting asthma severity (National Heart, Lung, and Blood Institute, 2007). There are at least 19 studies that have assessed the effectiveness of asthma education and/or self-management and nine that have assessed reduction of triggers to improve quality of life and reduce medical costs (Labre, Herman, Dumitru, Valenzuela, & Cechman, 2012). The evidence for the effectiveness of a home-based multicomponent intervention is most compelling for children (Dumitru, 2011). A review of nine outpatient asthma self-education management and 17 home-based 000 2019
1
ARTICLE IN PRESS intervention studies found the majority reported a positive return on investment (Hsu, Wilhelm, Lewis, & Herman, 2016). Home-based and school-based environmental assessments and modifications for asthma triggers are encouraged as important elements of asthma management programs by the U.S. Environmental Protection Agency (2019). National standards for asthma self-management education have been developed to meet the requirements of Current Procedural Terminology codes needed for reimbursement (Gardner et al., 2015). This quality-care improvement project was designed as a randomized case-control intervention to assess the effectiveness of a multicomponent asthma intervention in a population of children in reference to the standard care received. The health-care organization had not instituted a formal education/asthma management program across its multiple clinics that cared for children with asthma. A randomized methodology was selected as being the strongest approach to assess the possible benefits. METHODS The methodology was a randomized intervention/control quality improvement project with comparison made before and after intervention and with the control group (Ogrinc et al., 2008). The aims of the proposal were to compare the number of clinic visits, emergency department (ED) visits, urgent care visits, hospitalizations, and spirometry results before and after intervention. The short-term goal of this project was to demonstrate the capability to set up and initiate a quality improvement program for the management of patients with asthma. The long-term goal was to develop a program for self-education management as a reimbursable Current Procedural Terminology code 98960-98962 (Gardner et al., 2015). Cohort Selection Between April 1, 2014 and March 31, 2015, patients with asthma were identified in the Michigan State University (MSU) clinic system located in the Lansing, Michigan metropolitan area. There are five clinics in the system that care for children with asthma: three pediatric, one family practice, and one pediatric pulmonary clinic. Patients were included in the asthma quality-care improvement project if they had at least two office visits in the billing database during the previous year with a diagnosis of asthma (ICD -9 code 493.0-493.9) listed as one of the diagnostic codes and aged less than 18 years. Although asthma was listed as a diagnostic code, it was not necessarily the reason for the clinic visit. There were no exclusion criteria; all patients meeting the criteria were included (Figure 1). A random number generator was applied to a listing of eligible patients to assign them either to the control or intent to intervene group. Reviewing the medical charts after randomization, it was noted that 29 patients with asthma diagnosis codes were described in their medical record as having reactive airway disease. These patients were considered to have asthma and were not excluded. The nurse educator checked the clinic schedule for the upcoming week before approaching the parent/guardian at 2
Volume 000 Number 000
the planned visit. The nurse educator notified the patient’s health-care provider via the electronic medical record (EMR) that their patient would be approached at their next clinic visit and offered asthma education. The patient’s parent/guardian was not approached to participate if the patient’s health-care provider notified the asthma educator that they did not want the patient to participate. The patient’s parent/guardian provided verbal agreement for their child to receive the intervention. Intervention As part of the intervention, patients and their families were offered a home visit to assess the presence of environmental asthma triggers. Each participant’s family received a 66-page handout. The intervention potentially included multiple home visits to evaluate possible environmental triggers and education on the basic pathophysiology of asthma, self-management techniques, modification of asthma triggers, and proper use of asthma medications. The education was provided by a registered nurse, who was certified by the National Asthma Education Board as an asthma educator. The nurse tailored her conversation with the family/patient based on her review of the medical record, questions asked by the family, and age of the child. Content varied by willingness of the family to talk and ask questions. The encounter could be brief in the clinic or more extensive as part of a home visit. The nurse educator provided feedback to the patient’s health-care provider by writing a note in the EMR on all encounters she had with the patient/patient’s guardian. A board-certified occupational, environmental, and internal medicine physician met with the asthma education nurse on a regular basis to review patient cases. Interventions took place over 20 months from March 1, 2016 to December 31, 2017. Data Collection In the eleventh month of the project, charts were abstracted by two nursing students who were not involved in the intervention. The nursing students abstracted medical data from both the control and intent to intervene group charts for one year before the date of intervention. For those in the intent to intervene group, this date was the date the nurse educator met with the patient and parent/guardian. For control patients and patients who were randomized to the intent to intervene group but never received intervention, April 1, 2016 was arbitrarily used as the cutoff date for evaluating the before and after medical data. The nursing students abstracted the following information from the EMR for the year before and after the intervention date or April 1, 2016 as appropriate: asthma severity/control, the total number of general clinic visits, unscheduled visits for asthma, ED visits, urgent care visits, intensive care unit (ICU) admissions, hospitalizations, use and type of respiratory medication, number of steroid bursts, and predicted percent of forced expiratory volume in one second (FEV1%). The records from the two local hospitals were also accessed for information regarding emergency visits
Journal of Pediatric Health Care
ARTICLE IN PRESS FIGURE 1. Study FLOW diagram.
(This figure appears in color online at www.jpedhc.org) and hospitalizations. Asthma severity/control was assessed using the Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma Summary Report 2007 (National Heart, Lung, and Blood Institute, 2007). Based on whether the first clinic visit reviewed was the patient’s initial asthma or a follow-up visit, we used the age-appropriate Figures 3−4a, b, or c to classify severity and Figures 3−5a, b, or c to classify control depending on whether the first clinic visit reviewed was the patient’s initial asthma or a follow-up visit. All outcomes were based on information abstracted from medical records. Statistical Analysis Based on randomization, patients were divided into two groups: Group 1−controls, patients receiving usual care; and Group 2−intent to intervene group, patients randomized to receive intervention. Based on participation in the educational intervention, Group 2 was divided into two groups: Group 3−intervention group, patients in the intent to intervene group that actually received intervention; and
www.jpedhc.org
Group 4−not intervened among intent to intervene group, patients in the intent to intervene group who did not receive intervention. Continuous variables were reported as mean and standard deviation and categorical variables as count and percentage. Each variable was compared between the control and intent to intervene groups using t test for continuous variables and chi-square or Fisher’s exact test for categorical variables. The demographics of the control and intent to intervene groups were compared. The main outcome was the comparison of the control (Group 1) and intent to intervene group (Group 2). However, we also evaluated the outcome among patients who actually received the asthma education, labeled the intervention group (Group 3). For comparison, we also examined those patients randomized for intervention who did not receive intervention (Group 4). Because there were many zeros in clinical visits, a Poisson regression analysis was used to compare the number of visits expressed as either zero, one, or greater than one between the intent to intervene and control groups before and after
000 2019
3
ARTICLE IN PRESS intervention. We controlled for severity/control of asthma, sex, and patient age in this regression. The incidence rate ratio (IRR) of the interaction term of group times the indicator of before/after intervention captured the difference of visits between the groups before and after intervention. All statistical analysis was carried out by SAS version 9.4. This project was approved for funding by the administrators of the MSU clinic system and approved by the MSU Institutional Review Board as a quality improvement project that did not require a signed consent form. RESULTS A total of 901 patients were randomized with 458 in the control group (Group 1) and 443 in the intent to intervene group (Group 2) based on simple randomization (each patient had an equal probability to be assigned to the control or treatment group). Of the study group patients, 172 did not receive the intervention because of one of the following reasons: 126 (73%) patients did not have a follow-up clinic visit in the allotted time frame; 22 (13%) patients noted in the clinic records to have left the practice; 13 (8%) patients’ guardians refused; 10 (6%) providers refused; and one (0.1%) patient died where the cause of death was asthma (Figure 1). Of the 271 patients who received the educational intervention (Group 3), 100% received the educational material and met with asthma educator in clinic at least twice (13 met with asthma educator in clinic three times), and 28 of the 271 (10.3%) allowed a home visit (three had two home visits). Table 1 shows the demographics of the 458 control (Group 1) and the 443 intent to intervene (Group 2) patients. The groups were similar in mean age, type of clinic, and asthma severity. However, there were significantly more females in the intent to intervene group. The majority of the patients received their asthma care in a pediatric pulmonary office. Table 1 also displays the demographics of control patients (Group 1) compared with patients who actually received the asthma education (Group 3). The two groups were similar in age, sex, and type of clinic that managed their asthma. However, there were significantly more patients with moderate and severe asthma who received asthma education (Group 3) compared with the control patients (Group 1), which is consistent with the fact that those patients who were sicker presented for care more often than those who were well controlled, and therefore had a greater chance of being contacted by the nurse educator. The individuals randomized to receive intervention but who did not receive intervention were similar to those who received intervention except they were classified at lower severity/better control (p < .01). Table 2 shows asthma morbidity in the control (Group 1) and intent to intervene patients (Group 2) before and after intervention. The intent to intervene patients (Group 2) had significantly more total clinic visits before the intervention (p < .01) compared with the control patients (Group 1). The groups were similar for the number of unscheduled asthma visits, urgent care visits, total hospitalizations, and ICU admissions before the intervention. The median values for 4
Volume 000 Number 000
each of these variables were not statistically different between the groups either before or after intervention. All medians were one, except five that were two and one that was 2.5. The control patients (Group 1) had significantly more emergency department visits before the intervention, whereas the intent to intervene patients (Group 2) had significantly more patients who required steroid bursts. In addition, the intent to intervene patients (Group 2) had an average percent predicted FEV1 of 93.9%, which was significantly lower than the 95.5% of the control patients (Group 1; p = .016). The finding of more steroid bursts and decreased FEV1% suggest that the asthma of the intent to intervene patients (Group 2) was not as well controlled as the control patients (Group 1). After the intervention, the groups had similar numbers of total clinic visits, unscheduled asthma visits, urgent care visits, ED visits, hospitalizations, ICU admissions, intubations, and steroid bursts (Table 2). Although, the magnitude of difference in predicted FEV1% after the intervention was the same as before intervention, there were fewer patients (387) with these parameters in both groups after the study compared with the 788 before the intervention and thus, the difference in the predicted FEV1% between the two groups was no longer statistically significant (Table 2). Table 3 compares similar measures of morbidity as in Table 2 except that the control patients (Group 1) are compared with patients who actually received the asthma education intervention (Group 3). Although the patients who actually received the asthma education (Group 3) had more severe asthma than control patients (Group 1) after controlling for age, sex, and discrepancy in the severity of asthma, there were significantly less total clinic visits (IRR = 0.49, p < .01), urgent care visits (IRR = 0.33, p < .01), and steroid bursts (IRR = 0.44, p < .01) in patients who received asthma education (Group 3) compared with control patients (Group 1). Table 4 reports the estimates of interaction term of group times the indicator of before/after intervention for the intent to intervene patients (Group 2) versus the control patients (Group 1). Controlling for patient sex, there were significantly less total clinic visits (IRR = 0.53, p < .01) and steroid bursts (IRR = 0.47, p < .01) after the intervention in the intent to intervene patients (Group 2) in comparison with the control patients (Group 1). Table 4 also reports before/after intervention results for patients who actually received intervention (Group 3) versus control patients (Group 1). The magnitude of changes were greater than when we compared the intent to intervene patients (Group 2) with the control patients (Group 1; Table 4). A comparison among those patients who received intervention (Group 3) and those patients in the intent to treat group who did not receive intervention (Group 4) showed similar results as the Group 3 versus Group 1 comparisons. Only 28 (10%) of those patients who received the asthma education group had a home visit by the asthma nurse educator. The other parent/guardians of those who received the asthma education group declined a home visit.
Journal of Pediatric Health Care
ARTICLE IN PRESS TABLE 1. Comparing demographics in control (Group 1), intent to intervene (Group 2), and intervened patients (Group 3) before intervention Characteristic
Sex Male Female Mean ages, years (SD) Type of clinic Pediatrics Family medicine Pediatric pulmonary Missing Severity/Control Unknown Mild intermittent Mild persistent Moderate persistent Severe
Control (Group 1), n = 458
Intent to intervene (Group 2), n = 443
Intervened (Group 3), n = 271
Not intervened among intent to intervene (Group 4), n = 172
288 (62.9) 170 (37.1) 10.8 (3.6)
244 (55.1)* 199 (44.9) 11.1 (3.7)
151 (55.7) 120 (44.3) 11.1 (3.5)
93 (54.1) 79 (45.9) 11.1 (3.9)
154 (33.6) 3 (0.6) 294 (64.2) 7 (1.5)
148 (33.4) 2 (0.5) 288 (65) 5 (1.1)
106 (39.1) 1 (0.4) 164 (60.5) 0 (0)
42 (27.0) 1 (0.6) 124 (72.1) 5 (3.0)
67 (14.6) 82 (17.9) 176 (38.5) 110 (24.0) 23 (5.0)
64 (14.4) 71 (16.0) 168 (25.3) 112 (25.3) 28 (6.3)
14 (5.2)** 36 (13.3) 112 (41.3) 84 (31) 25 (9.2)
50 (29.1) 35 (20.4) 56 (32.6) 28 (16.3) 3 (1.7)
Note. Values are n (%), unless otherwise specified. *p = .02; Sex distribution is significantly different between Group 1 and 2 patients. **p < .01; Severity/Control distribution is significantly different between Group 1 and 3 patients.
DISCUSSION There are many published studies in pediatric patients with asthma which have demonstrated decreased morbidity after an education-based intervention (Dor et al., 2018; Karnick
et al., 2007; Kelly et al., 2000; Kercsmar et al., 2017; Largo, Borgialli, Wisinski, Wahl, & Priem, 2011; Rapp et al., 2018; Shani et al., 2015). The majority of these involved urban, low income, predominantly Medicaid populations. Our
TABLE 2. Measures of asthma morbidity in the control (Group 1) and intent to intervene (Group 2) patients before and after intervention Control (Group 1), n = 458
Measure
Before intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1a (SD) After intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1 (SD)
Intent to intervene (Group 2), n = 443
0
1
>1
0
1
>1
299 (65.3) 318 (69.4)
114 (24.9) 63 (13.8)
45 (9.8) 77 (16.8)
239 (54.0) 295 (66.6)
154 (34.8) 68 (15.4)
50 (11.3) 80 (18.1)
.01 .65
404 (88.2) 388 (84.7) 440 (96.1) 451 (98.5) 389 (84.9)
36 (7.9) 45 (9.8) 14 (3.1) 5 (1.1) 44 (9.6) 95.5 (9.2), n = 393
18 (3.3) 25 (5.5) 4 (1.4) 2 (0.4) 25 (5.5)
386 (87.1) 389 (87.8) 421 (95.0) 436 (98.4) 345 (77.9)
42 (9.5) 45 (10.2) 16 (3.6) 6 (1.4) 65 (14.8) 93.9 (9.7) n = 395
15 (3.4) 9 (2.0) 6 (1.35) 1 (0.2) 33 (7.5)
.64 .03 .70 .81 .02 .02
300 (65.5) 321 (70.1)
102 (22.3) 78 (17.0)
56 (12.2) 59 (12.9)
322 (72.7) 336 (75.9)
39 (8.8) 56 (12.6)
.06 .05
37 (8.1) 19 (4.2) 26 (5.7) 14 (3.1) 5 (1.1) 2 (0.4) 1 (0.22) 1 (0.22) 67 (14.7) 31 (6.8) 95.2 (16.1), n = 211
393 (88.7) 412 (93) 435 (98.2) 440 (99.3) 364 (83.1)
11 (2.5) 9 (2.0) 0 (0) 0 (0) 24 (5.5)
.36 .55 .26 .36 .22 .24
402 (87.8) 418 (91.3) 451 (98.5) 456 (99.6) 359 (78.6)
82 (18.5) 51 (11.5) 39 (8.8) 22 (5.0) 8 (1.8) 3 (0.68) 50 (11.4) 93.4 (13.8), n = 176
p value
Note. Values are n (%), unless otherwise specified. a % predicted FEV1 are reported as mean values in Groups 1 and 2 that had FEV1 values tested. ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit.
www.jpedhc.org
000 2019
5
ARTICLE IN PRESS TABLE 3. Measures of asthma morbidity in the control (Group 1) and intervened (Group 3) patients before and after intervention Control (Group 1), n = 458
Measure
Before intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1a (SD) After intervention Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts % predicted FEV1 (SD)
Intervened (Group 3), n = 271
0
1
>1
0
1
>1
299 (65.3) 318 (69.4)
114 (24.9) 63 (13.8)
45 (9.8) 77 (16.8)
115 (42.4) 166 (61.3)
112 (41.3) 47 (17.3)
44 (16.2) 58 (21.4)
< .01 .08
404 (88.2) 388 (84.7) 440 (96.1) 451 (98.5) 389 (84.9)
36 (7.9) 45 (9.8) 14 (3.1) 5 (1.1) 44 (9.6) 95.5 (9.2), n = 393
18 (3.3) 25 (5.5) 4 (1.4) 2 (0.4) 25 (5.5)
232 (85.6) 234 (86.4) 257 (95.6) 269 (99.3) 182 (67.2)
27 (10.0) 12 (4.4) 31 (11.4) 6 (2.2) 10 (3.7) 4 (1.5) 2 (0.7) 0 (0) 59 (21.8) 30 (11.1) 93.7 (10.1), n = 255
.58 .10 .67 .49 < .01 .02
300 (65.5) 321 (70.1)
102 (22.3) 78 (17.0)
56 (12.2) 59 (12.9)
185 (66.5) 192 (70.9)
37 (8.1) 19 (4.2) 26 (5.7) 14 (3.1) 5 (1.1) 2 (0.4) 1 (0.22) 1 (0.22) 67 (14.7) 31 (6.8) 95.2 (16.1), n = 211
250 (92.3) 249 (91.9) 265 (97.8) 269 (99.3) 210 (77.8)
402 (87.8) 418 (91.3) 451 (98.5) 456 (99.6) 359 (78.6)
55 (20.3) 37 (13.7)
31 (11.4) 42 (15.5)
17 (6.3) 4 (1.5) 14 (5.2) 8 (3.0) 6 (2.2) 0 (0) 2 (0.7) 0 (0) 38 (14.1) 22 (8.2) 93.6 (13.1), n = 156
P value
.74 .35 .08 .95 .27 .42 .78 .29
Note. Values are n (%), unless otherwise specified. a % predicted FEV1 are reported as mean values in the Groups 1 and 3 that had FEV1 values tested. ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit.
study is one of the few in a university-based community practice, which demonstrated a decrease in acute care services for pediatric patients with asthma after an education program. We were able to show this decrease in acute asthma care services despite the fact that only 10% of the intervention group allowed a home visit to assess possible environmental triggers. Although approximately 64% of the children with asthma in our system are treated by specialists in a pediatric pulmonology clinic, they still benefited from the educational intervention. Examples of Previously Published Studies Five examples of the studies with largely Medicaid-covered patients are summarized here. Rapp et al. (2018) found a decrease in ED visits for asthma from 41% to 21% and fewer asthma symptoms after an education intervention. The education was provided by a certified asthma educator in a Louisiana clinic setting at both baseline and 12-month follow-up visits and included an environmental evaluation by a community health worker (Rapp et al., 2018). The study did not provide data regarding the number of environmental evaluations or what asthma triggers were identified. The majority of the 187 children had less severe disease than those in our study with 70% having mild intermittent asthma. Twenty percent of the participants’ primary insurance was private or self-pay with the remainder being Medicaid. The Community Healthcare for Asthma Management and Prevention of Symptoms study in Arizona, Michigan, and Puerto Rico analyzed the cost-effectiveness of an 6
Volume 000 Number 000
evidence-based childhood asthma intervention (Dor et al., 2018). No information about the content of the intervention was provided. These patients were in Medicaid managed care organizations. They noted that the Community Healthcare for Asthma Management and Prevention of Symptoms study reduced asthma symptoms in pediatric patients by 29.75 days per child year, which equated to a cost savings of $28.76 per symptom free day. Unlike our results, a decrease in ED visits after a pediatric asthma intervention has been reported (Kercsmar et al., 2017). This intervention consisted of hospital-based care, outpatient care enhancements, and community-based support and resulted in a decrease in ED visits from 21.5 to 12.4, which was sustained over the subsequent 12 months. The inpatient asthma management for these patients focused on identifying and mitigating triggers for asthma exacerbations. It included aftercare with five in-home visits. The outpatient enhancements included case management for all patients with one or more asthma related hospitalizations or two or more ED visits in the past year. The community support focused on building relationships with local school nurses to increase completion of an asthma control test and to ensure documentation of an asthma control plan. The population for this study was similar to our study with only about a third being insured by Medicaid. Karnick et al. (2007) described a decrease in hospitalizations, ED visits, and clinic visits in a study of children with asthma from an inner-city area of Chicago. They compared outcomes before and after a single 20−30-minute asthma
Journal of Pediatric Health Care
www.jpedhc.org
b
% predicted FEV1
a
Regression analysis controls for sex as a covariate. Regression analysis controls for severity/control as a covariate. CI, confidence interval; ED, emergency department; FEV1, forced expiratory volume in one second; ICU, intensive care unit; IRR, incidence rate ratio.
.14 .29 < .01 .25 .37 .11 .10 p value .48
p value IRR (95% CI)
0.72 (0.46−1.12) 1.23 (0.84−1.82) 0.16 (0.08−0.31) 1.63 (0.70−3.81) 2.21 (0.38−12.6) 10.0 (0.58−170.7) 0.52 (0.24−1.15) Estimates (95% CI) 2.03 ( 3.53−7.59) < .01 .34 < .01 .66 .47 .49 < .01 p value .93 < .01 .14 .27 .92 .20 .48 < .01 p value .87
p value IRR (95% CI)
0.49 (0.37−0.65) 0.88 (0.69−1.14) 0.33 (0.19−0.57) 1.13 (0.66−1.95) 0.66 (0.21−2.07) 2.20 (0.24−20.4) 0.44 (0.31−0.63) Estimates (95% CI) 0.13 ( 2.92−3.18)
p value IRR (95% CI)
0.53 (0.41−0.68) 0.84 (0.66−1.06) 0.79 (0.51−1.20) 0.98 (0.60−1.59) 0.51 (0.18−1.42) 0.55 (0.11−2.86) 0.47 (0.33−0.66) Estimates (95% CI) 0.24 ( 3.07−2.60) Total clinic visits Unscheduled asthma visits Urgent care visits ED visits Hospitalizations ICU admissions Steroid bursts
Intervened (Group 3) versus control (Group 1) patientsb Intent to intervene (Group 2) versus control (Group 1) patientsa Measure
TABLE 4. Regression results of measures of asthma morbidity before and after intervention
Intervened (Group 3) versus not intervened among intent to intervene (Group 4) patientsb
ARTICLE IN PRESS education program, a reinforced asthma education program, and the reinforced asthma education program plus case management. The estimated cost savings from the intervention were » $4,000 per child. Kelly et al. (2000) had reported similar results among pediatric patients with asthma enrolled in Medicaid in Virginia. Shani et al. (2015) assessed the effectiveness of an innercity home-based asthma education program of 132 children in an environmental justice community and also found a significant reduction in ED visits. The actual number of homes visited were 41 of the 179 because of families dropping out of the study and incomplete surveys. We also found a reluctance of parents/guardians to allow a home visit, which may have limited our ability to decrease a patient’s environmental triggers. Our asthma nurse educator generally provided comprehensive asthma education in two office visits (85.2%) with only 13 patients having more than two encounters (4.8%). Despite the limited number of encounters our asthma educator had with guardians/patients, we still found that the intervention was effective. Of the seven published outpatient pediatric asthma education programs, ours is one of two whose patients had predominantly private insurance. One of the studies did not specify the type of insurance, but the population was identified as low income (Shani et al., 2015). The other studies involved patients whose primary payer was Medicaid. Because of the fee structure of the pediatric patients in the MSU clinics, there was no direct financial benefit to the MSU clinics to employ a health educator. To make hiring an asthma educator viable, it is essential to be able to bill insurance companies for this service. The cost benefits accrue to the insurance company, and studies such as ours can be used to convince the insurance company to reimburse for this service. Limitations One limitation of our study was the number of patients who were randomized to the intent to intervene group but who did not receive the asthma education intervention owing to the lack of an appointment during the study period. The group who attended clinic had more severe asthma but also were presumably more adherent to their treatment regimen because they came more frequently to clinic and might be more adherent to the educational intervention. However, Table 4 demonstrates that after controlling for asthma severity, the asthma education intervention was effective. A second limitation was that if the health-care provider in clinic did not record an urgent care visit, or the patient with asthma was not treated for an exacerbation at either of the two local hospitals, we would not know about these events. However, there is no reason to believe that events among intervention patients were less likely to be recorded or that they were more likely to seek care at other than the two local hospitals. The Health Homes University home-based environmental intervention and education program for families with pediatric asthma in Michigan also had 34% attrition rate (Reddy, Gomez, & Dixon, 2017). This study population was in the same city as ours but used three home visits over a
000 2019
7
ARTICLE IN PRESS six-month period. Medicaid was the primary payer for 81% of the study patients, whereas in our study, it only accounted for 29% of the patients in the control group and those who received the asthma education. Insurance information was not abstracted in our project on the patients who were randomized to the study group but did not receive the asthma education intervention. The Health Homes University found a 47% decrease in unscheduled health-care visits for asthma (Shani et al., 2015). One reported factor, which affected dropout attrition in a pediatric asthma intervention study, was younger age of caregiver (Zebracki et al., 2003). Other published pediatric asthma intervention studies have noted large study dropout rates ranging up to 75% (Wood, Tumiel-Berhalter, Owen, Taylor, & Kattan, 2006). CONCLUSION Asthma remains a significant contributor to childhood morbidity and use of health care in this country. There was a discrepancy in sex between the control and intent to intervene groups. However, after controlling for sex, there was a decrease in total clinic visits and the need for steroid bursts after the intervention. The effect was even greater in those patients who actually received asthma education, as there was also a decrease in urgent care visits. These results show that patients in the intent to intervene group benefited from an asthma education program. Although most families did not allow for a home evaluation, education in the clinic alone still demonstrated a measurable improvement in their asthma. Based on our results, even a modest intervention can improve clinically important outcomes. The benefit of this intervention was not limited to clinics predominantly caring for children with Medicaid insurance. Our results showed that a dedicated asthma educator also reduced morbidity from asthma in a predominantly pulmonary pediatric university-based community practice. We acknowledge the hard work and dedication of the asthma nurse educator, Amy Transue, RN. The Michigan State University Health team funded this quality improvement project. SUPPLEMENTARY MATERIALS Supplementary material associated with this article can be found in the online version at https://doi.org/10.1016/j. pedhc.2019.09.004. REFERENCES Health risk behaviors within the State of Michigan. (2015) Retrieved from http://www.michigan.gov/mdhhs/0,5885,7339-71550_5104_5279_39424-134600−,00.htm Callo, S. L., Dinh, P., Fussman, C., & Wahl, R. (2008). Asthma control and clinical management in children and adults. Retrieved from http://www.michigan.gov/brfswww.michigan.gov/asthma Dor, A., Luo, Q., Gerstein, M. T., Malveaux, F., Mitchell, H., & Markus, A. R. (2018). Cost-effectiveness of an evidencebased childhood asthma intervention in real-world primary care settings. Journal of Ambulatory Care Management, 41, 213–224. Dumitru, G. G. (2011). Recommendations from the task force on community preventive services to decrease asthma
8
Volume 000 Number 000
morbidity through home-based, multi-trigger, multicomponent interventions. American Journal of Preventive Medicine, 41 (Suppl. 2S1), s1–s4. Gardner, A., Kaplan, B., Brown, W., Krier-Morrow, D., Rappaport, S., & Marcus, L. (2015). National standards for asthma self-management education. Annals of Allergy, Asthma and Immunology, 114, 178–186.e1. Hsu, J., Wilhelm, N., Lewis, L., & Herman, E. (2016). Economic evidence for US asthma self-management education and homebased interventions. Journal of Allergy and Clinical Immunology. in Practice, 4, 1123–1134.e27. Karnick, P., Margellos-Anast, H., Seals, G., Whitman, S., Aljadeff, G., & Johnson, D. (2007). The pediatric asthma intervention: A comprehensive cost-effective approach to asthma management in a disadvantaged inner-city community. Journal of Asthma, 44, 39–44. Kelly, C. S., Morrow, A. L., Shults, J., Nakas, N., Strope, G. L., & Adelman, R. D. (2000). Outcomes evaluation of a comprehensive intervention program for asthmatic children enrolled in Medicaid. Pediatrics, 105, 1029–1035. Kercsmar, C. M., Beck, A. F., Sauers-Ford, H., Simmons, J., Wiener, B., Crosby, L., . . . Mansour, M. (2017). Association of an asthma improvement collaborative with health care utilization in Medicaid-insured pediatric patients in an urban community. JAMA Pediatrics, 171, 1072–1080. Labre, M. P., Herman, E. J., Dumitru, G. G., Valenzuela, K. A., & Cechman, C. L. (2012). Public health interventions for asthma: An umbrella review, 1990−2010. American Journal of Preventive Medicine, 42, 403–410. Largo, T. W., Borgialli, M., Wisinski, C. L., Wahl, R. L., & Priem, W. F. (2011). Healthy homes university: A home-based environmental intervention and Education Program for families with pediatric asthma in Michigan. Public Health Reports, 126 (Suppl. 1), S14–S26. National Heart, Lung, and Blood Institute. (2007). National Heart, Lung, and Blood Institute National Asthma Education and Prevention Program Expert Panel Report 3: Guidelines for the Diagnosis and Management of Asthma Full Report 2007. Retrieved from https://www.nhlbi.nih.gov/sites/default/files/ media/docs/asthgdln_1.pdf Ogrinc, G., Mooney, S. E., Estrada, C., Foster, T., Goldmann, D., Hall, L. W., . . . Watts, B. (2008). The Squire (Standards for QUality Improvement Reporting Excellence) guidelines for quality improvement reporting: Explanation and elaboration. Quality and Safety in Health Care, 17(Suppl. 1), i13–i32. Rapp, K. I., Jack, L., Wilson, C., Hayes, S. C., Post, R., McKnight, E., & Malveaux, F. (2018). Improving asthma-related outcomes among children participating in the head-off environmental asthma in Louisiana (HEAL), Phase II study. Health Promotion Practice, 19, 233–239. Reddy, A. L., Gomez, M., & Dixon, S. L. (2017). An evaluation of a state-funded healthy homes intervention on asthma outcomes in adults and children. Journal of Public Health Management and Practice, 23, 219–228. Shani, Z., Scott, R. G., Schofield, L. S., Johnson, J. H., Williams, E. R., Hampton, J., & Ramprasad, V. (2015). Effect of a home intervention program on pediatric asthma in an environmental justice community. Health Promotion Practice, 16, 291–298. U. S. Environmental Protection Agency. (2019). Indoor Air Quality (IAQ). Retrieved from https://www.epa.gov/indoor-air-quality-iaq Wood, P., Tumiel-Berhalter, L., Owen, S., Taylor, K., & Kattan, M. (2006). Implementation of an asthma intervention in the inner city. Annals of Allergy, Asthma and Immunology, 97, S20–S24. Zebracki, K., Drotar, D., Kirchner, H. L., Schluchter, M., Redline, S., Kercsmar, C., & Walders, N. (2003). Predicting attrition in a pediatric asthma intervention study. Journal of Pediatric Psychology, 28, 519–528.
Journal of Pediatric Health Care