- Email: [email protected]
Inhaled Corticosteroid Claims and Outpatient Visits After Hospitalization for Asthma Among Commercially Insured Children
Inhaled Corticosteroid Claims and Outpatient Visits After Hospitalization for Asthma Among Commercially Insured Children Annie Lintzenich Andrews, MD, MSCR; David G. Bundy, MD, MPH; Kit N. Simpson, DrPH; Ronald J. Teufel, II, MD, MSCR; Jillian Harvey, PhD; Annie N. Simpson, PhD From the Department of Pediatrics, Medical University of South Carolina College of Medicine (Drs Andrews, Bundy, and Teufel), and Department of Healthcare Leadership and Management, Medical University of South Carolina College of Health Professions (Drs K. N. Simpson, Harvey, and A. N. Simpson), Charleston, SC Conflict of Interest: The authors declare that they have no conflict of interest. Address correspondence to Annie Lintzenich Andrews, MD, MSCR, Department of Pediatrics, Medical University of South Carolina College of Medicine, 135 Rutledge Ave, MSC 561, Charleston, SC 29425 (e-mail: [email protected]). Received for publication July 20, 2016; accepted October 28, 2016.
ABSTRACT OBJECTIVE: To determine rates of inhaled corticosteroid (ICS) claims and outpatient follow-up after asthma hospitalization among commercially insured children. METHODS: We conducted a retrospective cohort analysis of children hospitalized for asthma using 2013 national Truven MarketScan data. Covariates included age, sex, region, length of stay, and having an ICS claim within 35 days before hospitalization. Outcome variables were a claim for any ICS-containing medication and outpatient visit within 30 days after discharge. Multivariable analysis used logistic regression. RESULTS: A total of 5471 children aged 2 to 17 were included; 61% were boys, and mean age was 6.8 years. Forty-one percent had a claim for an ICS and 76% had an outpatient visit within 30 days after hospital discharge. In multivariable analysis, children who had an ICS claim within 35 days before the hospitalization were more likely to have an ICS claim within 30 days after discharge (relative risk [RR] 1.3, 95% confidence interval
[CI] 1.2–1.5). The strongest predictor of an ICS claim within 30 days after discharge was attendance at an outpatient appointment (RR 1.4, 95% CI 1.3–1.6). Children aged 2 to 6 were more likely to attend an outpatient appointment (RR 1.1, 95% CI 1.1–1.2). Children with an ICS claim before admission were also more likely to attend an outpatient appointment (RR 1.1, 95% CI 1.004–1.1). CONCLUSIONS: In this national sample of commercially insured children with asthma, rates of ICS claims after hospitalization are low despite high rates of outpatient visits. Both inpatient and outpatient physicians must play a role in increasing ICS adherence in this high-risk population of children with asthma.
WHAT’S NEW
children, including those with asthma.7,9–12 Inpatient physicians must determine the appropriate home medication regimen and communicate this effectively to both the family and the primary care physician. One strategy to minimize risk and improve continuity of care during this high-risk time is to recommend outpatient follow-up at the time of hospital discharge.13 Attendance at an outpatient appointment shortly after hospital discharge for asthma provides additional opportunities for ICS prescribing, reinforcement of the importance of controller medications in the long-term management of asthma, and for providers to address parental questions and concerns regarding medication management. Recent evidence suggests that improved hospital to home care transitions can increase medication adherence in the posthospitalization period.14 Unfortunately, this model of care does not appear to be achieving acceptable rates of ICS adherence among publicly insured children with asthma.7,15 In 2011 we found
KEYWORDS: asthma; controller medications; hospital followup; transitions
ACADEMIC PEDIATRICS 2017;17:212–217
Less than half of commercially insured children with an asthma hospitalization receive inhaled corticosteroids despite high rates of outpatient visits within 30 days of hospital discharge. Patients who attend outpatient visits are 40% more likely to receive these guidelinerecommended medications.
ASTHMA IS THE most common chronic condition affecting children, and the incidence of asthma-related emergency department visits and hospitalizations continues to be high.1,2 Inhaled corticosteroids (ICS), when used regularly, can reduce the symptoms of asthma, improve patient-reported quality of life, and prevent costly emergency department visits and hospitalizations.3–6 However, these evidence-based, guideline-recommended medications remain underutilized.6–8 The transition from hospital to home is a particularly vulnerable time for ACADEMIC PEDIATRICS Copyright ª 2016 by Academic Pediatric Association
212
Volume 17, Number 2 March 2017
ACADEMIC PEDIATRICS
among a statewide population of Medicaid-insured children that only 52% had a pharmacy claim for an ICS within 60 days of hospital discharge, and only 48% attended an outpatient appointment in this same time frame.7 Publicly insured children face significant barriers to both outpatient appointment attendance and medication adherence, largely due to lack of a medical home, lack of transportation, inflexible work schedules, and perceived financial barriers.16–19 Less is known about commercially insured children’s barriers and resultant utilization behavior during this high-risk transition from hospital to home after asthma exacerbation. Previous studies suggest that commercially insured families are more likely to have a medical home, are more likely to fill written prescriptions for asthma medications, and have overall higher rates of asthma controller medication adherence compared to publicly insured children.20–22 The objective of this study was to determine rates of ICS claims and outpatient follow-up appointment attendance among a national sample of commercially insured children hospitalized with acute asthma exacerbation. We also aimed to determine predictors of ICS claims that may inform future interventions to improve medication adherence in children with asthma. Consistent with the current literature, we hypothesized that rates of both ICS claims and outpatient appointment attendance in this commercially insured population would be higher than the previously published rates in publicly insured patients as a result of the presence of fewer socioeconomic barriers.
PATIENTS AND METHODS DATA We conducted a retrospective analysis using 2012–2014 Truven MarketScan claims data. These data represent over 19 million commercially insured children from the 100 largest commercial insurers in the United States. Patients aged 2 to 17 years with a hospitalization with a primary diagnosis of asthma in 2013 were identified using International Classification of Diseases, 9th Revision (ICD-9), codes 493.XX. For patients with multiple admissions for asthma in the same year, we analyzed the first admission only. Claims for ICS-containing medications were identified using NDC codes and drug names. The initial list of ICS-containing medications was reviewed by 2 authors (KS and AA) for completeness; drugs were included regardless of mode of administration (eg, metered dose inhaler, nebulizer solution). The MarketScan database includes exact dates of dispensing; therefore, we were able to determine the exact number of days between hospital discharge and any medication dispensing. Outpatient visits were identified using Current Procedural Terminology (CPT) codes 99201 to 99205 and 99211 to 99215. Visit dates were compared to hospital discharge date to identify visits that occurred within 30 days after discharge. Any outpatient visit, regardless of ICD-9 code, was included. 2012 data were used to identify prehospitalization ICS claims for those patients with hospitalizations in the first
INHALED CORTICOSTEROID CLAIMS
213
34 days of 2013, and 2014 data were used to identify outpatient appointment attendance and ICS claims within 30 days after hospitalization for those hospitalized in December 2013. VARIABLES Patient-specific variables included age in years, sex, geographic region, time of year of admission, hospital length of stay, and whether or not the patient had a claim for an ICS-containing medication within the 35 days before the hospitalization (we chose 35 days rather than 30 days here to allow a bit of flexibility in refill time and to capture more patients who had recently filled an ICS). Race/ ethnicity is not available in the MarketScan database. Consistent with previous studies, and because of differences in disease phenotype by age, age was categorized into 3 groups: 2- to 6-year-olds (preschool/kindergarten), 7- to 12-year-olds (school age), and 12- to 17-year-olds (adolescents).8,23 Geographic region, as defined in the MarketScan database, is categorized into 5 groups: Northeast, North Central, South, West, and unknown. On the basis of our calculated mean and median length of stay, for the purpose of this analysis, length of stay was dichotomized into #2 days and >2 days. Additionally, any patient with a length of stay of >7 days was excluded from our analysis, as they were thought not to represent the typical asthma hospitalization. These patients made up less than 0.6% of our total population. In order to determine if time of year of admission (which may correspond to whether or not families have met their deductible) is associated with ICS claims or outpatient visit rates, we included the quarter of admission in our analysis. This variable may also help elucidate any link between peak flu/respiratory virus season or seasonal allergies and ICS claims. Our primary outcome variables were the presence of a pharmacy claim for any ICS-containing medication and the presence of a claim for an outpatient visit within 30 days of hospital discharge. We also analyzed a combined outcome of having both an ICS claim and an outpatient visit within 30 days after hospital discharge. Secondary analysis extended the outcome time period to 60 days after hospital discharge. This strategy helped account for the fact that a patient might receive an ICS-containing medication during the hospitalization and that this would not show up as an individual pharmacy claim; rather, it might be bundled with the hospitalization claim. Using a 60-day outcome window allows patients in this situation additional time to fill their first outpatient prescription as well as additional time to attend an outpatient follow-up appointment. ANALYSIS All statistical analyses were performed by SAS 9.4 (SAS Institute, Cary, NC). We first determined overall rates of ICS claims and outpatient appointment attendance in both the 30-day outcome window and the 60-day outcome window. Because the vast majority of
214
ANDREWS ET AL
ACADEMIC PEDIATRICS
both ICS claims and outpatient visits occurred within 30 days, further analyses focused on the 30-day outcomes. We used chi-square tests to compare characteristics of patients who did and did not have a pharmacy claim for any ICS-containing medication and who did and did not have a claim for an outpatient visit within 30 days of hospital discharge. Multiple logistic regression models were built for our 2 outcome variables in addition to a combined outcome variable that represents success with both outcomes (claim for an ICS and attendance at an outpatient appointment). Covariates were selected for inclusion and examination for confounding on the basis of clinical relevance. Traditional model-fitting procedures (log-rank tests and Akaike information criterion) were then used to assess the value of each independent variable in the model.24 The final parsimonious models included independent variables (Table 3). All multiple logistic regression models were then re-run using a modified Poisson regression model to provide more robust adjusted relative risk (RR) and corresponding 95% confidence interval estimates. This is especially important in this study as these are cohort data where the event outcomes are not rare, thus reporting odds ratios could exaggerate results.25 The institutional review board of the Medical University of South Carolina reviewed this study protocol and approved it as nonhuman subjects research, given the deidentified nature of the source data. Table 1. Demographics and Inhaled Corticosteroid (ICS) Prescription Claims for Children in the Truven MarketScan Commercial Database Admitted to Hospital for Acute Asthma Exacerbation in 2013 (N ¼ 5471) Characteristic
Value
Age, y, mean/median (range) Age category 2–6 y 7–12 y 13–17 y Length of stay, mean/median (range) 1–2 d >2d Gender Male Female Geographic region Northeast North Central South West Unknown Quarter of admission Q1 (Jan–Mar) Q2 (Apr–Jun) Q3 (Jul–Sept) Q4 (Oct–Dec) ICS-containing medication claim within 35 d before admission ICS-containing medication claim by 30 d ICS-containing medication claim by 60 d Outpatient visit by 30 d Outpatient visit by 60 d ICS claim and outpatient visit by 30 d Outpatient visit by 30 d and no ICS Claim
6.7/6 (2–17) 3105 (57%) 1750 (32%) 616 (11%) 1.9/2 (1–7) 4202 (77%) 1269 (23%) 61% 39% 1399 (26%) 1029 (19%) 1842 (34%) 1041 (19%) 160 (3%) 1407 (26%) 1387 (25%) 1016 (19%) 1661 (30%) 544 (10%) 2221 (41%) 2508 (46%) 4149 (76%) 4439 (81%) 1823 (33%) 2326 (43%)
RESULTS A total of 5471 children aged 2 to 17 years had a hospitalization with a primary diagnosis code of asthma and a length of stay of #7 days in 2013 (Table 1). Forty-one percent of patients had a pharmacy claim for any ICS-containing medication within 30 days after hospital discharge. By 60 days, a total of 46% of patients had a pharmacy claim for any ICS-containing medication. A total of 76% of patients had a claim for an outpatient visit within 30 days after hospital discharge. By 60 days, 81% of patients had a claim for an outpatient visit. Only 33% of patients had both an ICS claim and an outpatient visit by 30 days. Additionally, 2326 patients or 43% of all patients attended an outpatient visit but did not have a claim for an ICS within 30 days (Table 1). As previously mentioned, because most ICS claims and outpatient visits occurred within 30 days of hospital discharge, we focused the bivariate and multivariable analysis on the 30-day outcomes. Chi-square testing and regression models were also run on the 60-day outcomes; however, there were no additional significant findings and effect sizes remained similar. In bivariate analysis, age was significantly associated with ICS claims, outpatient visits, and the combined outcome of ICS claim/outpatient visit. For all outcomes, children aged 2 to 6 were the most likely to have success. The presence of an ICS claim in the 35 days before hospital admission was significantly associated with ICS claim after hospitalization and the combined outcome of ICS claim/outpatient visit. Attending an outpatient appointment within 30 days of hospital discharge was associated with higher rates of ICS claims (Table 2). In logistic regression analysis, younger children (aged 2 to 6) were 10% more likely to have a claim for an ICS (RR 1.1 95% confidence interval [CI] 1.01–1.3), 10% more likely to attend an outpatient visit (RR 1.1, 95% CI 1.1–1.2), and 30% more likely to achieve success at both outcomes (RR 1.3, 95% CI 1.2–1.5) compared to children aged 13 to 17. Additionally, children with a prehospitalization claim for an ICS were 30% more likely to have a posthospitalization ICS claim (RR 1.3, 95% CI 1.2–1.5), 10% more likely to attend an outpatient visit (RR 1.1, 95% CI 1.004–1.1), and 40% more likely to achieve success at both outcomes (RR 1.4, 95% CI 1.3–1.6). Children who attended an outpatient appointment within 30 days after hospital discharge were 40% more likely to have a claim for an ICS (RR 1.4, 95% CI 1.3–1.6) (Table 3). DISCUSSION Among a large national sample of commercially insured children admitted to the hospital for acute asthma exacerbation, we identified low rates of ICS claims in the 2 months immediately after hospital discharge. The rates identified in this population, 41% within 30 days and 46% within 60 days, are even lower than previously reported rates among a statewide Medicaid population, where we found 52% had a controller claim within 2 months of discharge.7 Multivariable analysis revealed
ACADEMIC PEDIATRICS
INHALED CORTICOSTEROID CLAIMS
215
Table 2. Bivariate Analysis Assessing for Factors Associated With Inhaled Corticosteroid (ICS) Prescription Claims and Outpatient Visits After Hospitalization for Acute Asthma Exacerbation (N ¼ 5471)* Characteristic Age category 2–6 y 7–12 y 13–17 y Gender Male Female Region Northeast North Central South West Unknown Time of year admitted Q1 Q2 Q3 Q4 Length of stay #2 d >2 d ICS claim in 35 d before admission Yes No Outpatient visit by 30 d Yes No
Pharmacy Claim for ICS by 30 D
P
Outpatient Visit by 30 D 79% 72% 70%
.31 41% 40%
<.01
.09
.07
.11
.28
<.0001
.83 33% 34% <.0001
.07 79% 75%
<.0001 44% 30%
.52 34% 32% 33% 34%
76% 75%
53% 39%
.33 31% 35% 33% 35% 34%
78% 75% 77% 74%
40% 43%
.49 34% 33%
78% 77% 74% 74% 84%
39% 39% 40% 43%
<.0001
.77
.01
.
P
36% 31% 27%
76% 76%
37% 44% 41% 42% 38%
Both Outcomes by 30 D
<.0001
.002 43% 38% 37%
P
45% 32% .
.
.
*Chi-square tests with significance level of P < .05.
that younger age, the presence of an ICS claim in the month leading up to the hospitalization, and attendance at an outpatient appointment were all predictors of having an ICS claim after the hospitalization. Consistent with existing literature, we hypothesized the rate of ICS claims after hospitalization for acute asthma exacerbation would be higher in this commercially insured population compared to our previous findings in a statewide Medicaid population.7,20,21,26,27 Previous studies supporting this hypothesis describe the discrepancy in adherence to asthma controller medications by insurance status as being multifactorial in etiology but strongly related to socioeconomic barriers faced by publicly insured families including inadequate health care access (difficulty obtaining and/or attending outpatient appointments), lack of transportation, and single parenting with the parent working outside the home leading to competing priorities.19,26,28 We were surprised to find a lower rate of ICS claims in this commercially insured population of children compared to Medicaidinsured children. One reason that ICS claim rates might be lower in this commercially insured population is the high out-of-pocket cost for ICS-containing medications commercially insured families often face at the pharmacy.29 Unlike publicly insured families who have no copay, deductible, or coinsurance responsibilities, families with commercial insurance often incur high copays and deductibles when trying to fill these prescriptions.30 Other possible explanations for the low rate of ICS claims after hospitalization include low rates of physician prescribing
(both inpatient and outpatient), inadequate asthma education during the hospitalization (leading to low prescription fill rates), inappropriate provider messaging regarding the chronicity of the disease leading to a belief that asthma is an acute and/or episodic condition, and concerns regarding the safety of the medications related to parental experience with and/or knowledge of systemic steroids. In this study, the strongest predictor of having an ICS claim after hospitalization was attending an outpatient appointment. Children who attended outpatient appointments were 40% more likely to have a claim for an ICS. This highlights a previously unreported tangible benefit to outpatient follow-up after hospitalization for asthma and is consistent with a recent adult study of over 20,000 Medicare-insured patients that found attending an outpatient appointment within 6 weeks of hospital discharge was associated with higher rates of medication adherence.14 Despite almost universal recommendations for timely outpatient follow-up after hospitalization, there is a lack of evidence supporting direct benefits of this practice. Additionally, a recent qualitative study of parents and providers of hospitalized children revealed that parents might fail to see the benefit in outpatient hospital follow-up appointments, as noted in comments like, “Her doctor did not know what was going on” and “I didn’t think it was necessary to keep that appointment.”18 In 2015 Solan et al9 noted a similar sentiment among parents of hospitalized children where parents identified the problem of poor communication between the inpatient team and the primary care provider. In a subsequent study, Solan et al10
216
ANDREWS ET AL
ACADEMIC PEDIATRICS
Table 3. Logistic Regression Models* Predicting Presence of Claim for Any Inhaled Corticosteroid (ICS) Containing Medication and Attendance at Outpatient Appointment Within 30 D After Hospital Discharge Independent Variable
Claim for ICS by 30 D, RR (95% CI)
Age category 2–6 y 7–12 y 13–17 y Geographic region Northeast North Central South West Unknown Time of year admitted Q1 Q2 Q3 Q4 Length of stay #2 d >2 d ICS claim within 35 d before admission Yes No Outpatient visit by 30 d Yes No
Outpatient Visit by 30 D, RR (95% CI)
Both Outcomes by 30 D, RR (95% CI)
1.1 (1.01–1.3)† 1.0 (0.9–1.2) Ref
1.1 (1.1–1.2)† 1.0 (0.98–1.1) Ref
1.3 (1.2–1.5)† 1.1 (0.97–1.3) Ref
Ref 1.1 (0.96–1.1) 1.0 (0.92–1.1) 1.2 (0.96–1.5) 1.2 (1.1–1.3)
Ref 1.0 (0.99–1.1) 1.0 (0.99–1.1) 0.9 (0.9–0.99)† 1.0 (0.95–1.01)
. . . . .
Ref 1.0 (0.9–1.1) 0.9 (0.8–0.98)† 1.0 (0.9–1.1)
Ref 1.0 (0.9–1.02) 1.0 (0.97–1.1) 1.0 (0.9–1.01)
. . . .
Ref 1.1 (0.99–1.2)
. .
. .
1.3 (1.2–1.5)† Ref
1.1 (1.004–1.1)† Ref
1.4 (1.3–1.6)† Ref
1.4 (1.3–1.6)† Ref
. .
. .
RR indicates relative risk; and CI, confidence interval. *Covariates were selected for inclusion and examination for confounding on the basis of clinical relevance. Traditional model-fitting procedures (log-rank tests and Akaike information criterion) were then used to assess the value of each independent variable in the model. ICS claim model included age, region, length of stay, prehospitalization claim, and outpatient visit by 30 d. Outpatient appointment attendance model included age, region, quarter, and prehospitalization claim. Combined outcome model included age and prehospitalization claim. †Statistically significant.
noted a lack of consensus among primary care providers and hospitalists regarding the utility of a follow-up visit for all hospitalized children. Despite this, our current study found that 76% of patients attended an outpatient appointment after hospital discharge. This suggests, at least for some populations, that the potential area for improvement is not increasing rates of appointment attendance but better defining the purpose of the follow-up appointment (ie, for the outpatient physician to ensure the patient is on the appropriate maintenance medications). Communication between the discharging physician and the outpatient physician is a crucial step in ensuring the relevance and effectiveness of the follow-up appointment. Future research efforts centering around technological solutions to these communication barriers may be beneficial. Another strong predictor of having an ICS claim after hospitalization was having an ICS claim before the hospitalization. Possible explanations for this could be that the children with prehospitalization claims are simply sicker than those without and therefore are more likely to remain adherent after the hospitalization, that these children have primary care physicians who prescribe ICS more frequently, or that these children have better access to care. Conversely, the patients who do not have a claim either before or after hospitalization could have a persistent problem with lack of provider recognition of the need for a controller medication or with family understanding and
buy-in regarding the diagnosis of persistent asthma and the importance of controller medication use. There are several limitations to this study. First, we used administrative claims data for this analysis. Claims data lack the clinical detail and granularity that might be found in other data sources taken directly from electronic health records. However, claims data are the most accurate source for determining, within large populations, what medication prescriptions patients are actually filling. We are unable to determine prescription writing from these data; therefore, we do not know how much of the problem of ICS nonadherence after hospitalization is actually due to physicians failing to prescribe the guideline-recommended therapy or if the problem is due to a high rate of unfilled prescriptions. Similarly, we do not know if the prescriptions that were filled originated in the inpatient setting or at the outpatient followup appointment. Prescriptions that are paid directly by parents and thus not submitted for insurance would not be included in these data. It is also possible that some patients have ICS at home already that they can use in the immediate posthospitalization period. Additionally, our comparison of ICS claim rates after hospitalization in this national sample of commercially insured children from 2013 to a statewide sample of publicly insured children from 2007 to 2009 is not a direct comparison. However, the statewide Medicaid data are the closest
ACADEMIC PEDIATRICS
comparator group we could find that looks specifically at pharmacy claims after hospitalization for acute asthma exacerbation. Last, we were unable to determine if patients were leaving the hospital with an ICS in hand that was bundled within the hospitalization claim and therefore not showing up as a separate pharmacy claim. It is possible that a higher proportion of patients are receiving ICS within the first 30 days after hospitalization because of this factor. Additional specific limitations of the MarketScan data are the lack of a rural/urban indicator, race/ethnicity designation, and consistent presence of hospital identification.
CONCLUSIONS Less than half of commercially insured children with an asthma hospitalization receive inhaled corticosteroids after hospitalization despite high rates of outpatient appointment attendance. Although 43% of study patients attended an outpatient appointment and did not have an ICS claim, attendance at an outpatient appointment was the strongest predictor of an ICS claim. Inpatient and outpatient physicians must work in concert to increase prescribing rates and improve medication education, messaging, and communication around the time of hospital discharge. Additionally, further in-depth analysis of the out-ofpocket expenses commercially insured patients face when filling these medications will be important. ACKNOWLEDGMENTS Supported in part by South Carolina Clinical and Translational Research (SCTR) Institute, with an academic home at the Medical University of South Carolina, National Institutes of Health/National Center for Advancing Translational Sciences (grants KL2 TR001452 and UL1 TR001450), and the Doris Duke Charitable Foundation (grant 2015209).
REFERENCES 1. Akinbami LJ, Moorman JE, Bailey C, et al. Trends in asthma prevalence, health care use, and mortality in the United States, 2001–2010. NCHS Data Brief. 2012;94:1–8. 2. Nath JB, Hsia RY. Children’s emergency department use for asthma, 2001–2010. Acad Pediatr. 2015;15:225–230. 3. Adams RJ, Fuhlbrigge A, Finkelstein JA, et al. Impact of inhaled antiinflammatory therapy on hospitalization and emergency department visits for children with asthma. Pediatrics. 2001;107:706–711. 4. Cloutier MM, Hall CB, Wakefield DB, et al. Use of asthma guidelines by primary care providers to reduce hospitalizations and emergency department visits in poor, minority, urban children. J Pediatr. 2005; 146:591–597. 5. Rachelefsky G. Inhaled corticosteroids and asthma control in children: assessing impairment and risk. Pediatrics. 2009;123:353–366. 6. Expert Panel Report 3 (EPR-3). Guidelines for the diagnosis and management of asthma—summary report 2007. J Allergy Clin Immunol. 2007;120:S94–S138. 7. Lintzenich A, Teufel RJ 2nd, Basco WT Jr. Under-utilization of controller medications and poor follow-up rates among hospitalized asthma patients. Hosp Pediatr. 2011;1:8–14. 8. Lintzenich A, Teufel RJ 2nd, Basco WT Jr. Younger asthmatics are less likely to receive inhaled corticosteroids and asthma education after admission for exacerbation. Clin Pediatr. 2010;49:1111–1116.
INHALED CORTICOSTEROID CLAIMS
217
9. Solan LG, Beck AF, Brunswick SA, et al. The family perspective on hospital to home transitions: a qualitative study. Pediatrics. 2015;136: e1539–e1549. 10. Solan LG, Sherman SN, DeBlasio D, et al. Communication challenges: a qualitative look at the relationship between pediatric hospitalists and primary care providers. Acad Pediatr. 2016;16:453–459. 11. US Institute of Medicine; Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academies Press; 2001. 12. Davis MM, Devoe M, Kansagara D, et al. “Did I do as best as the system would let me?” Healthcare professional views on hospital to home care transitions. J Gen Intern Med. 2012;27:1649–1656. 13. Agency for Healthcare Research and Quality Research. Strategy 4: Care Transitions from Hospital to Home: IDEAL Discharge Planning. Available at: http://www.ahrq.gov/professionals/systems/hospital/ engagingfamilies/strategy4/index.html; 2013. Accessed August 22, 2016. 14. Faridi KF, Peterson ED, McCoy LA, et al. Timing of first postdischarge follow-up and medication adherence after acute myocardial infarction. JAMA Cardiol. 2016;1:147–155. 15. Kenyon CC, Rubin DM, Zorc JJ, et al. Childhood asthma hospital discharge medication fills and risk of subsequent readmission. J Pediatr. 2015;166:1121–1127. 16. Auger KA, Kahn RS, Davis MM, et al. Medical home quality and readmission risk for children hospitalized with asthma exacerbations. Pediatrics. 2013;131:64–70. 17. Hefner JL, Wexler R, McAlearney AS. Primary care access barriers as reported by nonurgent emergency department users: implications for the US primary care infrastructure. Am J Med Qual. 2015;30: 135–140. 18. Amin D, Ford R, Ghazarian SR, et al. Parent and physician perceptions regarding preventability of pediatric readmissions. Hosp Pediatr. 2016;6:80–87. 19. Wang NE, Gisondi MA, Golzari M, et al. Socioeconomic disparities are negatively associated with pediatric emergency department aftercare compliance. Acad Emerg Med. 2003;10:1278–1284. 20. Stanford RH, Shah MB, D’Souza AO, et al. Predicting asthma outcomes in commercially insured and Medicaid populations. Am J Manag Care. 2013;19:60–67. 21. Finkelstein JA, Barton MB, Donahue JG, et al. Comparing asthma care for Medicaid and non-Medicaid children in a health maintenance organization. Arch Pediatr Adolesc Med. 2000;154:563–568. 22. Zickafoose JS, Gebremariam A, Clark SJ, et al. Medical home disparities between children with public and private insurance. Acad Pediatr. 2011;11:305–310. 23. Andrews AL, Teufel RJ 2nd, Basco WT Jr. Low rates of controller medication initiation and outpatient follow-up after emergency department visits for asthma. J Pediatr. 2012;160:325–330. 24. Hosmer DW Jr, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: Wiley; 2000. 25. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–706. 26. Bloomberg GR, Banister C, Sterkel R, et al. Socioeconomic, family, and pediatric practice factors that affect level of asthma control. Pediatrics. 2009;123:829–835. 27. Ortega AN, Belanger KD, Paltiel AD, et al. Use of health services by insurance status among children with asthma. Med Care. 2001;39: 1065–1074. 28. Halterman JS, Aligne CA, Auinger P, et al. Inadequate therapy for asthma among children in the United States. Pediatrics. 2000;105: 272–276. 29. Karaca-Mandic P, Jena AB, Joyce GF, et al. Out-of-pocket medication costs and use of medications and health care services among children with asthma. JAMA. 2012;307:1284–1291. 30. Novack GD. What determines how much your patient pays for their medication in the United States? Am J Ophthalmol. 2016;167:48–51.
ABSTRACT OBJECTIVE: To determine rates of inhaled corticosteroid (ICS) claims and outpatient follow-up after asthma hospitalization among commercially insured children. METHODS: We conducted a retrospective cohort analysis of children hospitalized for asthma using 2013 national Truven MarketScan data. Covariates included age, sex, region, length of stay, and having an ICS claim within 35 days before hospitalization. Outcome variables were a claim for any ICS-containing medication and outpatient visit within 30 days after discharge. Multivariable analysis used logistic regression. RESULTS: A total of 5471 children aged 2 to 17 were included; 61% were boys, and mean age was 6.8 years. Forty-one percent had a claim for an ICS and 76% had an outpatient visit within 30 days after hospital discharge. In multivariable analysis, children who had an ICS claim within 35 days before the hospitalization were more likely to have an ICS claim within 30 days after discharge (relative risk [RR] 1.3, 95% confidence interval
[CI] 1.2–1.5). The strongest predictor of an ICS claim within 30 days after discharge was attendance at an outpatient appointment (RR 1.4, 95% CI 1.3–1.6). Children aged 2 to 6 were more likely to attend an outpatient appointment (RR 1.1, 95% CI 1.1–1.2). Children with an ICS claim before admission were also more likely to attend an outpatient appointment (RR 1.1, 95% CI 1.004–1.1). CONCLUSIONS: In this national sample of commercially insured children with asthma, rates of ICS claims after hospitalization are low despite high rates of outpatient visits. Both inpatient and outpatient physicians must play a role in increasing ICS adherence in this high-risk population of children with asthma.
WHAT’S NEW
children, including those with asthma.7,9–12 Inpatient physicians must determine the appropriate home medication regimen and communicate this effectively to both the family and the primary care physician. One strategy to minimize risk and improve continuity of care during this high-risk time is to recommend outpatient follow-up at the time of hospital discharge.13 Attendance at an outpatient appointment shortly after hospital discharge for asthma provides additional opportunities for ICS prescribing, reinforcement of the importance of controller medications in the long-term management of asthma, and for providers to address parental questions and concerns regarding medication management. Recent evidence suggests that improved hospital to home care transitions can increase medication adherence in the posthospitalization period.14 Unfortunately, this model of care does not appear to be achieving acceptable rates of ICS adherence among publicly insured children with asthma.7,15 In 2011 we found
KEYWORDS: asthma; controller medications; hospital followup; transitions
ACADEMIC PEDIATRICS 2017;17:212–217
Less than half of commercially insured children with an asthma hospitalization receive inhaled corticosteroids despite high rates of outpatient visits within 30 days of hospital discharge. Patients who attend outpatient visits are 40% more likely to receive these guidelinerecommended medications.
ASTHMA IS THE most common chronic condition affecting children, and the incidence of asthma-related emergency department visits and hospitalizations continues to be high.1,2 Inhaled corticosteroids (ICS), when used regularly, can reduce the symptoms of asthma, improve patient-reported quality of life, and prevent costly emergency department visits and hospitalizations.3–6 However, these evidence-based, guideline-recommended medications remain underutilized.6–8 The transition from hospital to home is a particularly vulnerable time for ACADEMIC PEDIATRICS Copyright ª 2016 by Academic Pediatric Association
212
Volume 17, Number 2 March 2017
ACADEMIC PEDIATRICS
among a statewide population of Medicaid-insured children that only 52% had a pharmacy claim for an ICS within 60 days of hospital discharge, and only 48% attended an outpatient appointment in this same time frame.7 Publicly insured children face significant barriers to both outpatient appointment attendance and medication adherence, largely due to lack of a medical home, lack of transportation, inflexible work schedules, and perceived financial barriers.16–19 Less is known about commercially insured children’s barriers and resultant utilization behavior during this high-risk transition from hospital to home after asthma exacerbation. Previous studies suggest that commercially insured families are more likely to have a medical home, are more likely to fill written prescriptions for asthma medications, and have overall higher rates of asthma controller medication adherence compared to publicly insured children.20–22 The objective of this study was to determine rates of ICS claims and outpatient follow-up appointment attendance among a national sample of commercially insured children hospitalized with acute asthma exacerbation. We also aimed to determine predictors of ICS claims that may inform future interventions to improve medication adherence in children with asthma. Consistent with the current literature, we hypothesized that rates of both ICS claims and outpatient appointment attendance in this commercially insured population would be higher than the previously published rates in publicly insured patients as a result of the presence of fewer socioeconomic barriers.
PATIENTS AND METHODS DATA We conducted a retrospective analysis using 2012–2014 Truven MarketScan claims data. These data represent over 19 million commercially insured children from the 100 largest commercial insurers in the United States. Patients aged 2 to 17 years with a hospitalization with a primary diagnosis of asthma in 2013 were identified using International Classification of Diseases, 9th Revision (ICD-9), codes 493.XX. For patients with multiple admissions for asthma in the same year, we analyzed the first admission only. Claims for ICS-containing medications were identified using NDC codes and drug names. The initial list of ICS-containing medications was reviewed by 2 authors (KS and AA) for completeness; drugs were included regardless of mode of administration (eg, metered dose inhaler, nebulizer solution). The MarketScan database includes exact dates of dispensing; therefore, we were able to determine the exact number of days between hospital discharge and any medication dispensing. Outpatient visits were identified using Current Procedural Terminology (CPT) codes 99201 to 99205 and 99211 to 99215. Visit dates were compared to hospital discharge date to identify visits that occurred within 30 days after discharge. Any outpatient visit, regardless of ICD-9 code, was included. 2012 data were used to identify prehospitalization ICS claims for those patients with hospitalizations in the first
INHALED CORTICOSTEROID CLAIMS
213
34 days of 2013, and 2014 data were used to identify outpatient appointment attendance and ICS claims within 30 days after hospitalization for those hospitalized in December 2013. VARIABLES Patient-specific variables included age in years, sex, geographic region, time of year of admission, hospital length of stay, and whether or not the patient had a claim for an ICS-containing medication within the 35 days before the hospitalization (we chose 35 days rather than 30 days here to allow a bit of flexibility in refill time and to capture more patients who had recently filled an ICS). Race/ ethnicity is not available in the MarketScan database. Consistent with previous studies, and because of differences in disease phenotype by age, age was categorized into 3 groups: 2- to 6-year-olds (preschool/kindergarten), 7- to 12-year-olds (school age), and 12- to 17-year-olds (adolescents).8,23 Geographic region, as defined in the MarketScan database, is categorized into 5 groups: Northeast, North Central, South, West, and unknown. On the basis of our calculated mean and median length of stay, for the purpose of this analysis, length of stay was dichotomized into #2 days and >2 days. Additionally, any patient with a length of stay of >7 days was excluded from our analysis, as they were thought not to represent the typical asthma hospitalization. These patients made up less than 0.6% of our total population. In order to determine if time of year of admission (which may correspond to whether or not families have met their deductible) is associated with ICS claims or outpatient visit rates, we included the quarter of admission in our analysis. This variable may also help elucidate any link between peak flu/respiratory virus season or seasonal allergies and ICS claims. Our primary outcome variables were the presence of a pharmacy claim for any ICS-containing medication and the presence of a claim for an outpatient visit within 30 days of hospital discharge. We also analyzed a combined outcome of having both an ICS claim and an outpatient visit within 30 days after hospital discharge. Secondary analysis extended the outcome time period to 60 days after hospital discharge. This strategy helped account for the fact that a patient might receive an ICS-containing medication during the hospitalization and that this would not show up as an individual pharmacy claim; rather, it might be bundled with the hospitalization claim. Using a 60-day outcome window allows patients in this situation additional time to fill their first outpatient prescription as well as additional time to attend an outpatient follow-up appointment. ANALYSIS All statistical analyses were performed by SAS 9.4 (SAS Institute, Cary, NC). We first determined overall rates of ICS claims and outpatient appointment attendance in both the 30-day outcome window and the 60-day outcome window. Because the vast majority of
214
ANDREWS ET AL
ACADEMIC PEDIATRICS
both ICS claims and outpatient visits occurred within 30 days, further analyses focused on the 30-day outcomes. We used chi-square tests to compare characteristics of patients who did and did not have a pharmacy claim for any ICS-containing medication and who did and did not have a claim for an outpatient visit within 30 days of hospital discharge. Multiple logistic regression models were built for our 2 outcome variables in addition to a combined outcome variable that represents success with both outcomes (claim for an ICS and attendance at an outpatient appointment). Covariates were selected for inclusion and examination for confounding on the basis of clinical relevance. Traditional model-fitting procedures (log-rank tests and Akaike information criterion) were then used to assess the value of each independent variable in the model.24 The final parsimonious models included independent variables (Table 3). All multiple logistic regression models were then re-run using a modified Poisson regression model to provide more robust adjusted relative risk (RR) and corresponding 95% confidence interval estimates. This is especially important in this study as these are cohort data where the event outcomes are not rare, thus reporting odds ratios could exaggerate results.25 The institutional review board of the Medical University of South Carolina reviewed this study protocol and approved it as nonhuman subjects research, given the deidentified nature of the source data. Table 1. Demographics and Inhaled Corticosteroid (ICS) Prescription Claims for Children in the Truven MarketScan Commercial Database Admitted to Hospital for Acute Asthma Exacerbation in 2013 (N ¼ 5471) Characteristic
Value
Age, y, mean/median (range) Age category 2–6 y 7–12 y 13–17 y Length of stay, mean/median (range) 1–2 d >2d Gender Male Female Geographic region Northeast North Central South West Unknown Quarter of admission Q1 (Jan–Mar) Q2 (Apr–Jun) Q3 (Jul–Sept) Q4 (Oct–Dec) ICS-containing medication claim within 35 d before admission ICS-containing medication claim by 30 d ICS-containing medication claim by 60 d Outpatient visit by 30 d Outpatient visit by 60 d ICS claim and outpatient visit by 30 d Outpatient visit by 30 d and no ICS Claim
6.7/6 (2–17) 3105 (57%) 1750 (32%) 616 (11%) 1.9/2 (1–7) 4202 (77%) 1269 (23%) 61% 39% 1399 (26%) 1029 (19%) 1842 (34%) 1041 (19%) 160 (3%) 1407 (26%) 1387 (25%) 1016 (19%) 1661 (30%) 544 (10%) 2221 (41%) 2508 (46%) 4149 (76%) 4439 (81%) 1823 (33%) 2326 (43%)
RESULTS A total of 5471 children aged 2 to 17 years had a hospitalization with a primary diagnosis code of asthma and a length of stay of #7 days in 2013 (Table 1). Forty-one percent of patients had a pharmacy claim for any ICS-containing medication within 30 days after hospital discharge. By 60 days, a total of 46% of patients had a pharmacy claim for any ICS-containing medication. A total of 76% of patients had a claim for an outpatient visit within 30 days after hospital discharge. By 60 days, 81% of patients had a claim for an outpatient visit. Only 33% of patients had both an ICS claim and an outpatient visit by 30 days. Additionally, 2326 patients or 43% of all patients attended an outpatient visit but did not have a claim for an ICS within 30 days (Table 1). As previously mentioned, because most ICS claims and outpatient visits occurred within 30 days of hospital discharge, we focused the bivariate and multivariable analysis on the 30-day outcomes. Chi-square testing and regression models were also run on the 60-day outcomes; however, there were no additional significant findings and effect sizes remained similar. In bivariate analysis, age was significantly associated with ICS claims, outpatient visits, and the combined outcome of ICS claim/outpatient visit. For all outcomes, children aged 2 to 6 were the most likely to have success. The presence of an ICS claim in the 35 days before hospital admission was significantly associated with ICS claim after hospitalization and the combined outcome of ICS claim/outpatient visit. Attending an outpatient appointment within 30 days of hospital discharge was associated with higher rates of ICS claims (Table 2). In logistic regression analysis, younger children (aged 2 to 6) were 10% more likely to have a claim for an ICS (RR 1.1 95% confidence interval [CI] 1.01–1.3), 10% more likely to attend an outpatient visit (RR 1.1, 95% CI 1.1–1.2), and 30% more likely to achieve success at both outcomes (RR 1.3, 95% CI 1.2–1.5) compared to children aged 13 to 17. Additionally, children with a prehospitalization claim for an ICS were 30% more likely to have a posthospitalization ICS claim (RR 1.3, 95% CI 1.2–1.5), 10% more likely to attend an outpatient visit (RR 1.1, 95% CI 1.004–1.1), and 40% more likely to achieve success at both outcomes (RR 1.4, 95% CI 1.3–1.6). Children who attended an outpatient appointment within 30 days after hospital discharge were 40% more likely to have a claim for an ICS (RR 1.4, 95% CI 1.3–1.6) (Table 3). DISCUSSION Among a large national sample of commercially insured children admitted to the hospital for acute asthma exacerbation, we identified low rates of ICS claims in the 2 months immediately after hospital discharge. The rates identified in this population, 41% within 30 days and 46% within 60 days, are even lower than previously reported rates among a statewide Medicaid population, where we found 52% had a controller claim within 2 months of discharge.7 Multivariable analysis revealed
ACADEMIC PEDIATRICS
INHALED CORTICOSTEROID CLAIMS
215
Table 2. Bivariate Analysis Assessing for Factors Associated With Inhaled Corticosteroid (ICS) Prescription Claims and Outpatient Visits After Hospitalization for Acute Asthma Exacerbation (N ¼ 5471)* Characteristic Age category 2–6 y 7–12 y 13–17 y Gender Male Female Region Northeast North Central South West Unknown Time of year admitted Q1 Q2 Q3 Q4 Length of stay #2 d >2 d ICS claim in 35 d before admission Yes No Outpatient visit by 30 d Yes No
Pharmacy Claim for ICS by 30 D
P
Outpatient Visit by 30 D 79% 72% 70%
.31 41% 40%
<.01
.09
.07
.11
.28
<.0001
.83 33% 34% <.0001
.07 79% 75%
<.0001 44% 30%
.52 34% 32% 33% 34%
76% 75%
53% 39%
.33 31% 35% 33% 35% 34%
78% 75% 77% 74%
40% 43%
.49 34% 33%
78% 77% 74% 74% 84%
39% 39% 40% 43%
<.0001
.77
.01
.
P
36% 31% 27%
76% 76%
37% 44% 41% 42% 38%
Both Outcomes by 30 D
<.0001
.002 43% 38% 37%
P
45% 32% .
.
.
*Chi-square tests with significance level of P < .05.
that younger age, the presence of an ICS claim in the month leading up to the hospitalization, and attendance at an outpatient appointment were all predictors of having an ICS claim after the hospitalization. Consistent with existing literature, we hypothesized the rate of ICS claims after hospitalization for acute asthma exacerbation would be higher in this commercially insured population compared to our previous findings in a statewide Medicaid population.7,20,21,26,27 Previous studies supporting this hypothesis describe the discrepancy in adherence to asthma controller medications by insurance status as being multifactorial in etiology but strongly related to socioeconomic barriers faced by publicly insured families including inadequate health care access (difficulty obtaining and/or attending outpatient appointments), lack of transportation, and single parenting with the parent working outside the home leading to competing priorities.19,26,28 We were surprised to find a lower rate of ICS claims in this commercially insured population of children compared to Medicaidinsured children. One reason that ICS claim rates might be lower in this commercially insured population is the high out-of-pocket cost for ICS-containing medications commercially insured families often face at the pharmacy.29 Unlike publicly insured families who have no copay, deductible, or coinsurance responsibilities, families with commercial insurance often incur high copays and deductibles when trying to fill these prescriptions.30 Other possible explanations for the low rate of ICS claims after hospitalization include low rates of physician prescribing
(both inpatient and outpatient), inadequate asthma education during the hospitalization (leading to low prescription fill rates), inappropriate provider messaging regarding the chronicity of the disease leading to a belief that asthma is an acute and/or episodic condition, and concerns regarding the safety of the medications related to parental experience with and/or knowledge of systemic steroids. In this study, the strongest predictor of having an ICS claim after hospitalization was attending an outpatient appointment. Children who attended outpatient appointments were 40% more likely to have a claim for an ICS. This highlights a previously unreported tangible benefit to outpatient follow-up after hospitalization for asthma and is consistent with a recent adult study of over 20,000 Medicare-insured patients that found attending an outpatient appointment within 6 weeks of hospital discharge was associated with higher rates of medication adherence.14 Despite almost universal recommendations for timely outpatient follow-up after hospitalization, there is a lack of evidence supporting direct benefits of this practice. Additionally, a recent qualitative study of parents and providers of hospitalized children revealed that parents might fail to see the benefit in outpatient hospital follow-up appointments, as noted in comments like, “Her doctor did not know what was going on” and “I didn’t think it was necessary to keep that appointment.”18 In 2015 Solan et al9 noted a similar sentiment among parents of hospitalized children where parents identified the problem of poor communication between the inpatient team and the primary care provider. In a subsequent study, Solan et al10
216
ANDREWS ET AL
ACADEMIC PEDIATRICS
Table 3. Logistic Regression Models* Predicting Presence of Claim for Any Inhaled Corticosteroid (ICS) Containing Medication and Attendance at Outpatient Appointment Within 30 D After Hospital Discharge Independent Variable
Claim for ICS by 30 D, RR (95% CI)
Age category 2–6 y 7–12 y 13–17 y Geographic region Northeast North Central South West Unknown Time of year admitted Q1 Q2 Q3 Q4 Length of stay #2 d >2 d ICS claim within 35 d before admission Yes No Outpatient visit by 30 d Yes No
Outpatient Visit by 30 D, RR (95% CI)
Both Outcomes by 30 D, RR (95% CI)
1.1 (1.01–1.3)† 1.0 (0.9–1.2) Ref
1.1 (1.1–1.2)† 1.0 (0.98–1.1) Ref
1.3 (1.2–1.5)† 1.1 (0.97–1.3) Ref
Ref 1.1 (0.96–1.1) 1.0 (0.92–1.1) 1.2 (0.96–1.5) 1.2 (1.1–1.3)
Ref 1.0 (0.99–1.1) 1.0 (0.99–1.1) 0.9 (0.9–0.99)† 1.0 (0.95–1.01)
. . . . .
Ref 1.0 (0.9–1.1) 0.9 (0.8–0.98)† 1.0 (0.9–1.1)
Ref 1.0 (0.9–1.02) 1.0 (0.97–1.1) 1.0 (0.9–1.01)
. . . .
Ref 1.1 (0.99–1.2)
. .
. .
1.3 (1.2–1.5)† Ref
1.1 (1.004–1.1)† Ref
1.4 (1.3–1.6)† Ref
1.4 (1.3–1.6)† Ref
. .
. .
RR indicates relative risk; and CI, confidence interval. *Covariates were selected for inclusion and examination for confounding on the basis of clinical relevance. Traditional model-fitting procedures (log-rank tests and Akaike information criterion) were then used to assess the value of each independent variable in the model. ICS claim model included age, region, length of stay, prehospitalization claim, and outpatient visit by 30 d. Outpatient appointment attendance model included age, region, quarter, and prehospitalization claim. Combined outcome model included age and prehospitalization claim. †Statistically significant.
noted a lack of consensus among primary care providers and hospitalists regarding the utility of a follow-up visit for all hospitalized children. Despite this, our current study found that 76% of patients attended an outpatient appointment after hospital discharge. This suggests, at least for some populations, that the potential area for improvement is not increasing rates of appointment attendance but better defining the purpose of the follow-up appointment (ie, for the outpatient physician to ensure the patient is on the appropriate maintenance medications). Communication between the discharging physician and the outpatient physician is a crucial step in ensuring the relevance and effectiveness of the follow-up appointment. Future research efforts centering around technological solutions to these communication barriers may be beneficial. Another strong predictor of having an ICS claim after hospitalization was having an ICS claim before the hospitalization. Possible explanations for this could be that the children with prehospitalization claims are simply sicker than those without and therefore are more likely to remain adherent after the hospitalization, that these children have primary care physicians who prescribe ICS more frequently, or that these children have better access to care. Conversely, the patients who do not have a claim either before or after hospitalization could have a persistent problem with lack of provider recognition of the need for a controller medication or with family understanding and
buy-in regarding the diagnosis of persistent asthma and the importance of controller medication use. There are several limitations to this study. First, we used administrative claims data for this analysis. Claims data lack the clinical detail and granularity that might be found in other data sources taken directly from electronic health records. However, claims data are the most accurate source for determining, within large populations, what medication prescriptions patients are actually filling. We are unable to determine prescription writing from these data; therefore, we do not know how much of the problem of ICS nonadherence after hospitalization is actually due to physicians failing to prescribe the guideline-recommended therapy or if the problem is due to a high rate of unfilled prescriptions. Similarly, we do not know if the prescriptions that were filled originated in the inpatient setting or at the outpatient followup appointment. Prescriptions that are paid directly by parents and thus not submitted for insurance would not be included in these data. It is also possible that some patients have ICS at home already that they can use in the immediate posthospitalization period. Additionally, our comparison of ICS claim rates after hospitalization in this national sample of commercially insured children from 2013 to a statewide sample of publicly insured children from 2007 to 2009 is not a direct comparison. However, the statewide Medicaid data are the closest
ACADEMIC PEDIATRICS
comparator group we could find that looks specifically at pharmacy claims after hospitalization for acute asthma exacerbation. Last, we were unable to determine if patients were leaving the hospital with an ICS in hand that was bundled within the hospitalization claim and therefore not showing up as a separate pharmacy claim. It is possible that a higher proportion of patients are receiving ICS within the first 30 days after hospitalization because of this factor. Additional specific limitations of the MarketScan data are the lack of a rural/urban indicator, race/ethnicity designation, and consistent presence of hospital identification.
CONCLUSIONS Less than half of commercially insured children with an asthma hospitalization receive inhaled corticosteroids after hospitalization despite high rates of outpatient appointment attendance. Although 43% of study patients attended an outpatient appointment and did not have an ICS claim, attendance at an outpatient appointment was the strongest predictor of an ICS claim. Inpatient and outpatient physicians must work in concert to increase prescribing rates and improve medication education, messaging, and communication around the time of hospital discharge. Additionally, further in-depth analysis of the out-ofpocket expenses commercially insured patients face when filling these medications will be important. ACKNOWLEDGMENTS Supported in part by South Carolina Clinical and Translational Research (SCTR) Institute, with an academic home at the Medical University of South Carolina, National Institutes of Health/National Center for Advancing Translational Sciences (grants KL2 TR001452 and UL1 TR001450), and the Doris Duke Charitable Foundation (grant 2015209).
REFERENCES 1. Akinbami LJ, Moorman JE, Bailey C, et al. Trends in asthma prevalence, health care use, and mortality in the United States, 2001–2010. NCHS Data Brief. 2012;94:1–8. 2. Nath JB, Hsia RY. Children’s emergency department use for asthma, 2001–2010. Acad Pediatr. 2015;15:225–230. 3. Adams RJ, Fuhlbrigge A, Finkelstein JA, et al. Impact of inhaled antiinflammatory therapy on hospitalization and emergency department visits for children with asthma. Pediatrics. 2001;107:706–711. 4. Cloutier MM, Hall CB, Wakefield DB, et al. Use of asthma guidelines by primary care providers to reduce hospitalizations and emergency department visits in poor, minority, urban children. J Pediatr. 2005; 146:591–597. 5. Rachelefsky G. Inhaled corticosteroids and asthma control in children: assessing impairment and risk. Pediatrics. 2009;123:353–366. 6. Expert Panel Report 3 (EPR-3). Guidelines for the diagnosis and management of asthma—summary report 2007. J Allergy Clin Immunol. 2007;120:S94–S138. 7. Lintzenich A, Teufel RJ 2nd, Basco WT Jr. Under-utilization of controller medications and poor follow-up rates among hospitalized asthma patients. Hosp Pediatr. 2011;1:8–14. 8. Lintzenich A, Teufel RJ 2nd, Basco WT Jr. Younger asthmatics are less likely to receive inhaled corticosteroids and asthma education after admission for exacerbation. Clin Pediatr. 2010;49:1111–1116.
INHALED CORTICOSTEROID CLAIMS
217
9. Solan LG, Beck AF, Brunswick SA, et al. The family perspective on hospital to home transitions: a qualitative study. Pediatrics. 2015;136: e1539–e1549. 10. Solan LG, Sherman SN, DeBlasio D, et al. Communication challenges: a qualitative look at the relationship between pediatric hospitalists and primary care providers. Acad Pediatr. 2016;16:453–459. 11. US Institute of Medicine; Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academies Press; 2001. 12. Davis MM, Devoe M, Kansagara D, et al. “Did I do as best as the system would let me?” Healthcare professional views on hospital to home care transitions. J Gen Intern Med. 2012;27:1649–1656. 13. Agency for Healthcare Research and Quality Research. Strategy 4: Care Transitions from Hospital to Home: IDEAL Discharge Planning. Available at: http://www.ahrq.gov/professionals/systems/hospital/ engagingfamilies/strategy4/index.html; 2013. Accessed August 22, 2016. 14. Faridi KF, Peterson ED, McCoy LA, et al. Timing of first postdischarge follow-up and medication adherence after acute myocardial infarction. JAMA Cardiol. 2016;1:147–155. 15. Kenyon CC, Rubin DM, Zorc JJ, et al. Childhood asthma hospital discharge medication fills and risk of subsequent readmission. J Pediatr. 2015;166:1121–1127. 16. Auger KA, Kahn RS, Davis MM, et al. Medical home quality and readmission risk for children hospitalized with asthma exacerbations. Pediatrics. 2013;131:64–70. 17. Hefner JL, Wexler R, McAlearney AS. Primary care access barriers as reported by nonurgent emergency department users: implications for the US primary care infrastructure. Am J Med Qual. 2015;30: 135–140. 18. Amin D, Ford R, Ghazarian SR, et al. Parent and physician perceptions regarding preventability of pediatric readmissions. Hosp Pediatr. 2016;6:80–87. 19. Wang NE, Gisondi MA, Golzari M, et al. Socioeconomic disparities are negatively associated with pediatric emergency department aftercare compliance. Acad Emerg Med. 2003;10:1278–1284. 20. Stanford RH, Shah MB, D’Souza AO, et al. Predicting asthma outcomes in commercially insured and Medicaid populations. Am J Manag Care. 2013;19:60–67. 21. Finkelstein JA, Barton MB, Donahue JG, et al. Comparing asthma care for Medicaid and non-Medicaid children in a health maintenance organization. Arch Pediatr Adolesc Med. 2000;154:563–568. 22. Zickafoose JS, Gebremariam A, Clark SJ, et al. Medical home disparities between children with public and private insurance. Acad Pediatr. 2011;11:305–310. 23. Andrews AL, Teufel RJ 2nd, Basco WT Jr. Low rates of controller medication initiation and outpatient follow-up after emergency department visits for asthma. J Pediatr. 2012;160:325–330. 24. Hosmer DW Jr, Lemeshow S. Applied Logistic Regression. 2nd ed. New York: Wiley; 2000. 25. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702–706. 26. Bloomberg GR, Banister C, Sterkel R, et al. Socioeconomic, family, and pediatric practice factors that affect level of asthma control. Pediatrics. 2009;123:829–835. 27. Ortega AN, Belanger KD, Paltiel AD, et al. Use of health services by insurance status among children with asthma. Med Care. 2001;39: 1065–1074. 28. Halterman JS, Aligne CA, Auinger P, et al. Inadequate therapy for asthma among children in the United States. Pediatrics. 2000;105: 272–276. 29. Karaca-Mandic P, Jena AB, Joyce GF, et al. Out-of-pocket medication costs and use of medications and health care services among children with asthma. JAMA. 2012;307:1284–1291. 30. Novack GD. What determines how much your patient pays for their medication in the United States? Am J Ophthalmol. 2016;167:48–51.