Pediatrics and Neonatology (2013) 54, 113e118

Available online at www.sciencedirect.com

journal homepage: http://www.pediatr-neonatol.com

ORIGINAL ARTICLE

Prevalence, Mortality, and the Disease

Burden of Pediatric Congenital Heart Disease
in Taiwan
Shu-Jen Yeh a,b, Hui-Chi Chen c, Chun-Wei Lu d, Jou-Kou Wang d,
Li-Min Huang b,d, Shin-Chung Huang e, San-Kuei Huang e, Mei-Hwan Wu d,*

a
Department of Pediatrics, Far Eastern Memorial Hospital, Taipei, Taiwan
b
Graduate Institute of Epidemiology and Preventive Medicine, College of Public Health, National Taiwan University,
Taipei, Taiwan
c
Genomics Research Center, Academia Sinica, National Taiwan University Hospital and Medical College, National Taiwan
University, Taipei, Taiwan
d
Department of Pediatrics, National Taiwan University Hospital and Medical College, National Taiwan University,
Taipei, Taiwan
e
Taiwan Bureau of National Health Insurance, Taipei, Taiwan

Received Feb 15, 2012; received in revised form Jun 15, 2012; accepted Jun 26, 2012

Key Words Background: In Taiwan, the incidence of congenital heart diseases (CHDs) and severe CHDs was
congenital heart 13.08 and 1.51 per 1000 live births, respectively. This study further elucidates the prevalence
disease; and mortality of pediatric CHD patients in Taiwan.
disease burden; Methods: From the National Health Insurance database 2000e2010, we retrieved the data of
pediatric; CHD patients (aged 0e18 years). Mortality data were obtained from the national death statis-
prevalence tics.
Results: In total, 45,119 pediatric CHD patients were identified, given the prevalence at 918.0
per 100,000 (107.1 for severe CHD and 853.8 for simple CHD). Ventricular septal defect, ostium
secundum-type atrial septal defect, patent ductus arteriosus, pulmonary stenosis, and
tetralogy of Fallot were the five most frequently diagnosed CHDs. In those aged 0e6 years,
the prevalence was 1233.7 per 100,000 (123.5 for severe CHD and 1149.6 for simple CHD).
The age-specific prevalence of both simple and severe CHDs declined rapidly after the age
of 10 years. From the death registry, we noted that more than 90% of CHD-related deaths
occurred before the age of 5 years. The probability of cardiac death in CHD patients during

* Corresponding author. Department of Pediatrics, National Taiwan University Hospital, Number 7, Chung-Shen South Road, Taipei 100,
Taiwan.
E-mail address: [email protected] (M.-H. Wu).

1875-9572/$36 Copyright ª 2012, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights reserved.
http://dx.doi.org/10.1016/j.pedneo.2012.11.010
114 S.-J. Yeh et al

infancy was 4.5%, with the cumulative probability reaching 5.44%, 5.68%, and 6.04% by the
ages of 5, 10, and 20 years, respectively.
Conclusion: Because most CHD deaths occurred within the first 5 years of life (mainly during
infancy), the relatively low prevalence of CHDs in the population aged 0e18 years (918/
100,000; 74% for those between 0 years and 6 years of age) and the rapid decline in the
age-specific prevalence of CHD after the age of 10 years was attributed to noncompliance
of the children to medical follow-up after they began schooling.
Copyright ª 2012, Taiwan Pediatric Association. Published by Elsevier Taiwan LLC. All rights
reserved.

1. Introduction limited our record retrieval to the birth cohort between

1992 and 2010. The completely computerized NHI database
Congenital heart disease (CHD) is a major concern in consists of health records from each medical visit,
pediatric health care and has a significant impact on infant a scrambled identification number, and information
morbidity and mortality worldwide.1e3 Recent advances in including the patient’s date of birth, gender, date of
fetal diagnosis and postnatal cardiac intervention may admission or outpatient department visits, and reimburse-
modify the CHD spectrum and subsequent medical ment. Patients diagnosed with any type of CHD (according
needs.4e7 The incidence of CHD varies between geographic to the International Classification of Diseases, Ninth Revi-
regions, and this variation has been attributed to the sion, Clinical Modification [ICD-9-CM] code) were selected.
number of patients with ventricular septal defects (VSD).8 To avoid errors caused by incorrect tentative diagnoses,
The reported incidence of CHD among infants ranges from insignificant asymptomatic CHD, and miscoding, we defined
four to 12 per 1000 live births and is estimated to be prevalent CHD patients as those hospitalized with CHD as
approximately 10 per 1000 live births.5,9 A national pop- the major disease or those who visited the outpatient
ulation study conducted from 2000 to 2006 in Taiwan department with a major diagnosis of CHD consistently at
showed that the incidence of CHD was 13.08 per 1000 live least three times within the study period. The CHD was
births, with incidences of severe CHDs and simple CHDs at further classified as “simple” or “severe” based on the
1.51 and 11.67 per 1000 live births, respectively.10 The diagnostic code (Table 1).10
reported prevalence of CHD varies because of differences
in the length of follow-up periods, as well as methods of
target population selection and case detection.2,5,11 2.2. Age-specific prevalence
Several multistage surveys have been conducted in
school-aged children in Taiwan12,13; the prevalence of CHD The gender-specific population sizes by a single year of age
ranged from 2.4 to 6.04 per 1000 children.12,13 The were obtained from the Statistical Yearbook of Interior,
discrepancy between the incidence and prevalence data Department of Statistics, Ministry of the Interior.15 In 2010,
may be attributed to disease-related deaths and case a total of 4,915,037 children between 0 years and 18 years
identification criteria, as set by the study methods. A made up the denominator used for calculating prevalence
population-based study using a registry database may avoid and age-specific prevalence.
referral and diagnostic bias caused by geographic regional
variations and small sample sizes. The National Health
Insurance (NHI) is managed by the Bureau of National 2.3. Death statistics
Health Insurance, Department of Health. Since 1995, more
than 98% of Taiwanese nationals have been obligated by law The number of deaths from CHD (ICD 10, Q20eQ28) and
to join the government-run, single-payer NHI program.14 all causes were available from the Death Statistics for
The Taiwan National Department of Health established 2008 and 2009 compiled by the Department of Health.16
a national mortality database, which includes data on age- For calculating age-specific CHD mortality, the pop-
specific mortality and causes of death. These national ulation size obtained in 2008 and 2009 were used as the
databases serve as ideal resources for nationwide epide- denominator. We constructed a hypothetic birth cohort of
miological studies. Using these databases, we investigated 100,000 newborns with CHD incidence at 1308 per
the overall and age-specific prevalence of CHD and CHD 100,00010 and calculated the age-specific mortality for
subtypes. The probability of death from CHD was estimated CHD and for all other causes from the death statistics
in order to elucidate its impact on disease prevalence. data 2008e2009. We estimated the number of CHD
deaths, the number of deaths from other causes, and the
number of patients alive after considering the competing
2. Methods risks in each age strata. We divided the number of CHD
deaths by the number of patients at risk of each age
2.1. NHI database strata to obtain age-specific estimates of death proba-
bility from CHD. The cumulative death probability from
Health care records for the period between January 1, 2000 CHD was defined as the cumulative number of CHD deaths
and December 31, 2010, were retrieved for analysis. We divided by 1308.
 Pediatric congenital heart disease in Taiwan
Table 1 CHD disease-specific prevalence rate (per 100,000) and test for gender predominance.
ICD-9-CM Disease Severe (S)/ Prevalence (10
5) p Prevalence (10
5) for children
code Simple (E) Total Male Female of 0e6 years old (95% CI)

745 Bulbus cordis anomalies and anomalies E 24.0 22.5 25.6 0.0265 21.3 (19.1e23.9)
of cardiac septal closure (excluding
745.0, 745.1, 745.2, 745.3, 745.4, 745.5,
745.6)
745.0 Common truncus S 7.3 7.0 7.7 0.6500 6.0 (4.8e7.4)
745.1 Transposition of great arteries (including S 15.5 19.8 10.9 <0.0001 18.7 (16.5e21.1)
745.10, but excluding 745.11, 745.12)
745.11 Double outlet right ventricle S 9.5 10.1 8.9 0.1517 13.7 (11.9e15.8)
745.12 Corrected transposition of great arteries S 1.9 2.2 1.5 0.0741 2.5 (1.8e3.5)
745.2 Tetralogy of Fallot S 51.4 55.9 46.5 <0.0001 54.9 (51.1e58.9)
745.3 Common ventricle S 6.5 7.0 6.0 0.1573 6.5 (5.3e8.0)
745.4 Ventricular septal defect E 329.9 316.2 344.8 <0.0001 426.1 (415.4e437.1)
745.5 Ostium secundum-type atrial septal E 318.0 277.9 361.5 <0.0001 477.8 (466.5e489.4)
defect
745.6 Endocardial cushion defect S 15.6 13.1 18.3 <0.0001 16.0 (14.1e18.3)
746.0 Pulmonary stenosis E 79.3 76.0 82.8 0.0075 94.4 (89.5e99.7)
746.1 Tricuspid atresia or stenosis S 3.5 4.1 2.9 0.0363 4.2 (3.2e5.4)
746.2 Congenital Ebstein’s anomaly E 3.7 3.1 4.4 0.0234 3.2 (2.4e4.3)
746.3 Aortic stenosis E 10.1 12.4 7.6 <0.0001 8.2 (6.8e9.8)
746.4 Congenital insufficiency of aortic valve E 3.0 3.5 2.5 0.0442 1.2 (0.8e2.0)
746.5 Congenital mitral stenosis S 1.4 1.3 1.5 0.6355 1.6 (1.0e2.4)
746.6 Congenital mitral insufficiency E 10.9 8.4 13.7 <0.0001 6.1 (4.9e7.6)
746.7 Hypoplastic left heart syndrome S 3.1 3.0 3.2 0.6586 5.1 (4.0e6.4)
746.81 Subaortic stenosis E 1.2 1.2 1.2 0.9509 1.4 (0.9e2.1)
746.82 Cor triatriatum E 1.2 1.1 1.4 0.3980 1.2 (0.8e2.0)
746.83 Infundibular pulmonic stenosis E 4.4 4.4 4.3 0.9513 5.3 (4.2e6.7)
747.0 Patent ductus arteriosus E 129.5 97.1 164.7 <0.0001 153.4 (147.0e160.0)
747.1 Coarctation of aorta E 21.4 22.4 20.3 0.1140 32.3 (29.5e35.5)
747.3 Anomalies of pulmonary artery E 45.5 44.4 46.7 0.2429 51.2 (47.6e55.1)
747.41 Total anomalous pulmonary venous S 7.4 7.8 7.1 0.3603 11.5 (9.8e13.4)
connection
747.42 Partial anomalous pulmonary venous E 2.4 2.0 2.7 0.1168 2.9 (2.2e4.0)
connection
Subtotal of severe CHD 107.1 114.0 99.7 <0.0001 123.5 (117.8e129.5)
Subtotal of simple CHD 853.8 770.3 944.8 <0.0001 1149.6 (1132.0e1167.4)
Total 918.0 839.7 1003.2 <0.0001 1233.7 (1215.5e1252.1)
A p value <0.002 was defined as significant with Bonferroni correction for multiple comparisons.
CHD Z congenital heart disease.

115
116 S.-J. Yeh et al

2.4. Statistical analyses who reached the age of 14 years, during the following
4 years, i.e., approximately 35.2% of its peak value. The
We used the SAS software (SAS, Cary, NC, USA) for statis- prevalence after the age of 14 years decreased further until
tical analyses. The overall and age-specific prevalence of adulthood, but at a slower rate. The age-specific preva-
each CHD subtype were calculated. The age-specific and lence of CHD in 18-year-old individuals was 330.8 per
gender-specific prevalence rates were estimated based on 100,000, approximately 22.5% of the peak value (Figure 1).
the age of prevalent patients in 2010. We estimated prev- The prevalence of simple CHD and severe CHD showed the
alence and its 95% confidence intervals (CI) for each diag- same pattern as total CHD, i.e., a decrease after the age of
nostic code using PROC GENMOD with binomial distribution. 10 years. The prevalence of severe CHD decreased to 72.0
All tests were two-sided. A statistically significant p value at 14 years of age, i.e., approximately 41.8% of its peak
of <0.002 was defined, after including Bonferroni correc- value, and further decreased at a slower rate to 53.2 per
tion for multiple comparisons. 100,000 at the age of 18 years, approximately 30.8% of the
Age-specific mortality and overall mortality due to CHD peak prevalence (Figure 1).
were calculated. The CHD mortality fraction was obtained Because the age-specific prevalence of CHD decreased
by dividing the number of patients who died from CHD by significantly after the age of 10 years, the prevalence of
the total number of deaths. CHD in preschool children aged 0e6 years was also exam-
ined (Table 1). In the age group of 6e10 years, the preva-
lence of CHD was 1233.7 per 100,000, i.e., 1149.6 for
3. Results simple CHD and 123.5 for severe CHD. Severe CHD repre-
sented 10% of all CHD cases. We found that ASDII (477.8 per
A total of 45,119 prevalent patients who fulfilled the 100,000 children) was the most common CHD in this age
inclusion criteria were identified. group, followed by VSD (426.1 per 100,000), PDA (153.4 per
100,000), PS (94.4 per 100,000), and TOF (54.9 per 100,000)
3.1. Prevalence of CHD in pediatric populations (Table 1).

The overall prevalence of pediatric CHD per 100,000 chil- 3.4. Mortality from CHD in pediatric populations
dren was 918.0 (95% CI, 909.6e926.4), and the prevalence
of simple CHD and severe CHD was 853.8 (95% CI, To elucidate the impact of mortality on disease prevalence,
845.7e862.0) and 107.1 (95% CI, 104.3e110.1), respectively the CHD-specific mortality rate was calculated from the
(Table 1). Severe CHD represented 11.6% of all CHD cases. values obtained from the database of Death Statistics in
The overall prevalence was lower than the summation of Taiwan. The CHD-specific mortality was 12.9 per 100,000 in
the simple CHD and severe CHD because of the coexistence children aged 0e5 years; this decreased to 2.7 per 100,000
of the two CHD types in the same patient. The most in the population aged 0e20 years (Table 2). The preva-
common diseases, in descending order, were VSD (329.9/ lence of age-specific CHD-related mortality was the highest
100,000), ostium secundum-type atrial septal defect (ASDII; during infancy (58.8 per 100,000) and declined to 6.2, 2.4,
318.0/100,000), patent ductus arteriosus (PDA; 129.5/ and <1 in children aged 1e2, 2e3, and 3e4 years,
100,000), pulmonary stenosis (PS; 79.3/100,000), tetralogy respectively. More than 90% (137/157 in 2008 and 124/132
of Fallot (TOF; 51.4/100,000), coarctation of aorta (21.4/ in 2009) of pediatric CHD-related deaths occurred within
100,000), endocardial cushion defect (15.6/100,000), and the first 5 years of life. Among the all-cause mortality, the
transposition of great arteries (15.5/100,000) (Table 1). CHD mortality fraction accounted for 11.7% and 6.6% in
These disorders accounted for 35.9%, 34.6%, 14.1%, 8.6%, children aged 0e5 years and 0e20 years, respectively. For
5.6%, 2.3%, 1.7%, and 1.7% of the total CHD cases, each age group, the highest fraction was found in children
respectively. aged 0e1 years and 1e2 years (13% deaths in both age

3.2. Gender preference

Although the prevalence of transposition of great arteries

(19.8 vs. 10.9), TOF (55.9 vs. 46.5), and aortic stenosis (12.4
vs. 7.6) was higher in boys than in girls, the prevalence of
VSD (344.8 vs. 316.2), ASDII (365.1 vs. 277.9), PDA (164.7 vs.
97.1), and endocardial cushion defect (18.3 vs. 13.1) was
higher in girls than in boys (Table 1). Moreover, both simple
and total CHDs were more prevalent in girls than in boys,
and severe CHD was more prevalent in boys.

3.3. Age-specific prevalence in 2010

During infancy, an upward trend was observed for age-

specific prevalence of total CHD that was maintained until
the age of 3 years, followed by a plateau at 3e10 years. The Figure 1 Age-specific prevalence of total, simple, and
prevalence rapidly declined to 516.5 per 100,000 children, severe congenital heart diseases (CHDs) (per 100,000 children).
Pediatric congenital heart disease in Taiwan 117

Table 2 The number of deaths from all causes and congenital cardiac anomalies (ICD-10 death codes Q20eQ28) in 2008e2009.
Age (y) Number of deaths CHD Mortality In an assumed birth cohort
All cause CHD cause mortality fraction Death Cumulative death
(10
5) from CHD*(%) probability probability from
Total Male Female
from CHDy (%) CHD cause (%)
0e1 1675 218 108 110 58.8 13.0 4.50 4.50
2 193 25 9 16 6.2 13.0 0.54 5.05
3 149 10 3 7 2.4 6.7 0.22 5.26
4 115 2 1 1 0.5 1.7 0.04 5.31
5 100 6 2 4 1.4 6.0 0.13 5.44
6e10 432 11 2 9 0.4 2.5 0.24 5.68
11e15 467 6 4 2 0.2 1.3 0.13 5.81
16e20 1242 11 6 5 0.6 0.9 0.24 6.04
0e5 2232 261 123 138 12.9 11.7
0e20 4373 289 135 154 2.7 6.6
CHD Z congenital heart disease.
* Mortality fraction from CHD cause is a percentage that denotes the number of CHD-related deaths divided by the number of deaths
from all causes.
y
Death probability from CHD cause was derived by assuming a cohort with a CHD incidence of 1308 per 100,000 and competing risk of
death from other causes.

groups). The probability of cardiac death from CHD was the However, the proportions of PDA (14.1% vs. 15.4%) and PS
highest during infancy, with a probability of 4.5% that (8.6% vs. 9.3%) were high in the neonatal cohort. High
rapidly decreased to 0.54% in children aged 1e2 years and spontaneous remission rates and successful interventions at
0.22% in children aged 2e3 years. The probability was quite a young age might be the reason for the decreased
low after the age of 5 years. The cumulative probability of proportion of both types of CHD in subsequent follow-ups.
cardiac death by the ages of 5, 10, 15, and 20 years was As shown in Table 1, although the prevalence of CHD in
5.44%, 5.68%, 5.81%, and 6.04%, respectively. children aged 0e6 years was found to be 1233 per 100,000
[similar to total CHD incidence (1308 per 100,000)10], the
overall prevalence of CHDs in children aged 0e18 years was
4. Discussion only 918 per 100,000 (i.e., 70.1% of CHD incidence and
74.5% of CHD prevalence in the population aged 0e6 years).
For studying congenital anomalies, both incidence and Age-specific prevalence decreased dramatically after the
prevalence are investigated but represent different attri- age of 10 years. Less than 30% of patients complied with
butes. Incidence reflects the genetic variation or racial regular cardiac follow-ups up to the age of 18 years,
differences involved in diseases, and prevalence reflects regardless of whether they had simple or severe CHD. Such
the incidence, mortality, and duration of the disease, as a change in age-specific prevalence of CHD cannot be
well as the disease burden for the society in which it explained by the natural or unnatural course of the disease,
occurs. Although CHD prevalence should be close to its because the cumulative probability of death in CHD
incidence, prevalence may be lower after accounting for patients from birth to 5, 10, 15, and 20 years of age was
mortality and patients lost to follow-up. Based on the only 5.44%, 5.68%, 5.81%, and 6.04%, respectively. In
information obtained from our national databases, the addition, there are 19 medical centers in Taiwan providing
following two main findings were observed in the present pediatric cardiac programs under almost full NHI coverage;
study: (1) the prevalence of pediatric CHD patients who medical access for CHD care is easy. Therefore, the lack of
required/received medical care was 918 per 100,000, with willingness to attend continual cardiac follow-up after
a rapid decline after the age of 10 years; and (2) the entering school seems to be the major reason for the
cumulative probability of death by the time the children decline in age-specific prevalence. Similar trends have
were of the age of 10 years and 15 years was 5.68% and been observed in previous longitudinal follow-up stud-
5.81%, respectively. This decline in age-specific prevalence ies.17e19 Mackie et al19 reported that the compliance rate
after the age of 10 years is attributed to poor compliance of for continual cardiac follow-up was only 72% at 6e12 years
older children to medical follow-ups rather than to and 53% at 13e17 years in CHD patients who were diag-
mortality. nosed before the age of 6 years.19 Male patients and those
The most common types of CHDs were VSD, ASDII, PDA, with less severe CHD are less likely to pursue follow-up
PS, and TOF, but the incidence and prevalence differed for care.19 Despite this factor, the noncompliance rate is very
each CHD group.9 The prevalent patients had a higher high in Taiwan, the reasons for which should be investi-
proportion of VSD (35.9% vs. 30.7%), ASDII (34.6% vs. 24.7%), gated in the future. Such low adherence might cause
and TOF (5.6% vs. 4.8%) than the neonatal cohort.10 This unnecessary complications such as bacterial endocarditis,
difference might be explained by the longer natural history hyperviscosity syndrome, and iron deficiency anemia; these
and good survival rate after intervention for these CHDs.9 complications result in increased morbidity, subsequent
118 S.-J. Yeh et al

mortality, and disease burden in children with CHD as they children and adults in the United States, 1999 to 2006. Circu-
grow up. lation 2010;122:2254e63.
Infant mortality in Taiwan is similar to that in the United 4. Marelli AJ, Mackie AS, Ionescu-Ittu R, Rahme E, Pilote L.
States.9,13 Of all infant deaths, congenital malformations, Congenital heart disease in the general population: changing
prevalence and age distribution. Circulation 2007;115:
deformations, and chromosomal anomalies (ICD 10,
163e72.
Q00eQ99) are the leading causes in both the United States 5. Gillum RF. Epidemiology of congenital heart disease in the
and Taiwan, accounting for 19.9% (United States) and 22.8% United States. Am Heart J 1994;127:919e27.
(Taiwan).20 CHD-related infant mortality in Taiwan was 6. Khairy P, Ionescu-Ittu R, Mackie AS, Abrahamowicz M, Pilote L,
approximately 58.8 per 100,000, which was higher than that Marelli AJ. Changing mortality in congenital heart disease.
in the United States (41.46 per 100,000).3 The difference in J Am Coll Cardiol 2010;56:1149e57.
CHD-related infant mortality between these two countries 7. van der Bom T, Zomer AC, Zwinderman AH, Meijboom FJ,
in the age group of 1e5 years was still significant (2.59 and Bouma BJ, Mulder BJ. The changing epidemiology of congenital
1.38 per 100,000 in Taiwan and the United States, respec- heart disease. Nat Rev Cardiol 2011;8:50e60.
tively). In the age group of 6e15 (17) years, the differences 8. Hoffman JI, Kaplan S. The incidence of congenital heart
disease. J Am Coll Cardiol 2002;39:1890e900.
became insignificant (0.30 vs. 0.41 per 100,000). Overall,
9. Moons P, Sluysmans T, De Wolf D, Massin M, Suys B, Benatar A,
the CHD-related mortality rate in U.S. infants was only 71% et al. Congenital heart disease in 111 225 births in Belgium:
of the Taiwanese rate; in preschool children, it was only birth prevalence, treatment and survival in the 21st century.
53%. Because the incidence of severe CHD is approximately Acta Paediatr 2009;98:472e7.
1.5/1000 in both Taiwan and the United States,9 worse 10. Wu MH, Chen HC, Lu CW, Wang JK, Huang SC, Huang SK.
outcomes of CHD infants and children in Taiwan should not Prevalence of congenital heart disease at live birth in Taiwan.
be related to disease severity per se. This should be clari- J Pediatr 2010;156:782e5.
fied by further outcome-based studies. 11. Sadowski SL. Congenital cardiac disease in the newborn infant:
In summary, the prevalence of CHD in pediatric patients past, present, and future. Crit Care Nurs Clin North Am 2009;
who required/received medical care in Taiwan was 918 per 21:37e48.
12. Yu CH, Lue HC, Wu SJ, Chen MR. Heart disease screening of
100,000 for individuals aged 0e18 years and 1233 per
schoolchildren in Taiwan. Arch Pediatr Adolesc Med 2009;163:
100,000 for children aged 0e6 years. Because most (i.e., 233e7.
greater than 90%) of CHD deaths occurred within the first 5 13. Lue HC. Congenital heart disease in the East and West.
years of life (mainly during infancy), the decline in the age- Zhonghua Min Guo Xiao Er Ke Yi Xue Hui Za Zhi 1988;29:
specific prevalence of CHD after the age of 10 years was 39Be45B.
attributed to noncompliance to medical follow-ups because 14. Lee YC, Huang YT, Tsai YW, Huang SM, Kuo KN, McKee M, et al.
of the interference of schooling activity. Even in countries The impact of universal National Health Insurance on pop-
with a national health insurance system and easy medical ulation health: the experience of Taiwan. BMC Health Serv Res
access, poor compliance to medical follow-ups is a common 2010;10:225.
phenomenon in CHD children of school-going age. 15. Statistical Yearbook of Interior, Department of Statistics,
Ministry of Interior. Available at: http://sowf.moi.gov.tw/
stat/year/elist.htm Accessed June 12, 2011.
Acknowledgments 16. Cause of Death Statistics, Department of Health, Executive
Yuan, ROC. Available at: http://www.doh.gov.tw/EN2006/DM/
DM2.aspx?now_f od_ list_noZ10864&class_noZ390&level_
This study was supported by funding from the Far Eastern noZ2 Accessed June 12, 2011.
Memorial HospitaleNational Taiwan Hospital Joint Research 17. Gibbs JL, Monro JL, Cunningham D, Rickards A. Survival after
Program (100-FTN02). surgery or therapeutic catheterisation for congenital heart
disease in children in the United Kingdom: analysis of the
central cardiac audit database for 2000e1. BMJ 2004;328:611.
References 18. Billett J, Cowie MR, Gatzoulis MA, Vonder Muhll IF, Majeed A.
Comorbidity, healthcare utilisation and process of care
1. Centers for Disease Control and Prevention (CDC). Hospital measures in patients with congenital heart disease in the UK:
stays, hospital charges, and in-hospital deaths among infants cross-sectional, population-based study with caseecontrol
with selected birth defectsdUnited States, 2003. MMWR Morb analysis. Heart 2008;94:1194e9.
Mortal Wkly Rep 2007;56:25e9. 19. Mackie AS, Ionescu-Ittu R, Therrien J, Pilote L,
2. Dadvand P, Rankin J, Shirley MD, Rushton S, Pless-Mulloli T. Abrahamowicz M, Marelli AJ. Children and adults with
Descriptive epidemiology of congenital heart disease in congenital heart disease lost to follow-up: who and when?
Northern England. Paediatr Perinat Epidemiol 2009;23:58e65. Circulation 2009;120:302e9.
3. Gilboa SM, Salemi JL, Nembhard WN, Fixler DE, Correa A. 20. Heron M. Deaths: leading causes for 2007. Natl Vital Stat Rep
Mortality resulting from congenital heart disease among 2011;59:1e95.