Valvular and Congenital Heart Disease

Inpatient admissions and costs of congenital heart disease from adolescence to young adulthood Yang Lu, PhD, a Garima Agrawal, MPH, b Chia-Wei Lin, MS, b and Roberta G. Williams, MD c Torrance, and Los Angeles, CA

Background Health care utilization patterns and associated costs of congenital heart disease (CHD) in young adulthood are not well understood. Methods

California State Inpatient Databases 2005 to 2009 were used to conduct a retrospective study on inpatient admissions of CHD patients 10 to 29 years old (n = 7,419) and of all patients of the same age (n = 1,195,233).

Results

Compared with all patients, annual admission costs of CHD patients monotonically decreased, from $13.9 million at age 10 to 14 years to $7.2 million at age 25 to 29 years, which were due to lower costs per admission and somewhat fewer annual hospitalizations. Admissions from the emergency department in CHD patients increased with age regardless of insurance status: at age 25 to 29 years, 62% of admissions with public insurance, 43% with private insurance, and 78% with no insurance were admitted from the emergency department. Cardiac surgery, catheterization, and electrophysiologic study admissions decreased with age in CHD patients, whereas admissions due to arrhythmia and congestive heart failures became more prevalent. Results from multivariate linear regression in CHD patients showed that cardiac surgery was the most costly factor, associated with $29,543 more in costs per admission, followed by the use of a children's hospital, at $10,533. Factors predicting lower costs included female gender, low-complexity CHD, and shorter stay, all P b .001.

Conclusions

Compared with adolescents, young adults with CHD use fewer resources because the natural history of CHD results in fewer surgical admissions and more frequent but less expensive medical admissions. (Am Heart J 2014;168:948-55.)

Because of improved resources dedicated to treatment and diagnosis, congenital heart disease (CHD) has been transformed from a fatal disease to a chronic condition, enabling patients with moderate and complex CHD to survive into adulthood. 1 Hospitalizations of adult patients with CHD (ACHD) subsequently increased, accounting for 28.9% of all CHD hospitalizations in 1998 to 2004 and 36.5% in 2004 to 2010. 2 The total charges of ACHD hospitalizations rose from $691 million in 1998 to $3.16 billion in 2005 nationally. 3 However, little is known about the resource utilization patterns and health care costs of CHD in young adulthood, a transitional phase when most pediatric patients with CHD age out of specialty pediatric care. Many experience difficulty finding physicians with expertise in ACHD care in addition to losing public or private insurance coverage and therefore experiencing discontinuous medical care. 4-9 For example, Gurvitz et al 6 showed that ACHD patients usually From the aHarbor-UCLA Medical Center & Los Angeles Biomedical Research Institute, Torrance, California, bUniversity of Southern California, Los Angeles, CA, and cUniversity of Southern California, Los Angeles, CA & Children's Hospital Los Angeles, Los Angeles, CA. Submitted March 4, 2014; accepted August 2, 2014. Reprint requests: Roberta G. Williams, MD, Department of Pediatrics, Children's Hospital Los Angeles, 4650 W Sunset Boulevard, MS no. 34, Los Angeles, CA 90027. E-mail: [email protected] 0002-8703 © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ahj.2014.08.006

encounter their first gap in medical care at 19 years old and that 42% had a gap in medical care of ≥3 years before they first presented to an ACHD clinic. In addition, more than onethird of ACHD admissions originated in the emergency department (ED), nearly doubling that of adolescent CHD patients. 8 It has become apparent that this emerging population of ACHD patients has unique health care needs that are not being met. 4,5 As a result, 60% of ACHD patients lost to follow-up developed adverse outcomes or required urgent cardiac intervention, 6,10,11 potentially contributing to higher medical costs later in life. 12,13 Therefore, it is imperative to understand health care needs of emerging adults with CHD so that CHD-related adverse outcomes and complications can be prevented or delayed. Our study fills the knowledge gap in the literature by examining how hospitalization and costs change among CHD patients, as they transition from adolescence (10-19 years) to young adulthood (20-29 years). Understanding these changes in resource utilization and costs is the first step toward planning a health care system that will deliver quality care at sustainable costs for young adults with CHD.

Methods Data sources We used three databases: the California State Inpatient Databases, the Healthcare Cost and Utilization Project

American Heart Journal Volume 168, Number 6

(HCUP) cost-to-charge ratio (CCR) files, and the California Hospital Annual Utilization Data 2005 to 2009. By merging the first 2 databases, we obtained data for approximately 97% of all hospitalization records in California with imputed costs and administrative information including principal and secondary diagnoses, principal and secondary procedures, admission and discharge status, source of admission, basic patient demographics, total charges, and length of stay (LOS). By incorporating the third database, we obtained hospital characteristics for each admission, such as children's hospital status, teaching hospital status, bed size, and ownership models (public; private, not for profit; and private, for profit).

Study sample Pregnancy-related admissions and admissions with missing hospital charges were excluded, resulting in 1,195,233 all-patient (AP) admissions and 7,419 CHD admissions in the 10 to 29 age range (Figure 1). Pregnancy-related admissions were defined as admissions of obstetrical procedures, pregnancy complications, and postpartum stays. 14 Congenital heart disease admissions were identified as those with ≥1 principal or secondary International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes included in the Clinical Classifications Software Code 213 (for cardiac and circulatory congenital anomalies) with the exception of 747.5, 747.6, and 747.8. This definition has been used as the standard method for identifying CHD cases from claims data in existing literature. 3 Definition of costs Hospital costs of admissions were calculated by multiplying charges reported in State Inpatient Databases with CCRs from the official HCUP CCR files. The HCUP CCR files were constructed using information on all-payer inpatient costs and charges that hospitals reported to the Centers for Medicare and Medicaid Services. 15 Hospital-specific CCRs were available for two-thirds of the hospitals. For the remaining one-third of hospitals, weighted group average ratios were calculated based on group characteristics such as state, urban/rural, and bed size. In California, CCR were not available before 2006. We investigated the year-to-year changes in hospital CCR in California 2006 to 2009 and found that annual changes of CCR were small (1%-4%). To include year 2005 in our analysis, we assumed that the CCR in 2005 were similar to that of 2006 and used the 2006 CCR of the same hospitals to impute admission costs in 2005. All charges and costs were inflation adjusted and reported in 2010 dollars using annual consumer price indices published by the Bureau of Labor Statistics: cost adjusted ¼ cos 2010 t unadjusted  CC PI PI year . Statistical analyses Summary statistics and age trends were analyzed on and compared between admissions of CHD patients and APs

Lu et al 949

Figure 1

3,024,571 admissions of all patients 10–29 years old extracted from the California State Inpatient Databases 2005–2009, among which 9,107 were of patients with CHD.

Admissions with missing hospital charges (n = 290,182, among which 845 were of patients with CHD.)

Pregnancy admissions (n = 1,539,166, among which 843 were of patients with CHD.)

1,195,233 admissions of all patients, among which 7,419 admissions were of patients with CHD.

CHD, congenital heart disease.

Selection of inpatient admissions.

in 5-year groups (10-14, 15-19, 20-24, and 25-29 years). Prevalence of admissions for CHD patients with only patent foremen ovale (PFO) or atrial septal defect (ASD) was analyzed to better understand how the transition from adolescence to young adulthood differed by CHD complexity. Patent foremen ovale/ASD-only admissions were identified by ≥1 ICD-9 diagnosis codes of 745.5, 747.5, 747.6, or 747.8. Utilization by hospital type (teaching vs nonteaching, children's vs general acute), bed size, and ownership models (public; private, not for profit; and private, for profit) were also analyzed. We hypothesized that these hospital characteristics at least partially accounted for differences in quality of care, complexity of condition treated, and pricing. After examining the most common diagnoses and procedures in the admissions database, we identified 3 types of admissions characteristic of CHD patients in adolescence and young adulthood: (1) cardiac surgery, identified by a detailed list of ICD-9 cardiac surgery procedure codes from Chang et al 16; (2) nonsurgical cardiac interventions represented by catheterization or electrophysiologic (EP) study, identified by ≥1 ICD-9 procedure codes of 37.21 to 37.23 or 37.26; and (3) known sequalae of heart disease represented by arrhythmia or congestive heart failure (CHF), identified by ≥1 ICD-9 diagnosis codes of 426, 427, or 428. The following assigning hierarchy was used when N1 category applied in a single admission: when an admission indicated both a procedure of cardiac surgery and a procedure of catheterization/EP study, it was assigned to the category of cardiac surgery; and when an admission indicated both a procedure of catheterization/EP study and a diagnosis of arrhythmia/CHF, it was assigned as catheterization/EP study.

American Heart Journal December 2014

950 Lu et al

A) Admissions

500 450

438

2%

438

400 350

Millions

Figure 2

B) Costs

18 16 14

13.9 3.33%

1.29%

304

5%

4% 12.6

12

303

3%

300 10 250

1% 7.7

8 200

0.73%

7.2

1.78%

2%

6

150 0.46% 0.38%

100

1.00%

4

0.85%

1%

2

50 -

0% 10–14 yrs

15–19 yrs

annual admissions, CHD patients

20–24 yrs

25–29 yrs

% all admissions attributable to CHD patients

-

0% 10–14 yrs

15–19 yrs

annual admission costs, CHD patients

20–24 yrs

25–29 yrs

% all costs attributable to CHD patients

Annual admissions and costs of CHD patients.A, The left axis shows the number of annual admissions of CHD patients (displayed by bars), and the right axis shows the percent of all admissions in the general population that CHD patients accounted for (displayed by dots with 95% confidence intervals). B, The left axis shows the annual admission costs of CHD patients (displayed by bars), and the right axis shows the percent of all admission costs in the general population that CHD patients accounted for (displayed by square dots).

Multivariate linear regression was performed within CHD patients on the outcome of costs per admission. Explanatory variables included age group; gender; race/ ethnicity; insurance status; CHD complexity (PFO/ASD only or not); types of hospital admissions (cardiac surgery, catheterization/EP study, or arrhythmia/CHF); hospital characteristics, whether admitted from the ED, LOS, and year of admission. Stata (version 12; StataCorp, College Station, TX) was used to perform all data analyses. The study was approved by the Institutional Review Board of Children's Hospital Los Angeles in March 2012 as an exempt study due to the use of deidentified data only (CCI-12-00076). The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper, and its final contents.

Results Number of admissions and total costs The number of annual hospital admissions for patients with CHD stayed relatively stable in adolescence and in young adulthood; however, it dropped by about one-third from 438 per year in the 15 to 19 years age group to 304 per year in the 20 to 24 years age group (Figure 2A). In contrast, among APs of the same age, the number of hospital admissions more than doubled, from 34,059 per year in the 10 to 14 years age group to 79,014 per year in the 25 to 29

years age group. As a result, the proportion of AP admissions that was attributable to CHD patients steadily decreased with age, accounting for 1.29% admissions among the 10 to 14 year olds but for only 0.38% among the 25 to 29 year olds (Figure 2A). A decreasing trend in median hospital costs per admission was observed in CHD patients from $18,286 in the 10 to 14 years age group to $12,326 in the 25 to 29 years age group, whereas among APs, costs remained unchanged with age (Table I). Whereas the total annual admissions costs more than doubled in APs in the general population from $418 million in the 10 to 14 years age group to $856 million in the 25 to 29 years age group, annual costs of CHD patients monotonically decreased by age: in 2005 to 2009, their annual total costs were $13.9 million in the 10 to 14 years age group, $12.6 million in the 15 to 19 years age group, $7.7 million in the 20 to 24 years age group, and $7.2 million in the 25 to 29 years age group, representing 3.33%, 1.78%, 1.00%, and 0.85% of the annual total costs in APs of the same age (Figure 2B).

Trends in hospital type There was a clear shift from children's hospitals to nonchildren's hospitals for both CHD patients and APs when they transitioned from adolescence to young adulthood (Table I). The shift began in early-to-late adolescence and accelerated in young adulthood.

American Heart Journal Volume 168, Number 6

Lu et al 951

Table I. Summary statistics of inpatient admissions of CHD patients and APs Patients with CHD (n = 7419) Age group

10-14 (n = 2190) 45.2 1.6

15-19 (n = 2191) 42.6 1.8

20-24 (n = 1522) 47.4 1.4

APs (n = 1,195,233) 25-29 (n = 1516) 47.0 7.3

10-14 (n = 170,294) 43.3 7.5

15-19 (n = 298,622) 45.5 7.9

20-24 (n = 331,235) 44.6 4.1

25-29 (n = 395,072) 43.0 13.0

Female, % Unknown gender, % Race, % White 36.8 38.0 43.8 42.9 29.9 35.4 39.2 35.5 Black 5.1 7.3 6.8 6.1 6.6 6.8 8.5 8.4 Hispanics 40.6 34.6 30.5 30.7 35.1 29.8 31.0 28.1 Other 9.8 7.3 7.0 6.1 5.1 3.8 4.8 4.7 Unknown 7.7 12.7 12.0 14.1 23.4 24.2 16.5 23.3 PFO/ASD only 11.2 11.6 16.1 21.3 0.1 0.1 0.1 0.1 Children's 46.0 35.2 7.3 3.0 29.5 11.3 0.8 0.2 hospital, % Teaching 42.4 39.1 44.3 45.4 22.6 22.6 24.3 22.8 hospital, % Average bed size, % 0-99 1.2 2.0 3.8 2.2 12.1 14.0 10.6 11.9 100-299 31.8 28.3 19.2 20.8 33.0 31.9 30.6 30.9 300+ 57.8 59.0 71.4 72.0 46.6 45.7 52.0 50.2 Unknown 9.1 10.6 5.7 5.0 8.3 8.4 6.9 6.9 Hospital ownership, % Public 25.3 23.3 33.1 35.0 19.3 22.7 28.6 26.9 Private, not for 63.2 60.9 51.4 50.6 59.3 49.1 45.8 46.0 profit Private, for 2.3 5.1 9.8 9.5 13.2 19.9 18.6 20.2 profit Unknown 9.1 10.6 5.7 5.0 8.3 8.4 7.0 7.0 Cost per 18,286 15,344 12,244 12,326 6099 5961 6284 6185 admission, $ (8174-41,243) (7023-37,938) (5472-31,434) (5910-29,316) (3279-11,524) (3081-11,403) (3252-11,549) (3267-11,556) Insurance, % Public 42.7 41.2 40.9 41.6 44.5 39.4 33.4 35.5 Private 40.1 44.0 39.9 42.3 43.3 45.8 34.1 34.5 Uninsured 0.7 2.1 6.8 5.1 1.9 4.5 13.6 12.2 Unknown 16.4 12.6 12.5 11.0 10.2 10.3 18.9 17.8 Admissions 20.6 31.7 49.7 53.9 40.8 48.7 61.8 58.1 from ED, % LOS, d 3 (1-6) 3 (1-6) 3 (2-6) 3 (2-6) 3 (1-6) 3 (2-6) 3 (2-5) 3 (2-5) Abbreviations: CHD, congenital heart diseases; PFO, patent foremen ovale; ASD, atrial septal defect; ED, emergency department; LOS, length of stay. Data are presented as median interquartile range (IQR) or percent.

Among CHD patients, almost half of the admissions occurred in children's hospitals in the 10 to 14 years age group, but the utilization rate decreased by approximately 11% points in late adolescence and then sharply dropped to 7% and then 3% in young adulthood. Similar trends existed in APs, except for overall fewer admissions to children's hospitals. There was also a slight trend toward higher utilization of high-volume hospitals (bed size, ≥300); however, utilization of teaching hospitals stayed relatively unchanged for both CHD patients and APs.

Access to care Admissions of APs and CHD patients displayed similar age trends in the distribution of insurance type: compared with both younger and older cohorts, the 20 to 24 years olds had the highest percentage of no insurance and the lowest

percentage of private insurance (Table I). An important indicator of access to care, percentage of admissions from the ED, was consistently lower in CHD patients than in APs across all age groups (Table I). However, the gap was narrowed due to the percentage of ED admissions more rapidly increasing as CHD patients aged: for the 10 to 14 years age group, the rate of ED admissions was 23.3% points lower in CHD patients than in APs, yet the difference was reduced to a mere 4.2% points in the 25 to 29 years age group (Table I). Percentage of ED admissions differed by insurance type: across all age groups, those with private insurance were the least likely to be admitted from the ED in both APs and CHD patients (Figure 3). But across all insurance types, admissions from ED had a greater increase with age in CHD patients than in APs (Figure 3).

American Heart Journal December 2014

952 Lu et al

Figure 3 90%

10-14 yrs 15-19 yrs 20-24 yrs 25-29 yrs

80% 70%

83%

82%

81%

78%

78%

65% 62%

60%

62%

61%

56%

55%

54%

53% 49%

50% 44%

43%

40%

27%

25% 19%

20%

Public insurance

Private insurance

No insurance

Public insurance

Private insurance

CHD patients

n = 39,332

n = 37,413

n = 2,012

n = 10,566

n = 66,975

n = 59,811

n = 28,479

n = 58,056

n = 85,103

n = 71,697

n = 33,298

n = 63,808

n = 60

n = 84

n=*

n = 26

n = 276

n = 245

n = 231

n = 393

n = 252

n = 351

0%

n = 341

10%

16%

n = 144

30%

42% 39%

38%

38%

No insurance

All patients

Percent admissions from the ED by insurance type and age.⁎Sample size b10 is not reported for protection of patient privacy, stipulated by HCUP data publication rules.

Typical admissions among CHD patients Cardiac surgery was a major reason for admissions among adolescent CHD patients (Figure 4). However, cardiac surgery became less prevalent, as CHD patients transitioned into young adulthood, dropping from 33.2% of all CHD admissions in the 10 to 14 years age group to 20.8% in the 25 to 29 years age group. Meanwhile, similar yet less drastic decreases were observed in admissions due to catheterization/EP study as nonsurgical cardiac interventions. In contrast, admissions due to CHD sequalae, namely, arrhythmia/CHF, saw a large increase of prevalence with age: they accounted for 9.7% of all CHD admissions in the 10 to 14 years age group and rose to 25.7% in the 25 to 29 years age group. Similar to arrhythmia and CHF, admissions of PFO/ASDonly patients also greatly increased with age, from approximately 11% to 12% in adolescence to 16.1% in the 20 to 24 years age group and then to 21.3% in the 25 to 29 years age group (Table I). Cost factors within CHD patients Cardiac surgery had the largest cost impact of all factors and was associated with an increase of $29,543 in costs per admission (P b .001) (Table II). Other factors predicting increased costs included being admitted to a children's hospital ($10,355) or a teaching hospital ($8,315), admission due to catheterization/EP study

($7,803) or arrhythmia/CHF ($4,277), having private insurance ($2,636), and longer LOS ($2,595), all P b .001. At the same time, admission costs were negatively associated with being female (−$1,433) and having only PFO/ASD (−$5,860), P b .001. There were no significant racial variations in admission costs. Emergency department admissions, hospital bed size, or ownership model was not statistically insignificant as well. Age effects were largely nonexistent, when other factors such as plausible reasons of admission, CHD complexity, and hospital characteristics were being controlled for.

Discussion Although the number of admissions related to surgical and nonsurgical cardiac interventions progressively decreases, the number of medical admissions for arrhythmia and CHF increases from adolescence to young adulthood. Given the natural history of CHD, it is reasonable to assume that unlike surgical and catheterization/EP study admissions that are often planned, admissions for medical indications such as arrhythmias and CHF are more likely to occur unexpected and require urgent care. Our findings are consistent with the current literature that CHD patients without private insurance are more likely to use the ED as their usual source of care. 8 Our

American Heart Journal Volume 168, Number 6

Lu et al 953

Figure 4 40% Cardiac surgery 35%

33.2%

Catheterization or EP study Arrhythmia or CHF

30%

25.7%

25.0%

24.2%

25%

20.8% 19.8%

20% 15.8% 13.8%

13.1% 10.6%

10-14 yrs

15-19 yrs

20-24 yrs

n = 389

n = 316

n = 381

n = 162

n = 302

n = 346

n = 287

n = 530

n = 213

n = 302

0%

n = 727

5%

9.8%

9.7%

10%

n = 148

15%

25-29 yrs

Age trend of typical admissions among CHD patients.

study also confirmed the findings by Gurvitz et al 8 that the percentage of admissions from the ED doubles to N40%, as patients enter young adulthood, raising the issue of access to care. In our study, we found that the percentage of ED admissions increased for CHD patients, as they transitioned into young adulthood, and the increase was more pronounced in those uninsured and/ or with public insurance. Indeed, patients with public insurance had a 35%-point increase in ED admissions from adolescence to young adulthood, suggesting that limited access to care is a significant driver for ED admissions in CHD patients. This is important because limited access to care is common among young adults who age out of Medicaid and/or lose their dependent status in their parents' private insurance. In fact, young adults aged 19 to 39 years have the lowest percentage of health care insurance of any age group b65 years. 17 Because of the lack of access to coordinated care and/or delays in seeking appropriate outpatient ACHD management care, young adults with CHD would be more likely to use expensive inpatient care for avoidable conditions and develop secondary complications that will incur greater costs in the long term. Furthermore, even among CHD patients with private insurance who presumably had the highest access to care, we observe a 27%-point increase in ED admissions from the 10 to 14 to the 25 to 29 years age group, which was much N10% increase among APs with private insurance. This 27%-point increase in ED admissions

among privately insured CHD patients indicated that insurance coverage alone may not be able to fully meet all the special health care needs of young adult CHD patients, unlike their counterparts of the same age in the general population. The occurrence of CHD-related emergencies such as incident arrhythmia attacks and CHF often require emergent care. The natural history of CHD may result in lower inpatient costs from fewer cardiac surgeries, but the rise of CHD sequalae and secondary complications call for an in-depth look into whether proactive case management of young adults with CHD will reduce the need for urgent admission. The decline of cardiac surgical admissions and the rise of CHD sequalae with age suggest that the natural history of disease is an important factor for the changing trend in admission types, as young CHD patients emerge as adults. Cardiac surgery is the single most influential cost factor for hospitalizations in the CHD population, adding N$29,000 in costs per admission. Therefore, the decline of its prevalence in young adulthood by 12.4% points may explain to a great extent the significant decrease in total admission costs incurred by young ACHD patients, despite that the shift from the more expensive children's hospitals to general acute hospitals also partially explains a decline in costs. Fewer surgeries due to the natural history of CHD predict a relatively quiet period of CHD resource utilization in young adulthood, which is compounded by the lack of access to ACHD specialty care 4,5 and a rise of PFO/ASD admissions with lower CHD complexity. To our knowledge,

American Heart Journal December 2014

954 Lu et al

Table II. Multivariate regression results on costs per admission among CHD patients Coefficient (in 2010 dollars) Age group (reference: 10-14 years old) 15-19 years old 20-24 years old 25-29 years old Gender (reference: male) Female Race (reference: white) Black Hispanics Other PFO/ASD only Children's hospital Teaching hospital Average bed size (reference: 0-99) 100-299 300+ Hospital ownership (reference: public) Private, not for profit Private, for profit Plausible reasons of admission Cardiac surgery Catheterization or EP study Arrhythmia or CHF Insurance (reference: public insurance) Private Uninsured Unknown Admissions from ED LOS

95% CI

P

1094 2632 1057

−414 to 2603 559-4164 −799 to 2913

.155 .010 .264

−1433

−2611 to −255

.017

976 −1317 1123 −5860 10,355 8315

−1547 to 3499 −2770-137 −1149 to 3395 −7539 to −4181 8503-12,206 6702-9929

.448 .076 .332 b.001 b.001 b.001

4086 2512

−111 to 8282 −1653 to 6677

.056 .237

782 −126

−894 to 2458 −3110 to 2858

.361 .934

29,543 7803 4277

27,947-31,140 5866-9740 2656-5898

b.001 b.001 b.001

2636 84 −124 −1059 2595

1263-4009 −3237 to 3404 −1954 to 1705 −2481 to 362 2531-2658

b.001 .961 .894 .144 b.001

Abbreviations: CI, confidence intervals; CHD, congenital heart disease; PFO, patent foremen ovale; ASD, atrial septal defect; EP, electrophysiology; CHF, congestive heart failure; ED, emergency department; LOS, length of stay. Adjusted R2 = 0.56. Individual years were included in the regression but not shown in the table.

this article is among the first to demonstrate the role of natural history of disease as an important contributor to resource utilization and health care costs of adolescents and young adults with CHD in recent years. One direction for further research is to disentangle and measure the independent impacts on health care utilization and patient outcomes in CHD by natural history of CHD and access to care, respectively. A better understanding of these impacts will help guide us in optimal resource planning in the treatment and secondary prevention for emerging adults with CHD. The Patient Protection and Affordable Care Act allows individuals to be covered under their parents' private insurance plan up to age 26 years, effective September 2010, which provides a great natural experiment to study the impact of exogenously extended access to care. Another direction for future research is to conduct longitudinal analyses of adolescents and young adults with CHD and their health care utilization over time, which will shed more light on how natural history of disease and transition of care impact individual patients. More accurate parameters can be estimated for utilization and cost projections based on individual characteristics, disease complexity, and other risk factors, which in turn

could help improve ACHD case management and resource planning for these individuals in transition.

Study limitations The study is cross-sectional and retrospective in nature and addresses only the inpatient care of CHD patients. The literature is replete with evidence that a large body of adult CHD patients receive no care. 18 The ultimate health care costs due to this lack of care are beyond the scope of this study and may accumulate to substantial amounts in future years. In addition, the data we used report hospital costs and charges but do not include physician fees, so the costs reported in the study are an underestimate of total admissions costs. And because of the nature of administrative data, the accuracy of the study population selection and data analyses depends on how accurately diagnostic and procedural codes are assigned. For example, we were not able to observe the differences in how children's hospitals and other hospitals code CHD admissions, which could be one reason that fewer CHD admissions are observed in young adulthood. In addition, this article does not provide estimates on what fraction of the total decrease in CHD costs with age is due to lower utilization of children's hospitals and what

American Heart Journal Volume 168, Number 6

fraction is due to natural history of CHD, that is, lower demand for cardiac interventions. If utilization of the more costly children's hospitals with unique resources is positively associated with CHD complexity, higher costs at these hospitals may reflect actual needs of more complex patients for cardiac surgery other than their pricing practices. Further sophisticated decomposition analyses would be required to answer this question. Finally, our findings might not be generalizable to other states due to differences in demographics and factors that impact access to inpatient care such as availability of ACHD management programs and eligibility requirements for public insurance coverage. Nevertheless, one important factor identified by the study, natural history of disease, is important in planning the ultimate resource utilization for CHD patients, regardless of geographic location.

Conclusion In the CHD population, multiple factors influence health care utilization during the second and third decades of life, including the natural history of CHD and associated noncardiac diseases as well as access to inpatient care. Total inpatient health care costs in CHD decline with age in young adulthood due to fewer admissions for cardiac surgery and catheterization/EP study, among other factors. The shift from procedure-related to therapeutic/diagnostic admissions is reflective of the evolving natural history of CHD. Increased admissions from the ED is likely a result of urgent management of unplanned arrhythmia or CHF as well as the use of the ED as a source of care especially among publicly insured and noninsured CHD patients. This study provides an important baseline from which to understand and provide for the health care needs of emerging adults with CHD.

Acknowledgements This study was funded by the California Healthcare Foundation. The authors thank Drs Alan Lewis and Robertson Parkman at Children's Hospital Los Angeles for their helpful comments.

Disclosures The authors do not have any conflicts to disclose.

References 1. Hoffman JIE, Kaplan S, Liberthson RR. Prevalence of congenital heart disease. Am Heart J 2004;147(3):425-39. 2. O'Leary JM, Siddiqi OK, de Ferranti S, et al. The Changing Demographics of Congenital Heart Disease Hospitalizations in the United States, 1998 Through 2010. JAMA 2013;309(10):984-6.

Lu et al 955

3. Opotowsky AR, Siddiqi OK, Webb GD. Trends in hospitalizations for adults with congenital heart disease in the U.S. J Am Coll Cardiol 2009;54(5):460-7. 4. Fernandes SM, Khairy P, Fishman L, et al. Referral patterns and perceived barriers to adult congenital heart disease care: results of a survey of US pediatric cardiologists. J Am Coll Cardiol 2012;60(23): 2411-8. 5. Kovacs AH, Verstappen A. The whole adult congenital heart disease patient. Prog Cardiovasc Dis 2011;53(4):247-53. 6. Gurvitz M, Valente AM, Broberg C, et al. Prevalence and predictors of gaps in care among adult congenital heart disease patients: HEART-ACHD (The Health, Education, and Access Research Trial). J Am Coll Cardiol 2013;61(21):2180-4. 7. Banerjee R, Ziegenfuss JY, Shah ND. Impact of discontinuity in health insurance on resource utilization. BMC Health Serv Res 2010;10:195. 8. Gurvitz MZ, Inkelas M, Lee M, et al. Changes in hospitalization patterns among patients with congenital heart disease during the transition from adolescence to adulthood. J Am Coll Cardiol 2007;49 (8):875-82. 9. Moons P, Pinxten S, Dedroog D, et al. Expectations and experiences of adolescents with congenital heart disease on being transferred from pediatric cardiology to an adult congenital heart disease program. J Adolesc Health 2009;44(4):316-22. 10. Moons P, Hilderson D, Van Deyk K. Implementation of transition programs can prevent another lost generation of patients with congenital heart disease. Eur J Cardiovasc Nurs 2008;7(4):259-63. 11. Yeung E, Kay J, Roosevelt GE, et al. Lapse of care as a predictor for morbidity in adults with congenital heart disease. Int J Cardiol 2008;125(1):62-5. 12. Wray J, Frigiola A, Bull C. Adult congenital heart disease research N. Loss to specialist follow-up in congenital heart disease; out of sight, out of mind. Heart 2013;99(7):485-90. 13. Casey PH, Lyle RE, Bird TM, et al. Effect of hospital-based comprehensive care clinic on health costs for Medicaid-insured medically complex children. Arch Pediatr Adolesc Med 2011;165(5): 392-8. 14. James AH, Jamison MG, Biswas MS, et al. Acute myocardial infarction in pregnancy: a United States population-based study. Circulation 2006;113(12):1564-71. 15. Agency for Healthcare Research and Quality. Cost-to-charge ratio files. http://www.hcup-us.ahrq.gov/db/state/costtocharge.jsp. [Accessed January 10, 2014]. 16. Chang R-KR, Chen AY, Klitzner TS. Female sex as a risk factor for in-hospital mortality among children undergoing cardiac surgery. Circulation 2002;106(12):1514-22. 17. Nicholson JL, Collins SR. Young, uninsured, and seeking change: health coverage of young adults and their views on health reform. Findings from the Commonwealth Fund Survey of Young Adults (2009). Issue Brief (Commonw Fund) 2009;73:1-22. 18. Warnes CA, Williams RG, Bashore TM, et al. ACC/AHA 2008 Guidelines for the Management of Adults with Congenital Heart Disease: Executive Summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (writing committee to develop guidelines for the management of adults with congenital heart disease). Circulation 2008;118(23): 2395-451.

Inpatient admissions and costs of congenital heart disease from adolescence to young adulthood.

Health care utilization patterns and associated costs of congenital heart disease (CHD) in young adulthood are not well understood...
703KB Sizes 0 Downloads 7 Views