- Email: [email protected]
Congenital Heart Disease Hospitalizations in Canada: A 10-Year Experience
Canadian Journal of Cardiology
-
(2015) 1e7
Clinical Research
Congenital Heart Disease Hospitalizations in Canada: A 10-Year Experience Sunjidatul Islam, MBBS, MSc,a Yutaka Yasui, PhD,a Padma Kaul, PhD,a,b Ariane J. Marelli, MD, MPH,c and Andrew S. Mackie, MD, SMa,d,e a b c
School of Public Health, University of Alberta, Edmonton, Alberta, Canada
Department of Medicine, University of Alberta, Edmonton, Alberta, Canada
McGill Adult Unit for Congenital Heart Disease Excellence, McGill University, Montreal, Quebec, Canada d e
Stollery Children’s Hospital, Edmonton, Alberta, Canada
Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada
ABSTRACT
RESUM E
Background: The effect of the growing population of children and adults with congenital heart disease (CHD) on inpatient services in Canada is not known. We sought to assess temporal changes in hospitalizations of CHD patients. Methods: We identified all patients with a CHD diagnosis who received inpatient care in Canada between fiscal years 2003 and 2012 according to the Discharge Abstract Database of the Canadian Institute for Health Information. Poisson regression was performed to assess temporal changes in the annual hospitalization rate. Hospitalization rates were indexed to the general population and the estimated CHD population. Results: A total of 103,034 hospitalizations occurred in 61,051 patients from fiscal years 2003 to 2012. The absolute number of hospitalizations increased by 4.0% per year in adults and 1.3% per year in children. The greatest increase was in patients aged 65 years
Introduction : L’incidence de la population croissante d’enfants et nitale sur les services hospid’adultes atteints de cardiopathie conge à e valuer les taliers canadiens est inconnue. Nous avons cherche variations temporelles en matière d’hospitalisation chez les patients nitale. atteints de cardiopathie conge thodes : À l’aide de la base de donne es sur les conge s hospitaliers Me , nous avons recense de l’Institut canadien d’information sur la sante nitale ayant e te hostous les patients atteints de cardiopathie conge s au Canada entre les anne es financières 2003 et 2012. Une pitalise gression de Poisson a e te effectue e afin d’e valuer les variations re temporelles des taux d’hospitalisation annuels. Ces taux d’hospitali te indexe s en fonction de la population ge ne rale et de la sation ont e e de patients atteints de cardiopathie conge nitale. population estime sultats : Entre les anne es financières 2003 et 2012, il y a eu au Re total 103 034 hospitalisations pour 61 051 patients. Le nombre
With improvements in survival,1 the prevalence of congenital heart disease (CHD) has increased rapidly in adults, and adults with CHD now outnumber children with CHD.2,3 In addition, the distribution of CHD has changed with an increased prevalence of severe CHD among adults relative to children.2-4 Adult patients with CHD are at risk of development of late cardiac complications, which sometimes require repeat interventions.4-6 Moreover, they might acquire other comorbid conditions with aging. The effect of this changing
demographic characteristic on the number of adult CHD clinic visits has been previously described in Canada.7 However, the effect on inpatient care is not known. Therefore, the objectives of this study were: (1) to determine temporal changes in the hospitalization rate among CHD patients in Canada, overall and according to age, sex, and severity of CHD; (2) to determine temporal changes in length of hospital stay, overall and among specific patient subgroups; and (3) to identify predictors of length of stay (LOS) > 14 days. Methods
Received for publication February 24, 2015. Accepted May 29, 2015. Corresponding author: Dr Andrew S. Mackie, Division of Cardiology, Stollery Children’s Hospital, 4C2 Walter C. Mackenzie Center, 8440-112th St NW, Edmonton, Alberta T6G 2B7, Canada. Tel.: þ1-780-407-8361; fax: þ1-780-407-3954. E-mail: [email protected] See page 7 for disclosure information.
Study design and data source We conducted an observational retrospective cohort study using the Discharge Abstract Database from the Canadian Institute for Health Information, which consists of hospitalization records from all acute care hospitals in Canada, except
http://dx.doi.org/10.1016/j.cjca.2015.05.022 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
2
Canadian Journal of Cardiology Volume - 2015
followed by those 40-64 years. However, the hospitalization rate in adults varied between 39 and 55 per 1000 CHD population with a reduction of 4% per year (95% confidence interval, 0.95-0.96; P < 0.001). The hospitalization rate in children ranged from 79 to 87 per 1000 CHD population and did not change significantly over time (rate ratio, 1.00; 95% confidence interval, 1.00-1.01; P ¼ 0.035). Men accounted for 53.5% of hospitalizations. Conclusions: The absolute number of hospitalizations of patients with CHD increased over time in children and adults. However, the hospitalization rate relative to the CHD population decreased among adults, possibly reflecting improved outpatient management. The absolute increase in CHD hospitalizations will pose a financial burden on health care systems.
e chez les adultes absolu d’hospitalisations s’est accru de 4 % par anne e chez les enfants. La plus forte augmentation a et de 1,3 % par anne te enregistre e chez les patients âge s de 65 ans et plus suivie des e s de 40 à 64 ans. On a toutefois observe un taux d’hospatients âge pitalisation variant entre 39 et 55 par 1000 patients adultes atteints nitale et une re duction de ce taux de 4 % par de cardiopathie conge e (intervalle de confiance à 95 % 0,95 à 0,96; P < 0,001). Chez anne les enfants, le taux d’hospitalisation se situait entre 79 et 87 par 1000 nitale et n’a pas varie de patients atteints de cardiopathie conge manière significative au fil du temps (ratio des taux, 1,00; intervalle de confiance à 95 % 1,00 à 1,01; P ¼ 0,035). Les hommes ont fait l’objet de 53,5 % des hospitalisations. Conclusions : Au fil du temps, le nombre absolu d’hospitalisations de nitale a augmente tant chez les patients atteints de cardiopathie conge une diminution du adultes que chez les enfants. On a toutefois observe taux d’hospitalisation chez les adultes atteints de cardiopathie connitale, ce qui pourrait s’expliquer par une meilleure prise en charge ge ambulatoire. Il est à noter que l’augmentation absolue du nombre d’hospitalisations chez les patients atteints de cardiopathie connitale imposera un fardeau financier au système de sante du pays. ge
those from the province of Quebec. The database includes data on admission and discharge dates, patient demographic variables, diagnoses (a primary and up to 24 secondary diagnoses), diagnostic and therapeutic procedures (up to 16 procedures), and discharge disposition. Multiple hospitalizations for the same patient can be tracked with the use of a unique anonymous patient identification number.
CHD was applied to the calculation of annual hospitalization rates according to age, sex, and severity of CHD. In addition, we indexed the annual hospitalization rate to the CHD population of Canada (less the Yukon, Northwest Territories, Nunavut, and Quebec), which was estimated using recent Quebec CHD prevalence data, published by Marelli et al.3 They reported the prevalence of CHD in Quebec for the years 2005 and 2010 for children and adults. We estimated the size of the CHD population on an annual basis using CHD prevalence data from 2005 and midyear general population statistics from Statistics Canada for fiscal years 2003-2007, and prevalence data from 2010 for fiscal years 2008-2012. The CHD population for children and adults were estimated separately using the same method. LOS was the secondary outcome of interest. In the event of a hospitalization within 24 hours of discharge, the LOS was measured as the duration between the first admission date and the second discharge date. We also assessed the total LOS, defined as the total duration of hospital days (over multiple hospitalizations) per person over the 10-year study period.
Study population Inclusion criteria. Our study population consisted of all patients, irrespective of their age, with a diagnosis of CHD who received inpatient care during the fiscal years 2003-2012 in Canada, less the Yukon, Northwest Territories, Nunavut, and Quebec. CHD was identified based on International Classification of Disease (ICD) Revisions 9 and 10, from any of 25 diagnostic fields. Exclusion criteria. We excluded patients with isolated patent ductus arteriosus and those who lived in the Yukon, Northwest Territories, Nunavut, or Quebec even if they received care in hospitals of other provinces. We also excluded patients who were discharged alive on the same day of admission, with the assumption that they were admitted for a day procedure. Hospitalizations within 24 hours of discharge were considered hospital-to-hospital transfers and therefore counted as single hospitalizations. Outcome measures Our primary outcome of interest was the annual number and rate of hospitalizations with CHD as a primary or secondary diagnosis and the annual number and rate of hospitalizations according to age group, sex, severity of CHD, and therapeutic intervention. For each year, we calculated the hospitalization rate by dividing the number of hospitalizations by the population size of Canada (less the Yukon, Northwest Territories, Nunavut, and Quebec) in the given year. The same procedure stratified according to age, sex, and severity of
Independent variables CHD lesions were grouped into 3 complexities: simple, moderate, and complex, defined according to the 32nd Bethesda conference.8 We classified patients with multiple CHD diagnoses based on the hierarchies of complexities. For example, if a patient had complex lesions and moderate or simple lesions, the patient was classified as a complex CHD patient. Similarly, patients with moderate and simple CHD diagnoses were categorized as moderate CHD patients. For patients having multiple hospitalizations during the study period, all CHD diagnoses coded across the multiple hospitalizations were used in classifying the patient as simple, moderate, or complex. Age was categorized into 6 categories: infants (< 1 year), 1-4 years, 5-17 years, 18-39 years, 40-64 years, and 65 years. We used Canadian Classification of Health Interventions codes corresponding to ICD-10 codes and Canadian
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
Classification of Diagnostic, Therapeutic, and Surgical Procedures codes corresponding to ICD-9 codes to identify the therapeutic cardiovascular intervention performed during hospitalizations. The hospitalizations were grouped into 4 categories based on the intervention procedure performed: no cardiac intervention, cardiac surgery only, cardiac catheterization only, or cardiac surgery and catheterization. Statistical analysis Characteristics of index hospitalizations (ie, characteristics of patients at the first hospitalization during the study period) and all hospitalizations were described according to frequency distributions (proportions) for categorical variables and median and interquartile range (IQR) for LOS, which was treated as a continuous variable. The 10 most common primary diagnoses during index and all hospitalizations were described with frequency distributions for children and adults separately. We expressed the annual hospitalization rate per 100,000 persons and per 1,000 of the CHD population. Poisson regression analysis was performed to assess temporal changes in the annual hospitalization rate. The numbers and rates of annual hospitalizations were also analyzed stratified according to age group, sex, and severity of CHD. The hospitalization rates indexed to the CHD population were stratified according to child or adult hospitalization. LOS during the index hospitalization and total LOS over the 10year study period were compared using Kruskal-Wallis (if more than 2 groups; eg, age) and Wilcoxon rank sum tests (if 2 groups; eg, sex) across the categories of age, sex, and severity of CHD. A multivariable logistic generalized estimating equations model was used to identify factors associated with LOS > 14 days. Two sided P values < 0.05 were considered statistically significant. All analyses were performed using SAS statistical software (version 9.4; SAS Institute, Cary, NC). Ethics All procedures of the study were approved by the Health Research Ethics Board at the University of Alberta. Results A total of 103,034 inpatient admissions with CHD as a primary or secondary diagnosis occurred between fiscal years 2003 and 2012 among 61,051 patients. Patient characteristics of the cohort are described in Table 1. In-hospital death occurred in 3730 (6.1%) patients and 1965 (52.7%) of all inhospital deaths occurred at the time of the index hospitalization. The most common primary diagnoses responsible for hospitalizations are reported in Table 2 for children and adults. Infants accounted for 39.2% of all hospitalizations followed by 40- to 64-year-olds (18.3%), and patients aged 65 years (16.3%). There were more hospitalizations among men (53.5%). Of all hospitalizations, 55.8% were associated with simple CHD, 32.3% were associated with moderate CHD, and 12.0% were associated with complex CHD. Among children, 28,837 (77.3%) had a single hospitalization, 4477 (12.0%) had 2 hospitalizations, and 3988 (10.7%) had 3 or more hospitalizations. Among adults, 14,028 (59.1%) had a single hospitalization, 4838 (20.4%) had 2 hospitalizations,
3
Table 1. Demographic and clinical characteristics of study subjects Characteristic Fiscal year 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Age group < 1 year 1-4 years 5-17 years 18-39 years 40-64 years 65 years Sex Female Male Median length of stay (IQR), days Severity of CHD Simple Moderate Complex Cardiac intervention No intervention Cardiac surgery Cardiac catheterization Cardiac surgery and catheterization
Index hospitalizations
All hospitalizations
7267 6460 6335 6155 5820 5968 5769 5783 5895 5599
(11.9) (10.6) (10.4) (10.1) (9.5) (9.8) (9.5) (9.5) (9.7) (9.2)
9217 9530 9964 9984 9832 10,449 10,628 10,909 11,338 11,183
(9.0) (9.3) (9.7) (9.7) (9.5) (10.1) (10.3) (10.6) (11.0) (10.9)
30,791 2629 4138 6262 10,614 6617
(50.4) (4.3) (6.8) (10.3) (17.4) (10.8)
40,420 8864 7818 10,283 18,882 16,767
(39.2) (8.6) (7.6) (10.0) (18.3) (16.3)
28,499 (46.7) 32,552 (53.3) 6.2 (2.9-15.7)
47,887 (46.5) 55,147 (53.5) 5.8 (2.6-13.2)
38,316 (62.8) 18,661 (30.6) 4074 (6.7)
57,455 (55.8) 33,271 (32.3) 12,308 (12.0)
43,072 12,993 3978 1008
72,325 21,721 6950 2038
(70.6) (21.3) (6.5) (1.7)
(70.2) (21.1) (6.8) (2.0)
Data are presented as n (%) except where otherwise stated. CHD, congenital heart disease; IQR, interquartile range.
and 4883 (20.6%) had 3 hospitalizations. The absolute numbers of annual hospitalizations from fiscal years 20032012 are summarized according to age group, sex, severity of CHD, and intervention in Supplemental Table S1. The most frequent CHD diagnoses and the most frequent combinations of 2 CHD diagnoses are presented in Supplemental Tables S2 and S3. The absolute number of hospitalizations increased from 9217 in 2003 to 11,183 in 2012, an average increase of 2.4% per year. The annual hospitalization rate varied between 38 and 43 per 100,000 persons during the study period with an increase of 1.1% per year (95% confidence interval [CI], 1.00-1.013; P < 0.001). The overall annual hospitalization rate varied between 55 and 70 per 1000 CHD population with a reduction of 3% per year from 2003 to 2012 (95% CI, 0.97-0.98; P < 0.001). Age groups The number of hospitalizations increased by 4.0% per year in adults (Fig. 1). The number of hospitalizations increased by 6.5% per year for patients aged 65 years, 3.0% per year for 40- to 64-year-olds, and 2.1% per year for 18- to 39-yearolds. In children, the number of hospitalizations increased by 1.3% per year (Fig. 1); an increase of 2.7% was observed in 1to 4-year-olds, and an increase of 2.1% in 5- to 17-year-olds. In contrast, the number of hospitalizations did not change
4
Canadian Journal of Cardiology Volume - 2015
Table 2. Most frequent primary diagnosis Index hospitalizations
All Hospitalizations
Children (< 18 years old) N Cardiac Ventricular septal defect Ostium secundum atrial septal defect Coarctation of the aorta Other specified anomalies of heart Other anomalies of heart Tetralogy of Fallot Transposition of great arteries Endocardial cushion defect Noncardiac Down syndrome Acute bronchiolitis Adults N Cardiac diagnoses Nonrheumatic aortic valve disorder Ostium secundum atrial septal defect Ischemic heart disease Aortic valve abnormality Atrial tachyarrhythmia* Congestive heart failure Atherosclerosis of coronary artery without angina Nonrheumatic mitral valve disorder Infective endocarditis Noncardiac Pneumonia
37,558 2570 1867 935 845 793 728 661 547
57,102
(6.8) (5.0) (2.5) (2.2) (2.1) (1.9) (1.8) (1.5)
683 (1.8) 596 (1.6)
Ventricular septal defect Ostium secundum atrial septal defect Tetralogy of Fallot Coarctation of the aorta Other specified anomalies of heart Transposition of the great arteries Endocardial cushion defect Pulmonary artery stenosis
3331 2186 1419 1213 1116 1090 1058 1005
Pneumonia Acute bronchiolitis
1544 (2.7) 1380 (2.4)
23,493 3142 2036 1527 1212 1207 956 870 561 432
(5.8) (3.8) (2.5) (2.1) (2.0) (1.9) (1.9) (1.8)
45,932
(13.4) (8.7) (6.5) (5.2) (5.1) (4.1) (3.7) (2.4) (1.8)
Congestive heart failure Nonrheumatic aortic valve disorder Atrial tachyarrhythmia* Ischemic heart disease Ostium secundum atrial septal defect Aortic valve abnormality Atherosclerosis of coronary artery without angina Nonrheumatic mitral valve disorder Infective endocarditis
262 (1.1)
Pneumonia
4065 4035 3588 2756 2564 1474 1465 824 727
(8.9) (8.8) (7.8) (6.0) (5.6) (3.2) (3.2) (1.8) (1.6)
734 (1.6)
Data are presented as n (%) except where otherwise stated. * Includes atrial fibrillation, atrial flutter, supraventricular tachycardia, atrial premature beats, sinoatrial node dysfunction, or pre-excitation syndrome.
significantly in infants (< 1 year of age). However, infants consistently had the highest annual hospitalization rate over the 10-year period followed by 1- to 4-year-olds and patients aged 65 years (Fig. 2). With respect to changes in hospitalization rates, the hospitalization rate in infants decreased by a factor of 0.992 per year (95% CI, 0.989-0.996; P < 0.001). However, the annual hospitalization rate increased per fiscal year by 1.7% in patients aged 1-4 years (95% CI, 1.010-1.025; P < 0.001), 3.1% in patients aged 5-17 years (95% CI, 1.002-1.039; P < 0.001), 1.7% in patients aged 18-39 years (95% CI, 1.0101.024; P < 0.001), 1.0% in patients aged 40-64 years (95% CI, 1.005-1.015; P < 0.001), and 2.5% in patients aged 65 years (95% CI, 1.020-1.031; P < 0.001). The
hospitalization rate in children ranged from 79 to 87 per 1000 CHD population and did not change significantly over time (rate ratio, 1.00; 95% CI, 1.00-1.01; P ¼ 0.035). In contrast, the hospitalization rate in adults varied between 39 and 55 per 1000 CHD population with a reduction of 4% per year from 2003 to 2012 (95% CI, 0.95-0.96; P < 0.001). Sex The number of hospitalizations increased by 2.7% per year for men and by 2.0% per year for women. The annual hospitalization rate increased by 1.5% per year in men (95% CI, 1.012-1.018; P < 0.001) and by 0.7% in women (95% CI, 1600
Annual hospitalization rate per 100,000 persons
1500 7000
1400
Number of hospitalizations
6000
1300 100
5000 4000 Children
3000
Adults 2000 1000 0 2003
2004
2005
2006
2007 2008 Fiscal year
2009
2010
2011
2012
Figure 1. Changes in number of annual hospitalizations from 2003 to 2012 for children and adults.
80
< 1 year 1-4 years
60
5-17 years
40
18-39 years 40-64 years
20
65+years
0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Fiscal Year
Figure 2. Age-specific annual hospitalization rate in congenital heart disease patients in Canada (2003-2012). Dashed line indicates 95% confidence interval.
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
5
Table 3. Length of stay during index hospitalization Variable
Median length of stay (IQR), days
Age group < 1 year 9 (3-26) 1-4 years 3 (2-5) 5-17 years 3 (1-5) 18-39 years 4 (2-7) 40-64 years 6 (3-10) 65 years 8 (5-14) Sex Female 5 (3-15) Male 6 (3-16) Severity of CHD Simple 6 (3-14) Moderate 6 (3-16) Complex 9 (3-27) Primary diagnosis* according to greatest length of stay Hypoplastic left heart syndrome 22 (8-43) Infective endocarditis 16 (9-31) Transposition of great arteries 15 (7-26) Total anomalous pulmonary 14 (6-26) venous connection Single ventricle 13 (7-24) Down syndrome 11 (5-23) Pulmonary embolism 11 (6-20) Ventricular tachycardia 9 (3-17) Tricuspid atresia 9 (5-18) Cardiomyopathy 8 (5-17) Ischemic heart disease 8 (4-17) Atherosclerosis of coronary 8 (5-14) artery without angina Congestive heart failure 7 (4-15)
Table 4. Total length of stay between 2003 and 2012 inclusive P < 0.0001
< 0.0001 < 0.0001
CHD, congenital heart disease. * To present the maximum resource utilization, the principal diagnoses with frequency >100 were chosen.
1.004-1.040; P < 0.001). The annual hospitalization rate was consistently higher for men with 41-48 hospitalizations per 100,000 persons; for women, the rate was 36-39 per 100,000 persons (P < 0.001). CHD severity The number of hospitalizations increased by 2.8% per year among simple CHD patients, by 1.3% in moderate CHD patients, and by 3.2% for complex CHD patients. The annual hospitalization rate increased by 1.3% per year among simple CHD patients (95% CI, 1.011-1.016; P < 0.001) and by 2.5% among complex CHD patients (95% CI, 1.019-1.032; P < 0.001). However, there was no significant change among moderate CHD patients (rate ratio, 1.002; 95% CI, 0.9991.006; P ¼ 0.22). The annual hospitalization rate was higher for simple CHD patients ranging from 21 to 24 per 100,000 persons, followed by 13-14 per 100,000 persons for moderate CHD patients, and 4-5 per 100,000 persons for complex CHD patients. LOS The median LOS during the index hospitalization was 6 days (interquartile range, 3-16). Index hospitalizations with a diagnosis of hypoplastic left heart syndrome followed by infective endocarditis and transposition of great arteries had the longest LOS (Table 3). The total LOS per person over the 10-year study period was longest in men, those with complex
Characteristics of patients
Median total length of stay (IQR), days
Age group at time of index hospitalization < 1 year 11.4 (3.6-33.4) 1-4 years 4.3 (2.5-9.3) 5-17 years 4.3 (1.9-8.5) 18-39 years 5.3 (2.7-11.3) 40-64 years 8.3 (5.1-19.4) 65 years 16.2 (7.7-36.8) Sex Female 7.9 (3.2-24.7) Male 9.6 (4.3-26.0) Severity of CHD Simple 7.8 (3.3-22.0) Moderate 9.6 (4.1-26.9) Complex 21.8 (8.1-55.0)
P < 0.0001
< 0.0001 < 0.0001
CHD, congenital heart disease.
CHD, patients aged 65 years, and infants (Table 4). Risk factors for LOS > 14 days were age < 1 year, male sex, complex CHD, and having cardiac surgery and catheterization during the same hospitalization (Supplemental Table S4). Discussion To our knowledge, this is the first study to assess temporal trends of hospitalizations among CHD patients across Canada. The absolute number of annual hospitalizations increased an average of 2.4% per year from 2003 to 2012. Hospitalizations with CHD increased more than that observed among all-cause hospitalizations in Canada, which increased annually by 0.6%, 0.4%, and 0.2% over 3 consecutive years from 2003 and 2004 to 2005 and 2006.9 A population based study in the United Kingdom reported an average increase of 1.2% per year in the number of hospitalizations with a primary diagnosis of CHD between 1995 to 1996 and 2003 to 2004.10 This is consistent with the observed increase in the number of hospitalizations among our study population. However, the annual number of hospitalizations in adults with CHD in the United States more than doubled between 1998 and 2005, with an average increase of 14.6% per year.11 The larger increase in hospitalizations, relative to our findings, observed in the US study might be because of the difference in study methods. The US administrative data sources report cases rather than singular patients. In the absence of a unique identifier such as exists in Canada, it is not possible to ascertain multiple hospitalizations per patient, such that each case constitutes a hospitalization episode. Hospitalizations with admission and discharge on the same date were excluded from our study. In contrast, the US study included hospitalizations with day procedures, which might have been performed with increasing frequency during their study period (1998-2005). The annual hospitalization rate for persons with CHD in our study was 38-43 per 100,000 persons including all age groups and increased by 1.1% per year; this increment was modest compared with that observed in absolute numbers of hospitalizations (2.4% per year) from 2003 to 2012. We calculated the annual hospitalization rate indexed to the Canadian population, which experienced an average of 1.2% growth per year from 2003 to 2012.12 This is the likely
6
explanation for the smaller increase in the annual hospitalization rate relative to the increase in absolute hospitalization numbers during the period studied. The number of hospitalizations increased more in adults (4% per year) than in children (1.3% per year). This is consistent with the growing population of adult CHD survivors including the geriatric CHD population.3,13 The population of children with CHD is also increasing over time, although to a lesser extent than the population increase among adults.3 Although adults accounted for fewer hospitalizations than children overall, our 10-year study demonstrates that there is an important evolving change in the age distribution of hospitalized CHD patients, with a significant increase in the absolute number of hospitalizations among adults over time. Nonetheless, we observed a decline in the incidence of CHD hospitalizations indexed to the adult CHD population, which might relate to increasing number of referrals to specialized adult CHD centres and declining mortality among adults.4,14 We found a consistently higher number and rate of hospitalizations among men compared with women, despite the overall prevalence of CHD being higher in women than in men.2 This is in contrast to the findings of Zomer and colleagues, who reported that after excluding pregnancy-related admissions, the hospitalization rate was similar for women and men (42% vs 40%, respectively).15 Our data source does not allow us to conclude the reason that more hospitalizations were observed among men. Several factors might play a role; men might develop more complications and/or have more comorbid conditions compared with women.16 Female patients have milder lesions17 and experience lower mortality compared with men,18 consistent with results reported by Zomer and colleagues.15 Men with CHD are more likely to undergo surgery in infancy and adulthood and have a worse outcome after repeated surgery in adulthood compared with women.18,19 Other factors might include a possible lower compliance rate among men with medical therapy, or a higher proportion of men not under CHD care. We found more hospitalizations associated with simple CHD compared with moderate or complex CHD. Although the proportion of patients with complex CHD was less than those with simple or moderate CHD, there were disproportionately more admissions and more total hospital days among complex CHD patients, reflecting their predisposition to develop late complications and require reintervention.8 In adults, late cardiac events such as congestive heart failure, ischemic heart disease, atrial fibrillation, and atrial flutter were the most common reasons for hospitalizations. This finding is consistent with the report by Verheugt and colleagues, who indicated that supraventricular arrhythmia, arrhythmias “not otherwise specified,” chest pain, and heart failure were the most common cardiac admission diagnoses/ symptoms among adult CHD patients in The Netherlands.20 Zomer and colleagues also found cardiac arrhythmia, heart failure, and coronary artery disease to be the most common cardiac diagnoses among adult CHD patients of Quebec and The Netherlands.15 Atrial arrhythmia was the primary diagnosis for 8% of hospitalizations in the current study. Many patients with arrhythmia are stabilized and discharged within 24 hours, which might have decreased the hospitalization number with atrial arrhythmia. However, Bouchardy et al.
Canadian Journal of Cardiology Volume - 2015
found a wide range of incidence for development of atrial arrhythmia (ranging from 1% to 57%) among adult CHD patients depending on the CHD severity and patient age21; our result is consistent with this finding, acknowledging the differences in methodology. Our data demonstrate that ischemic heart disease is an important reason for hospitalization among adults, and is consistent with other studies that demonstrated that ischemic heart disease is an emerging problem in this population.22 The median LOS that we observed of 6 days is consistent with the median LOS of 5 days reported among adult CHD patients in Belgium.23 Patients with complex CHD had the highest LOS, consistent with the findings of a populationbased study of adults with CHD in Quebec.24 Limitations We used an administrative data source that was not constructed for research purposes, is prone to misclassification, and lacked data on socioeconomic status and hospital-specific information (ie, tertiary vs community hospital). However, we used a large nationally representative study population and examined CHD patients from all age groups, allowing a comparison of inpatient resource utilization across age groups and implementation of appropriate strategies in allocating ageappropriate resources. Hospitalizations from Quebec and the 3 territories were not included in the study. Of 34.8 million Canadians according to 2012 statistics,12 data on 76.4% (26.6 million) were captured in this study. We used the general population as a denominator in calculating the hospitalization rate, and an estimated rather than measured CHD population as an alternative denominator. Our study population was limited to patients who received inpatient care and does not represent the complete burden of CHD on health care systems because many CHD patients receive only outpatient care. Many CHD lesions are corrected or repaired during childhood and many patients survive to adulthood with a misconception of being “cured.” Therefore, when these patients are admitted for reasons other than CHD, there is the possibility that the CHD might not be coded as a diagnosis. Similarly, a diagnosis of simple CHD might be missed if a patient with simple CHD was hospitalized because of a reason not related to CHD, such as injury. However, complex CHD patients were more likely to be admitted with complications of CHD, which might lead to a greater likelihood of their CHD diagnosis being coded. This might result in differential misclassification of CHD hospitalizations; consequently our results might underestimate hospitalizations with simple CHD and hospitalizations among adults. Conclusions The absolute number of hospitalizations with CHD increased over time in children and adults, with a greater increase in adults. Although there were more hospitalizations associated with simple CHD, patients with complex lesions had disproportionately more admissions. However, the hospitalization rate indexed to the CHD population decreased in adults and remained constant in children during the study period. Increasing numbers of hospitalizations among adults pose an increasing financial burden associated with the
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
inpatient care of this population. Adult cardiology programs will need to provide increasing allocation of inpatient health care resources for patients living with CHD. Funding Sources Dr Mackie is funded by the Women and Children’s Health Research Institute, an academic institute affiliated with the University of Alberta and the Stollery Children’s Hospital Foundation. Disclosures The authors have no conflict of interest to disclose. References 1. Moons P, Bovijn L, Budts W, Belmans A, Gewillig M. Temporal trends in survival to adulthood among patients born with congenital heart disease from 1970 to 1992 in Belgium. Circulation 2010;122:2264-72. 2. Marelli AJ, Mackie AS, Ionescu-Ittu R, Rahme E, Pilote L. Congenital heart disease in the general population: changing prevalence and age distribution. Circulation 2007;115:163-72. 3. Marelli AJ, Ionescu-Ittu R, Mackie AS, et al. Lifetime prevalence of congenital heart disease in the general population from 2000 to 2010. Circulation 2014;130:749-56. 4. Avila P, Mercier LA, Dore A, et al. Adult congenital heart disease: a growing epidemic. Can J Cardiol 2014;30(12 suppl):S410-9. 5. Somerville J. Management of adults with congenital heart disease: an increasing problem. Annu Rev Med 1997;48:283-93. 6. Perloff JK. Adults with surgically treated congenital heart disease. Sequelae and residua. JAMA 1983;250:2033-6.
7
12. Statistics Canada. Table 051-0001-Estimates of population, by age group and sex for July 1, Canada, provinces and territories, annual (persons unless otherwise noted), CANSIM (database). Available at: http://www5.statcan. gc.ca/cansim/pick-choisir?lang¼eng&p2¼33&id¼0510001. Accessed March 18, 2015. 13. Afilalo J, Therrien J, Pilote L, et al. Geriatric congenital heart disease: burden of disease and predictors of mortality. J Am Coll Cardiol 2011;58:1509-15. 14. Mylotte D, Pilote L, Ionescu-Ittu R, et al. Specialized adult congenital heart disease care: the impact of policy on mortality. Circulation 2014;129:1804-12. 15. Zomer AC, Ionescu-Ittu R, Vaartjes I, et al. Sex differences in hospital mortality in adults with congenital heart disease: the impact of reproductive health. J Am Coll Cardiol 2013;62:58-67. 16. Verheugt CL, Uiterwaal CS, van der Velde ET, et al. Gender and outcome in adult congenital heart disease. Circulation 2008;118:26-32. 17. Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle WT, Correa A. Prevalence of congenital heart defects in metropolitan Atlanta, 19982005. J Pediatr 2008;153:807-13. 18. Marelli A, Gauvreau K, Landzberg M, Jenkins K. Sex differences in mortality in children undergoing congenital heart disease surgery: a United States population-based study. Circulation 2010;122(11 suppl): S234-40. 19. Zomer AC, Verheugt CL, Vaartjes I, et al. Surgery in adults with congenital heart disease. Circulation 2011;124:2195-201. 20. Verheugt CL, Uiterwaal CS, van der Velde ET, et al. The emerging burden of hospital admissions of adults with congenital heart disease. Heart 2010;96:872-8. 21. Bouchardy J, Therrien J, Pilote L, et al. Atrial arrhythmias in adults with congenital heart disease. Circulation 2009;120:1679-86.
7. Gatzoulis MA, Hechter S, Siu SC, Webb GD. Outpatient clinics for adults with congenital heart disease: increasing workload and evolving patterns of referral. Heart 1999;81:57-61.
22. Giannakoulas G, Dimopoulos K, Engel R, et al. Burden of coronary artery disease in adults with congenital heart disease and its relation to congenital and traditional heart risk factors. Am J Cardiol 2009;103: 1445-50.
8. Warnes CA, Liberthson R, Danielson GK, et al. Task force 1: the changing profile of congenital heart disease in adult life. J Am Coll Cardiol 2001;37:1170-5.
23. Moons P, Siebens K, De Geest S, et al. A pilot study of expenditures on, and utilization of resources in, health care in adults with congenital heart disease. Cardiol Young 2001;11:301-13.
9. Canadian Institute for Health Information. Trends in inpatient hospitalization and day surgery in Canada from 1995/1996 to 2005/2006. 2007. Available at: https://secure.cihi.ca/free_products/cad_analysis_in_ brief_e.pdf. Accessed August 7, 2015.
24. Mackie AS, Pilote L, Ionescu-Ittu R, Rahme E, Marelli AJ. Health care resource utilization in adults with congenital heart disease. Am J Cardiol 2007;99:839-43.
10. Billett J, Majeed A, Gatzoulis M, Cowie M. Trends in hospital admissions, in-hospital case fatality and population mortality from congenital heart disease in England, 1994 to 2004. Heart 2008;94:342-8. 11. Opotowsky AR, Siddiqi OK, Webb GD. Trends in hospitalizations for adults with congenital heart disease in the U.S. J Am Coll Cardiol 2009;54:460-7.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2015.05.022.
-
(2015) 1e7
Clinical Research
Congenital Heart Disease Hospitalizations in Canada: A 10-Year Experience Sunjidatul Islam, MBBS, MSc,a Yutaka Yasui, PhD,a Padma Kaul, PhD,a,b Ariane J. Marelli, MD, MPH,c and Andrew S. Mackie, MD, SMa,d,e a b c
School of Public Health, University of Alberta, Edmonton, Alberta, Canada
Department of Medicine, University of Alberta, Edmonton, Alberta, Canada
McGill Adult Unit for Congenital Heart Disease Excellence, McGill University, Montreal, Quebec, Canada d e
Stollery Children’s Hospital, Edmonton, Alberta, Canada
Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada
ABSTRACT
RESUM E
Background: The effect of the growing population of children and adults with congenital heart disease (CHD) on inpatient services in Canada is not known. We sought to assess temporal changes in hospitalizations of CHD patients. Methods: We identified all patients with a CHD diagnosis who received inpatient care in Canada between fiscal years 2003 and 2012 according to the Discharge Abstract Database of the Canadian Institute for Health Information. Poisson regression was performed to assess temporal changes in the annual hospitalization rate. Hospitalization rates were indexed to the general population and the estimated CHD population. Results: A total of 103,034 hospitalizations occurred in 61,051 patients from fiscal years 2003 to 2012. The absolute number of hospitalizations increased by 4.0% per year in adults and 1.3% per year in children. The greatest increase was in patients aged 65 years
Introduction : L’incidence de la population croissante d’enfants et nitale sur les services hospid’adultes atteints de cardiopathie conge à e valuer les taliers canadiens est inconnue. Nous avons cherche variations temporelles en matière d’hospitalisation chez les patients nitale. atteints de cardiopathie conge thodes : À l’aide de la base de donne es sur les conge s hospitaliers Me , nous avons recense de l’Institut canadien d’information sur la sante nitale ayant e te hostous les patients atteints de cardiopathie conge s au Canada entre les anne es financières 2003 et 2012. Une pitalise gression de Poisson a e te effectue e afin d’e valuer les variations re temporelles des taux d’hospitalisation annuels. Ces taux d’hospitali te indexe s en fonction de la population ge ne rale et de la sation ont e e de patients atteints de cardiopathie conge nitale. population estime sultats : Entre les anne es financières 2003 et 2012, il y a eu au Re total 103 034 hospitalisations pour 61 051 patients. Le nombre
With improvements in survival,1 the prevalence of congenital heart disease (CHD) has increased rapidly in adults, and adults with CHD now outnumber children with CHD.2,3 In addition, the distribution of CHD has changed with an increased prevalence of severe CHD among adults relative to children.2-4 Adult patients with CHD are at risk of development of late cardiac complications, which sometimes require repeat interventions.4-6 Moreover, they might acquire other comorbid conditions with aging. The effect of this changing
demographic characteristic on the number of adult CHD clinic visits has been previously described in Canada.7 However, the effect on inpatient care is not known. Therefore, the objectives of this study were: (1) to determine temporal changes in the hospitalization rate among CHD patients in Canada, overall and according to age, sex, and severity of CHD; (2) to determine temporal changes in length of hospital stay, overall and among specific patient subgroups; and (3) to identify predictors of length of stay (LOS) > 14 days. Methods
Received for publication February 24, 2015. Accepted May 29, 2015. Corresponding author: Dr Andrew S. Mackie, Division of Cardiology, Stollery Children’s Hospital, 4C2 Walter C. Mackenzie Center, 8440-112th St NW, Edmonton, Alberta T6G 2B7, Canada. Tel.: þ1-780-407-8361; fax: þ1-780-407-3954. E-mail: [email protected] See page 7 for disclosure information.
Study design and data source We conducted an observational retrospective cohort study using the Discharge Abstract Database from the Canadian Institute for Health Information, which consists of hospitalization records from all acute care hospitals in Canada, except
http://dx.doi.org/10.1016/j.cjca.2015.05.022 0828-282X/Ó 2015 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
2
Canadian Journal of Cardiology Volume - 2015
followed by those 40-64 years. However, the hospitalization rate in adults varied between 39 and 55 per 1000 CHD population with a reduction of 4% per year (95% confidence interval, 0.95-0.96; P < 0.001). The hospitalization rate in children ranged from 79 to 87 per 1000 CHD population and did not change significantly over time (rate ratio, 1.00; 95% confidence interval, 1.00-1.01; P ¼ 0.035). Men accounted for 53.5% of hospitalizations. Conclusions: The absolute number of hospitalizations of patients with CHD increased over time in children and adults. However, the hospitalization rate relative to the CHD population decreased among adults, possibly reflecting improved outpatient management. The absolute increase in CHD hospitalizations will pose a financial burden on health care systems.
e chez les adultes absolu d’hospitalisations s’est accru de 4 % par anne e chez les enfants. La plus forte augmentation a et de 1,3 % par anne te enregistre e chez les patients âge s de 65 ans et plus suivie des e s de 40 à 64 ans. On a toutefois observe un taux d’hospatients âge pitalisation variant entre 39 et 55 par 1000 patients adultes atteints nitale et une re duction de ce taux de 4 % par de cardiopathie conge e (intervalle de confiance à 95 % 0,95 à 0,96; P < 0,001). Chez anne les enfants, le taux d’hospitalisation se situait entre 79 et 87 par 1000 nitale et n’a pas varie de patients atteints de cardiopathie conge manière significative au fil du temps (ratio des taux, 1,00; intervalle de confiance à 95 % 1,00 à 1,01; P ¼ 0,035). Les hommes ont fait l’objet de 53,5 % des hospitalisations. Conclusions : Au fil du temps, le nombre absolu d’hospitalisations de nitale a augmente tant chez les patients atteints de cardiopathie conge une diminution du adultes que chez les enfants. On a toutefois observe taux d’hospitalisation chez les adultes atteints de cardiopathie connitale, ce qui pourrait s’expliquer par une meilleure prise en charge ge ambulatoire. Il est à noter que l’augmentation absolue du nombre d’hospitalisations chez les patients atteints de cardiopathie connitale imposera un fardeau financier au système de sante du pays. ge
those from the province of Quebec. The database includes data on admission and discharge dates, patient demographic variables, diagnoses (a primary and up to 24 secondary diagnoses), diagnostic and therapeutic procedures (up to 16 procedures), and discharge disposition. Multiple hospitalizations for the same patient can be tracked with the use of a unique anonymous patient identification number.
CHD was applied to the calculation of annual hospitalization rates according to age, sex, and severity of CHD. In addition, we indexed the annual hospitalization rate to the CHD population of Canada (less the Yukon, Northwest Territories, Nunavut, and Quebec), which was estimated using recent Quebec CHD prevalence data, published by Marelli et al.3 They reported the prevalence of CHD in Quebec for the years 2005 and 2010 for children and adults. We estimated the size of the CHD population on an annual basis using CHD prevalence data from 2005 and midyear general population statistics from Statistics Canada for fiscal years 2003-2007, and prevalence data from 2010 for fiscal years 2008-2012. The CHD population for children and adults were estimated separately using the same method. LOS was the secondary outcome of interest. In the event of a hospitalization within 24 hours of discharge, the LOS was measured as the duration between the first admission date and the second discharge date. We also assessed the total LOS, defined as the total duration of hospital days (over multiple hospitalizations) per person over the 10-year study period.
Study population Inclusion criteria. Our study population consisted of all patients, irrespective of their age, with a diagnosis of CHD who received inpatient care during the fiscal years 2003-2012 in Canada, less the Yukon, Northwest Territories, Nunavut, and Quebec. CHD was identified based on International Classification of Disease (ICD) Revisions 9 and 10, from any of 25 diagnostic fields. Exclusion criteria. We excluded patients with isolated patent ductus arteriosus and those who lived in the Yukon, Northwest Territories, Nunavut, or Quebec even if they received care in hospitals of other provinces. We also excluded patients who were discharged alive on the same day of admission, with the assumption that they were admitted for a day procedure. Hospitalizations within 24 hours of discharge were considered hospital-to-hospital transfers and therefore counted as single hospitalizations. Outcome measures Our primary outcome of interest was the annual number and rate of hospitalizations with CHD as a primary or secondary diagnosis and the annual number and rate of hospitalizations according to age group, sex, severity of CHD, and therapeutic intervention. For each year, we calculated the hospitalization rate by dividing the number of hospitalizations by the population size of Canada (less the Yukon, Northwest Territories, Nunavut, and Quebec) in the given year. The same procedure stratified according to age, sex, and severity of
Independent variables CHD lesions were grouped into 3 complexities: simple, moderate, and complex, defined according to the 32nd Bethesda conference.8 We classified patients with multiple CHD diagnoses based on the hierarchies of complexities. For example, if a patient had complex lesions and moderate or simple lesions, the patient was classified as a complex CHD patient. Similarly, patients with moderate and simple CHD diagnoses were categorized as moderate CHD patients. For patients having multiple hospitalizations during the study period, all CHD diagnoses coded across the multiple hospitalizations were used in classifying the patient as simple, moderate, or complex. Age was categorized into 6 categories: infants (< 1 year), 1-4 years, 5-17 years, 18-39 years, 40-64 years, and 65 years. We used Canadian Classification of Health Interventions codes corresponding to ICD-10 codes and Canadian
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
Classification of Diagnostic, Therapeutic, and Surgical Procedures codes corresponding to ICD-9 codes to identify the therapeutic cardiovascular intervention performed during hospitalizations. The hospitalizations were grouped into 4 categories based on the intervention procedure performed: no cardiac intervention, cardiac surgery only, cardiac catheterization only, or cardiac surgery and catheterization. Statistical analysis Characteristics of index hospitalizations (ie, characteristics of patients at the first hospitalization during the study period) and all hospitalizations were described according to frequency distributions (proportions) for categorical variables and median and interquartile range (IQR) for LOS, which was treated as a continuous variable. The 10 most common primary diagnoses during index and all hospitalizations were described with frequency distributions for children and adults separately. We expressed the annual hospitalization rate per 100,000 persons and per 1,000 of the CHD population. Poisson regression analysis was performed to assess temporal changes in the annual hospitalization rate. The numbers and rates of annual hospitalizations were also analyzed stratified according to age group, sex, and severity of CHD. The hospitalization rates indexed to the CHD population were stratified according to child or adult hospitalization. LOS during the index hospitalization and total LOS over the 10year study period were compared using Kruskal-Wallis (if more than 2 groups; eg, age) and Wilcoxon rank sum tests (if 2 groups; eg, sex) across the categories of age, sex, and severity of CHD. A multivariable logistic generalized estimating equations model was used to identify factors associated with LOS > 14 days. Two sided P values < 0.05 were considered statistically significant. All analyses were performed using SAS statistical software (version 9.4; SAS Institute, Cary, NC). Ethics All procedures of the study were approved by the Health Research Ethics Board at the University of Alberta. Results A total of 103,034 inpatient admissions with CHD as a primary or secondary diagnosis occurred between fiscal years 2003 and 2012 among 61,051 patients. Patient characteristics of the cohort are described in Table 1. In-hospital death occurred in 3730 (6.1%) patients and 1965 (52.7%) of all inhospital deaths occurred at the time of the index hospitalization. The most common primary diagnoses responsible for hospitalizations are reported in Table 2 for children and adults. Infants accounted for 39.2% of all hospitalizations followed by 40- to 64-year-olds (18.3%), and patients aged 65 years (16.3%). There were more hospitalizations among men (53.5%). Of all hospitalizations, 55.8% were associated with simple CHD, 32.3% were associated with moderate CHD, and 12.0% were associated with complex CHD. Among children, 28,837 (77.3%) had a single hospitalization, 4477 (12.0%) had 2 hospitalizations, and 3988 (10.7%) had 3 or more hospitalizations. Among adults, 14,028 (59.1%) had a single hospitalization, 4838 (20.4%) had 2 hospitalizations,
3
Table 1. Demographic and clinical characteristics of study subjects Characteristic Fiscal year 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Age group < 1 year 1-4 years 5-17 years 18-39 years 40-64 years 65 years Sex Female Male Median length of stay (IQR), days Severity of CHD Simple Moderate Complex Cardiac intervention No intervention Cardiac surgery Cardiac catheterization Cardiac surgery and catheterization
Index hospitalizations
All hospitalizations
7267 6460 6335 6155 5820 5968 5769 5783 5895 5599
(11.9) (10.6) (10.4) (10.1) (9.5) (9.8) (9.5) (9.5) (9.7) (9.2)
9217 9530 9964 9984 9832 10,449 10,628 10,909 11,338 11,183
(9.0) (9.3) (9.7) (9.7) (9.5) (10.1) (10.3) (10.6) (11.0) (10.9)
30,791 2629 4138 6262 10,614 6617
(50.4) (4.3) (6.8) (10.3) (17.4) (10.8)
40,420 8864 7818 10,283 18,882 16,767
(39.2) (8.6) (7.6) (10.0) (18.3) (16.3)
28,499 (46.7) 32,552 (53.3) 6.2 (2.9-15.7)
47,887 (46.5) 55,147 (53.5) 5.8 (2.6-13.2)
38,316 (62.8) 18,661 (30.6) 4074 (6.7)
57,455 (55.8) 33,271 (32.3) 12,308 (12.0)
43,072 12,993 3978 1008
72,325 21,721 6950 2038
(70.6) (21.3) (6.5) (1.7)
(70.2) (21.1) (6.8) (2.0)
Data are presented as n (%) except where otherwise stated. CHD, congenital heart disease; IQR, interquartile range.
and 4883 (20.6%) had 3 hospitalizations. The absolute numbers of annual hospitalizations from fiscal years 20032012 are summarized according to age group, sex, severity of CHD, and intervention in Supplemental Table S1. The most frequent CHD diagnoses and the most frequent combinations of 2 CHD diagnoses are presented in Supplemental Tables S2 and S3. The absolute number of hospitalizations increased from 9217 in 2003 to 11,183 in 2012, an average increase of 2.4% per year. The annual hospitalization rate varied between 38 and 43 per 100,000 persons during the study period with an increase of 1.1% per year (95% confidence interval [CI], 1.00-1.013; P < 0.001). The overall annual hospitalization rate varied between 55 and 70 per 1000 CHD population with a reduction of 3% per year from 2003 to 2012 (95% CI, 0.97-0.98; P < 0.001). Age groups The number of hospitalizations increased by 4.0% per year in adults (Fig. 1). The number of hospitalizations increased by 6.5% per year for patients aged 65 years, 3.0% per year for 40- to 64-year-olds, and 2.1% per year for 18- to 39-yearolds. In children, the number of hospitalizations increased by 1.3% per year (Fig. 1); an increase of 2.7% was observed in 1to 4-year-olds, and an increase of 2.1% in 5- to 17-year-olds. In contrast, the number of hospitalizations did not change
4
Canadian Journal of Cardiology Volume - 2015
Table 2. Most frequent primary diagnosis Index hospitalizations
All Hospitalizations
Children (< 18 years old) N Cardiac Ventricular septal defect Ostium secundum atrial septal defect Coarctation of the aorta Other specified anomalies of heart Other anomalies of heart Tetralogy of Fallot Transposition of great arteries Endocardial cushion defect Noncardiac Down syndrome Acute bronchiolitis Adults N Cardiac diagnoses Nonrheumatic aortic valve disorder Ostium secundum atrial septal defect Ischemic heart disease Aortic valve abnormality Atrial tachyarrhythmia* Congestive heart failure Atherosclerosis of coronary artery without angina Nonrheumatic mitral valve disorder Infective endocarditis Noncardiac Pneumonia
37,558 2570 1867 935 845 793 728 661 547
57,102
(6.8) (5.0) (2.5) (2.2) (2.1) (1.9) (1.8) (1.5)
683 (1.8) 596 (1.6)
Ventricular septal defect Ostium secundum atrial septal defect Tetralogy of Fallot Coarctation of the aorta Other specified anomalies of heart Transposition of the great arteries Endocardial cushion defect Pulmonary artery stenosis
3331 2186 1419 1213 1116 1090 1058 1005
Pneumonia Acute bronchiolitis
1544 (2.7) 1380 (2.4)
23,493 3142 2036 1527 1212 1207 956 870 561 432
(5.8) (3.8) (2.5) (2.1) (2.0) (1.9) (1.9) (1.8)
45,932
(13.4) (8.7) (6.5) (5.2) (5.1) (4.1) (3.7) (2.4) (1.8)
Congestive heart failure Nonrheumatic aortic valve disorder Atrial tachyarrhythmia* Ischemic heart disease Ostium secundum atrial septal defect Aortic valve abnormality Atherosclerosis of coronary artery without angina Nonrheumatic mitral valve disorder Infective endocarditis
262 (1.1)
Pneumonia
4065 4035 3588 2756 2564 1474 1465 824 727
(8.9) (8.8) (7.8) (6.0) (5.6) (3.2) (3.2) (1.8) (1.6)
734 (1.6)
Data are presented as n (%) except where otherwise stated. * Includes atrial fibrillation, atrial flutter, supraventricular tachycardia, atrial premature beats, sinoatrial node dysfunction, or pre-excitation syndrome.
significantly in infants (< 1 year of age). However, infants consistently had the highest annual hospitalization rate over the 10-year period followed by 1- to 4-year-olds and patients aged 65 years (Fig. 2). With respect to changes in hospitalization rates, the hospitalization rate in infants decreased by a factor of 0.992 per year (95% CI, 0.989-0.996; P < 0.001). However, the annual hospitalization rate increased per fiscal year by 1.7% in patients aged 1-4 years (95% CI, 1.010-1.025; P < 0.001), 3.1% in patients aged 5-17 years (95% CI, 1.002-1.039; P < 0.001), 1.7% in patients aged 18-39 years (95% CI, 1.0101.024; P < 0.001), 1.0% in patients aged 40-64 years (95% CI, 1.005-1.015; P < 0.001), and 2.5% in patients aged 65 years (95% CI, 1.020-1.031; P < 0.001). The
hospitalization rate in children ranged from 79 to 87 per 1000 CHD population and did not change significantly over time (rate ratio, 1.00; 95% CI, 1.00-1.01; P ¼ 0.035). In contrast, the hospitalization rate in adults varied between 39 and 55 per 1000 CHD population with a reduction of 4% per year from 2003 to 2012 (95% CI, 0.95-0.96; P < 0.001). Sex The number of hospitalizations increased by 2.7% per year for men and by 2.0% per year for women. The annual hospitalization rate increased by 1.5% per year in men (95% CI, 1.012-1.018; P < 0.001) and by 0.7% in women (95% CI, 1600
Annual hospitalization rate per 100,000 persons
1500 7000
1400
Number of hospitalizations
6000
1300 100
5000 4000 Children
3000
Adults 2000 1000 0 2003
2004
2005
2006
2007 2008 Fiscal year
2009
2010
2011
2012
Figure 1. Changes in number of annual hospitalizations from 2003 to 2012 for children and adults.
80
< 1 year 1-4 years
60
5-17 years
40
18-39 years 40-64 years
20
65+years
0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Fiscal Year
Figure 2. Age-specific annual hospitalization rate in congenital heart disease patients in Canada (2003-2012). Dashed line indicates 95% confidence interval.
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
5
Table 3. Length of stay during index hospitalization Variable
Median length of stay (IQR), days
Age group < 1 year 9 (3-26) 1-4 years 3 (2-5) 5-17 years 3 (1-5) 18-39 years 4 (2-7) 40-64 years 6 (3-10) 65 years 8 (5-14) Sex Female 5 (3-15) Male 6 (3-16) Severity of CHD Simple 6 (3-14) Moderate 6 (3-16) Complex 9 (3-27) Primary diagnosis* according to greatest length of stay Hypoplastic left heart syndrome 22 (8-43) Infective endocarditis 16 (9-31) Transposition of great arteries 15 (7-26) Total anomalous pulmonary 14 (6-26) venous connection Single ventricle 13 (7-24) Down syndrome 11 (5-23) Pulmonary embolism 11 (6-20) Ventricular tachycardia 9 (3-17) Tricuspid atresia 9 (5-18) Cardiomyopathy 8 (5-17) Ischemic heart disease 8 (4-17) Atherosclerosis of coronary 8 (5-14) artery without angina Congestive heart failure 7 (4-15)
Table 4. Total length of stay between 2003 and 2012 inclusive P < 0.0001
< 0.0001 < 0.0001
CHD, congenital heart disease. * To present the maximum resource utilization, the principal diagnoses with frequency >100 were chosen.
1.004-1.040; P < 0.001). The annual hospitalization rate was consistently higher for men with 41-48 hospitalizations per 100,000 persons; for women, the rate was 36-39 per 100,000 persons (P < 0.001). CHD severity The number of hospitalizations increased by 2.8% per year among simple CHD patients, by 1.3% in moderate CHD patients, and by 3.2% for complex CHD patients. The annual hospitalization rate increased by 1.3% per year among simple CHD patients (95% CI, 1.011-1.016; P < 0.001) and by 2.5% among complex CHD patients (95% CI, 1.019-1.032; P < 0.001). However, there was no significant change among moderate CHD patients (rate ratio, 1.002; 95% CI, 0.9991.006; P ¼ 0.22). The annual hospitalization rate was higher for simple CHD patients ranging from 21 to 24 per 100,000 persons, followed by 13-14 per 100,000 persons for moderate CHD patients, and 4-5 per 100,000 persons for complex CHD patients. LOS The median LOS during the index hospitalization was 6 days (interquartile range, 3-16). Index hospitalizations with a diagnosis of hypoplastic left heart syndrome followed by infective endocarditis and transposition of great arteries had the longest LOS (Table 3). The total LOS per person over the 10-year study period was longest in men, those with complex
Characteristics of patients
Median total length of stay (IQR), days
Age group at time of index hospitalization < 1 year 11.4 (3.6-33.4) 1-4 years 4.3 (2.5-9.3) 5-17 years 4.3 (1.9-8.5) 18-39 years 5.3 (2.7-11.3) 40-64 years 8.3 (5.1-19.4) 65 years 16.2 (7.7-36.8) Sex Female 7.9 (3.2-24.7) Male 9.6 (4.3-26.0) Severity of CHD Simple 7.8 (3.3-22.0) Moderate 9.6 (4.1-26.9) Complex 21.8 (8.1-55.0)
P < 0.0001
< 0.0001 < 0.0001
CHD, congenital heart disease.
CHD, patients aged 65 years, and infants (Table 4). Risk factors for LOS > 14 days were age < 1 year, male sex, complex CHD, and having cardiac surgery and catheterization during the same hospitalization (Supplemental Table S4). Discussion To our knowledge, this is the first study to assess temporal trends of hospitalizations among CHD patients across Canada. The absolute number of annual hospitalizations increased an average of 2.4% per year from 2003 to 2012. Hospitalizations with CHD increased more than that observed among all-cause hospitalizations in Canada, which increased annually by 0.6%, 0.4%, and 0.2% over 3 consecutive years from 2003 and 2004 to 2005 and 2006.9 A population based study in the United Kingdom reported an average increase of 1.2% per year in the number of hospitalizations with a primary diagnosis of CHD between 1995 to 1996 and 2003 to 2004.10 This is consistent with the observed increase in the number of hospitalizations among our study population. However, the annual number of hospitalizations in adults with CHD in the United States more than doubled between 1998 and 2005, with an average increase of 14.6% per year.11 The larger increase in hospitalizations, relative to our findings, observed in the US study might be because of the difference in study methods. The US administrative data sources report cases rather than singular patients. In the absence of a unique identifier such as exists in Canada, it is not possible to ascertain multiple hospitalizations per patient, such that each case constitutes a hospitalization episode. Hospitalizations with admission and discharge on the same date were excluded from our study. In contrast, the US study included hospitalizations with day procedures, which might have been performed with increasing frequency during their study period (1998-2005). The annual hospitalization rate for persons with CHD in our study was 38-43 per 100,000 persons including all age groups and increased by 1.1% per year; this increment was modest compared with that observed in absolute numbers of hospitalizations (2.4% per year) from 2003 to 2012. We calculated the annual hospitalization rate indexed to the Canadian population, which experienced an average of 1.2% growth per year from 2003 to 2012.12 This is the likely
6
explanation for the smaller increase in the annual hospitalization rate relative to the increase in absolute hospitalization numbers during the period studied. The number of hospitalizations increased more in adults (4% per year) than in children (1.3% per year). This is consistent with the growing population of adult CHD survivors including the geriatric CHD population.3,13 The population of children with CHD is also increasing over time, although to a lesser extent than the population increase among adults.3 Although adults accounted for fewer hospitalizations than children overall, our 10-year study demonstrates that there is an important evolving change in the age distribution of hospitalized CHD patients, with a significant increase in the absolute number of hospitalizations among adults over time. Nonetheless, we observed a decline in the incidence of CHD hospitalizations indexed to the adult CHD population, which might relate to increasing number of referrals to specialized adult CHD centres and declining mortality among adults.4,14 We found a consistently higher number and rate of hospitalizations among men compared with women, despite the overall prevalence of CHD being higher in women than in men.2 This is in contrast to the findings of Zomer and colleagues, who reported that after excluding pregnancy-related admissions, the hospitalization rate was similar for women and men (42% vs 40%, respectively).15 Our data source does not allow us to conclude the reason that more hospitalizations were observed among men. Several factors might play a role; men might develop more complications and/or have more comorbid conditions compared with women.16 Female patients have milder lesions17 and experience lower mortality compared with men,18 consistent with results reported by Zomer and colleagues.15 Men with CHD are more likely to undergo surgery in infancy and adulthood and have a worse outcome after repeated surgery in adulthood compared with women.18,19 Other factors might include a possible lower compliance rate among men with medical therapy, or a higher proportion of men not under CHD care. We found more hospitalizations associated with simple CHD compared with moderate or complex CHD. Although the proportion of patients with complex CHD was less than those with simple or moderate CHD, there were disproportionately more admissions and more total hospital days among complex CHD patients, reflecting their predisposition to develop late complications and require reintervention.8 In adults, late cardiac events such as congestive heart failure, ischemic heart disease, atrial fibrillation, and atrial flutter were the most common reasons for hospitalizations. This finding is consistent with the report by Verheugt and colleagues, who indicated that supraventricular arrhythmia, arrhythmias “not otherwise specified,” chest pain, and heart failure were the most common cardiac admission diagnoses/ symptoms among adult CHD patients in The Netherlands.20 Zomer and colleagues also found cardiac arrhythmia, heart failure, and coronary artery disease to be the most common cardiac diagnoses among adult CHD patients of Quebec and The Netherlands.15 Atrial arrhythmia was the primary diagnosis for 8% of hospitalizations in the current study. Many patients with arrhythmia are stabilized and discharged within 24 hours, which might have decreased the hospitalization number with atrial arrhythmia. However, Bouchardy et al.
Canadian Journal of Cardiology Volume - 2015
found a wide range of incidence for development of atrial arrhythmia (ranging from 1% to 57%) among adult CHD patients depending on the CHD severity and patient age21; our result is consistent with this finding, acknowledging the differences in methodology. Our data demonstrate that ischemic heart disease is an important reason for hospitalization among adults, and is consistent with other studies that demonstrated that ischemic heart disease is an emerging problem in this population.22 The median LOS that we observed of 6 days is consistent with the median LOS of 5 days reported among adult CHD patients in Belgium.23 Patients with complex CHD had the highest LOS, consistent with the findings of a populationbased study of adults with CHD in Quebec.24 Limitations We used an administrative data source that was not constructed for research purposes, is prone to misclassification, and lacked data on socioeconomic status and hospital-specific information (ie, tertiary vs community hospital). However, we used a large nationally representative study population and examined CHD patients from all age groups, allowing a comparison of inpatient resource utilization across age groups and implementation of appropriate strategies in allocating ageappropriate resources. Hospitalizations from Quebec and the 3 territories were not included in the study. Of 34.8 million Canadians according to 2012 statistics,12 data on 76.4% (26.6 million) were captured in this study. We used the general population as a denominator in calculating the hospitalization rate, and an estimated rather than measured CHD population as an alternative denominator. Our study population was limited to patients who received inpatient care and does not represent the complete burden of CHD on health care systems because many CHD patients receive only outpatient care. Many CHD lesions are corrected or repaired during childhood and many patients survive to adulthood with a misconception of being “cured.” Therefore, when these patients are admitted for reasons other than CHD, there is the possibility that the CHD might not be coded as a diagnosis. Similarly, a diagnosis of simple CHD might be missed if a patient with simple CHD was hospitalized because of a reason not related to CHD, such as injury. However, complex CHD patients were more likely to be admitted with complications of CHD, which might lead to a greater likelihood of their CHD diagnosis being coded. This might result in differential misclassification of CHD hospitalizations; consequently our results might underestimate hospitalizations with simple CHD and hospitalizations among adults. Conclusions The absolute number of hospitalizations with CHD increased over time in children and adults, with a greater increase in adults. Although there were more hospitalizations associated with simple CHD, patients with complex lesions had disproportionately more admissions. However, the hospitalization rate indexed to the CHD population decreased in adults and remained constant in children during the study period. Increasing numbers of hospitalizations among adults pose an increasing financial burden associated with the
Islam et al. CHD Hospitalizations in Canada From 2003 to 2012
inpatient care of this population. Adult cardiology programs will need to provide increasing allocation of inpatient health care resources for patients living with CHD. Funding Sources Dr Mackie is funded by the Women and Children’s Health Research Institute, an academic institute affiliated with the University of Alberta and the Stollery Children’s Hospital Foundation. Disclosures The authors have no conflict of interest to disclose. References 1. Moons P, Bovijn L, Budts W, Belmans A, Gewillig M. Temporal trends in survival to adulthood among patients born with congenital heart disease from 1970 to 1992 in Belgium. Circulation 2010;122:2264-72. 2. Marelli AJ, Mackie AS, Ionescu-Ittu R, Rahme E, Pilote L. Congenital heart disease in the general population: changing prevalence and age distribution. Circulation 2007;115:163-72. 3. Marelli AJ, Ionescu-Ittu R, Mackie AS, et al. Lifetime prevalence of congenital heart disease in the general population from 2000 to 2010. Circulation 2014;130:749-56. 4. Avila P, Mercier LA, Dore A, et al. Adult congenital heart disease: a growing epidemic. Can J Cardiol 2014;30(12 suppl):S410-9. 5. Somerville J. Management of adults with congenital heart disease: an increasing problem. Annu Rev Med 1997;48:283-93. 6. Perloff JK. Adults with surgically treated congenital heart disease. Sequelae and residua. JAMA 1983;250:2033-6.
7
12. Statistics Canada. Table 051-0001-Estimates of population, by age group and sex for July 1, Canada, provinces and territories, annual (persons unless otherwise noted), CANSIM (database). Available at: http://www5.statcan. gc.ca/cansim/pick-choisir?lang¼eng&p2¼33&id¼0510001. Accessed March 18, 2015. 13. Afilalo J, Therrien J, Pilote L, et al. Geriatric congenital heart disease: burden of disease and predictors of mortality. J Am Coll Cardiol 2011;58:1509-15. 14. Mylotte D, Pilote L, Ionescu-Ittu R, et al. Specialized adult congenital heart disease care: the impact of policy on mortality. Circulation 2014;129:1804-12. 15. Zomer AC, Ionescu-Ittu R, Vaartjes I, et al. Sex differences in hospital mortality in adults with congenital heart disease: the impact of reproductive health. J Am Coll Cardiol 2013;62:58-67. 16. Verheugt CL, Uiterwaal CS, van der Velde ET, et al. Gender and outcome in adult congenital heart disease. Circulation 2008;118:26-32. 17. Reller MD, Strickland MJ, Riehle-Colarusso T, Mahle WT, Correa A. Prevalence of congenital heart defects in metropolitan Atlanta, 19982005. J Pediatr 2008;153:807-13. 18. Marelli A, Gauvreau K, Landzberg M, Jenkins K. Sex differences in mortality in children undergoing congenital heart disease surgery: a United States population-based study. Circulation 2010;122(11 suppl): S234-40. 19. Zomer AC, Verheugt CL, Vaartjes I, et al. Surgery in adults with congenital heart disease. Circulation 2011;124:2195-201. 20. Verheugt CL, Uiterwaal CS, van der Velde ET, et al. The emerging burden of hospital admissions of adults with congenital heart disease. Heart 2010;96:872-8. 21. Bouchardy J, Therrien J, Pilote L, et al. Atrial arrhythmias in adults with congenital heart disease. Circulation 2009;120:1679-86.
7. Gatzoulis MA, Hechter S, Siu SC, Webb GD. Outpatient clinics for adults with congenital heart disease: increasing workload and evolving patterns of referral. Heart 1999;81:57-61.
22. Giannakoulas G, Dimopoulos K, Engel R, et al. Burden of coronary artery disease in adults with congenital heart disease and its relation to congenital and traditional heart risk factors. Am J Cardiol 2009;103: 1445-50.
8. Warnes CA, Liberthson R, Danielson GK, et al. Task force 1: the changing profile of congenital heart disease in adult life. J Am Coll Cardiol 2001;37:1170-5.
23. Moons P, Siebens K, De Geest S, et al. A pilot study of expenditures on, and utilization of resources in, health care in adults with congenital heart disease. Cardiol Young 2001;11:301-13.
9. Canadian Institute for Health Information. Trends in inpatient hospitalization and day surgery in Canada from 1995/1996 to 2005/2006. 2007. Available at: https://secure.cihi.ca/free_products/cad_analysis_in_ brief_e.pdf. Accessed August 7, 2015.
24. Mackie AS, Pilote L, Ionescu-Ittu R, Rahme E, Marelli AJ. Health care resource utilization in adults with congenital heart disease. Am J Cardiol 2007;99:839-43.
10. Billett J, Majeed A, Gatzoulis M, Cowie M. Trends in hospital admissions, in-hospital case fatality and population mortality from congenital heart disease in England, 1994 to 2004. Heart 2008;94:342-8. 11. Opotowsky AR, Siddiqi OK, Webb GD. Trends in hospitalizations for adults with congenital heart disease in the U.S. J Am Coll Cardiol 2009;54:460-7.
Supplementary Material To access the supplementary material accompanying this article, visit the online version of the Canadian Journal of Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10. 1016/j.cjca.2015.05.022.