About Congenital Heart Defects and Our Program

Glidewell J, Farr SL, Book, WM, Botto L, Li JS, Soim AS, Downing K, Riehle-Colarusso T, D’Ottavio AA, Feldkamp ML, Khanna AD, Raskind-Hood CL, Sommerhalter KM, Crume TI. Individuals aged 1-64 years with documented congenital heart defects at healthcare encounters, five U.S. surveillance sites, 2011-2013. American Heart Journal. 2021 May 2;S0002-8703(21):00109-5. https://doi.org/10.1016/j.ahj.2021.04.007

Raskind-Hood C, Gray K, Morgan J, Book W. Hypertension and Heart Failure as Predictors of Mortality in the Adult Congenital Heart Defect (ACHD) Population. Congenital Heart Disease. 2021 April;16(4): 333-355. doi:10.32604/CHD.2021.014384

Everitt I, Hoffman T, Raskind-Hood C, Saraf A, Rodriguez F, Hogue C, Book W. Predictors of 30-Day Readmission Following Congenital Heart Surgery across the Lifespan. Cardiology in the Young. 2020 Sep;30(9):1297-1304. https://doi:10.1017/S1047951120002012. Epub 2020 Aug 5. PMID: 32753074.

Gurvitz M, Dunn J, Bhatt A, Book W, Glidewell J, Hogue C, Lin A, Lui G, McGarry C, Raskind-Hood C, Van Zutphen A, Zaidi A, Jenkins K, Riehle-Colarusso T. Characteristics of Adults with Congenital Heart Defects at Three U.S. Surveillance Sites. Journal of the American College of Cardiology. 2020 July; 76(2): 175-182. https://doi.org/10.1016/j.jacc.2020.05.025

Raskind-Hood C, Saraf A, Colarusso T, Glidewell J, Van Zutphen A, Gurvitz M, Dunn J, Lui G, McGarry C, Hogue C, Hoffman T, Rodriguez F, Book W. Assessing Pregnancy, Gestational Complications, and Comorbidities in Women with Congenital Heart Defects (Data from ICD-9-CM Codes in Three U.S. Surveillance Sites). American Journal of Cardiology. 2020 March; 125(5): 812-819. https://doi.org/10.1016/j.amjcard.2019.12.001

Gaydos, L, Sommerhalter K, Raskind-Hood C, Fapo O, Lui G, Hsu D, Van Zutphen A, Glidewell J, Farr S, Rodriguez F, Hoffman T, Book W. Health Care Transition Perceptions among Parents of Adolescents with Congenital Heart Defects in Georgia and New York. Pediatric Cardiology. 2020; 41:1220-1230. https://doi.org/10.1007/s00246-020-02378-z

Gerardin JF, Raskind-Hood C, Hoffman T, Well A, Rodriguez F, Kalogeropoulos A, Hogue, C, Book W. Lost in the system? Transfer to adult congenital heart disease care - Challenges and solutions. Congenital Heart Disease. 2019 July; 14(4): 541-548. https://doi:org/10.1111/chd.12780

Raskind-Hood C, John K, Book W, Hogue C. Estimates of Adolescent and Adult Congenital Heart Defect Prevalence in Metropolitan Atlanta, 2010, Using Capture–Recapture Applied to Administrative Records. Annals of Epidemiology. 2019 April; 32: 72-77. https://doi.org/10.1016/j.annepidem.2018.11.012

Lui G, McGarry C, Bhatt, A, Book, W., Riehle-Colarusso T, Dunn J, Glidewell J, Gurvitz M, Hoffman T, Hogue C, Hsu D, Obenhaus S, Raskind-Hood C, Rodriguez F, Zaidi A, Van Zutphen A. Surveillance of Congenital Heart Defects among Adolescents at Three Sites in the U.S. Sites. 2019 April. American Journal of Cardiology; 124(1): 137-143. https://doi.org/10.1016/j.amjcard.2019.03.044

Glidewell J, Book, W., Raskind-Hood C, Hogue C, Dunn J, Gurvitz M, Ozonoff A, McGarry C, Van Zutphen A, Lui G, Downing K, Riehle-Colarusso T. Population-Based Surveillance of Congenital Heart Defects among Adolescents and Adults. Birth Defects Research. 2018 November 15; 110(19): 1395-1403. https://doi.10.1002/bdr2.1400

Rodriguez F, Ephrem G, Gerardin J, Raskind-Hood C, Hogue C, Book W. The 745.5 Issue in Code-based, Adult Congenital Heart Disease Population Studies: Relevance to Current and Future ICD-9-CM and ICD-10-CM Studies. Congenital Heart Disease. 2018 January; 13(1): 59-64. https://DOI.10.1111/chd.12563

Manuscripts in Progress

Rodriguez III F, Raskind-Hood CL, Hoffman T, Saraf A, Farr S, Glidewell J, Li J, D’Ottavio A, Botto L, Van Zutphen A, Hsu D, Lui G, Book W. How Well Do ICD-9-CM Codes Predict True Congenital Heart Disease Using Integrated Clinical and Administrative Datasets? A CDC-Based Multi-Institutional Study.

Lui GK, Sommerhalter K, Xi Y, Botto L, Crume T., Farr S, Feldkamp M, Glidewell J, Hsu D, Khanna A, Krikov S, Li J, Raskind-Hood C, Sarno L, Van Zutphen AR, Zaidi A, Soim A, Book W. Healthcare Utilization among Adolescents with Congenital Heart Defects at Five Sites in the United States, 2011-2013

Raskind-Hood CL, Kancherla V, Glidewell J, Farr S, Hoffman T, Li J, D’Ottavio A, Botto L, Feldkamp M, Hsu D, Lui GK, Book WM. Racial, Disparities and Socioeconomic Mediators in Healthcare Utilization and Mortality for Patients with Congenital Heart Defects.

Spears T, D’Ottavio A, Book W, Raskind-Hood C, … Whitehead K, Hoskoppal A, Feldkamp M, Botto L. Pregnancy and Pregnancy Outcomes in Women with Congenital Heart Disease: A Population-based Assessment from Five Sites in the U.S.

Health Care Utilization Patterns among Adults with Congenital Heart Defects from an Integrated Surveillance System.

Spectrum and Prevalence of Heart Failure in Children and Adults with Congenital Heart Disease.

Prevalence of Co-morbidities in Congenital Heart Disease: An Analysis of Congenital Heart Defects Across the Lifespan.

Cognitive Disorders Among Individuals with Congenital Heart Defects.

Goldstein S, Li JS, D’Ottavio A, Spears T, Chiswell K, Hartman RJ, Kemper AR, Forestieri N, Hoffman TM, Walsh MJ, Sang CJ, Welke K, Krikov S, Raskind-Hood CL, Book WM, Botto L. Neonatal Outcomes of Pregnancy in Women with Congenital Heart Disease.

Healthcare Expenditures in Children with Congenital Heart Defects Insured by Medicaid in Georgia and New York.

Reynolds E, Raskind-Hood CL, Book W, Blake S. Mental Health Service Utilization among Congenital Heart Defect Medicaid Patients Experiencing Pregnancy.

Rink D, Kancherla V, Raskind-Hood C, Rodriguez F, … Book, W Association between Combined Anatomic and Physiologic Classification of Adults with Congenital Heart Defects and Selected Healthcare Utilization and Clinical Outcomes.

Alam Z, Raskind-Hood C, …. Book W. The Risk of Cerebrovascular Accident among Patients with Congenital Heart Disease.

Graham N, Raskind-Hood C, Book W, Rodriguez, F. Association between Diabetes Mellitus and Hospitalization in the Adult Congenital Heart Disease Population.

Bell MN, Book W, Hoffman T, Raskind-Hood CL, Downing K, Glidewell J, Farr SL, Kamaleswaran R. A Machine Learning-based Model for Predicting Congenital Heart Defects from Administrative Data.

Licitra G, Raskind-Hood C, Ivey, L, Hoffman T, Hogue C, Rodriguez F, Book W. Predictors Associated with 30-day Readmission among a Cohort of Adult Congenital Heart Disease Patients, Medicaid Claims Data 2010-2013.

Raskind-Hood C, Lowe J, Ivey L, Hogue C, Book W. Severity of Congenital Heart Disease as a Predictor for Preterm Birth.

Kancherla V, Raskind-Hood CL, Elani, H, Book W. Dental care utilization among individuals with congenital heart defects enrolled in Medicaid.

Validation of CCS Codes 177- 196 to Detect Pregnancy in the Georgia Congenital Heart Defect Repository, 2011-2013.

List of MPH Thesis and Discovery Projects from Emory RSPH and SOM

1. Kline, M. Care Retention Affects Pregnancy Outcomes in Women with Congenital Heart Disease (School of Medicine, expected 2022)

2. Chen, J. Risk of Hospitalization and Emergency Department Visits among Congenital Heart Defect Patients with Mental Health Disorders. (RSPH, Department of Epidemiology, expected 2022).

3. Zhang, Y. Association between Influenza and Healthcare Utilization among Children and Adolescents Aged 1-18 years with Congenital Heart Defects (RSPH, Department of Epidemiology, 2021)

   Abstract

   Background: The population with congenital heart defects (CHD) continues to grow due to improved survival, and so does the burden on the healthcare system as CHD cases require continuous and specialized care across the lifespan. CHD anatomic complexity is a risk factor for healthcare utilization of patients with CHD. Evidence of excess healthcare utilization attributable to influenza among pediatric patients with chronic conditions requires assessment of the association between CHD anatomic group and healthcare utilization after patients contract influenza.
   Methods: This retrospective secondary analysis assessed the association between CHD anatomic group and 30-day outpatient healthcare utilization among a pediatric and adolescent cohort with CHD and influenza. Clinical and administrative electronic healthcare records (eHR) between 2008-2013 were examined for 2,184 children and adolescents aged 1-19 years with CHD and an influenza diagnosis. CHD Anatomic complexity was categorized as complex, shunt, valve or shunt+valve, and outpatient utilization was determined from encounters that occurred within 30 days of an influenza diagnosis. Poisson regression models with robust variance estimates were applied to estimate crude and adjusted relative risks (cRR and aRR) and 95% confidence intervals (CIs).
   Results: Occurrence of any or none outpatient encounters within 30-days of an influenza diagnosis differed across CHD anatomic groups (shunt: 31.6% vs 32.1%; valve: 20.0% vs 26.7%; shunt+valve: 22.0% vs 17.4%; complex: 26.4% vs 23.8%). There was no association between CHD anatomic group and outpatient utilization after adjusting for age, race, ethnicity, hypertension, and heart failure, aside from comparison of the shunt+valve group with the shunt group. Patients with shunt+valve lesions were at a slightly increased risk of having outpatient visits within 30-days of an influenza diagnosis compared to patients with shunt lesions (aRR: 1.09; 95% CI:1.00-1.19), whereas no difference in risk existed between the valve group and shunt group (aRR: 0.92; 95% CI: 0.84-1.02) or the complex group and the shunt group (aRR: 0.98; 95% CI: 0.90-1.08).
   Conclusions: Findings suggest an association between CHD anatomic group and one-month outpatient healthcare utilization after an influenza diagnosis among children and adolescents with CHD. Future studies should further examine this association in other populations, and using prospective data.

4. Alam, Z. Risk of Cerebrovascular Accident among Patients with Congenital Heart Disease (School of Medicine, 2021)

   Abstract

   Background: Congenital heart disease (CHD) survivors have an increased risk of cerebrovascular accident (CVA), although it is unclear how this risk varies between patients with CHDs of varying anatomic complexity. This study investigated whether having anatomically complex CHD increased a subsequent risk of CVA compared to patients with less anatomically complex CHD.
   Methods: This was a retrospective cohort study of 13,971 patients aged 11-64 (mean age 26.8, 43% male) who were seen at either Emory Healthcare (EHC), Children’s Healthcare of Atlanta (CHOA), or Sibley Heart Center between 2008 and 2013. Patients were grouped based on CHD anatomy (27% complex, 24% shunt, 42% valve, 7% shunt + valve) with complex CHD including Transposition of the Great Arteries (TGA), Tetralogy of Fallot (ToF), Endocardial Cushion Defects, Pulmonary valve atresia, Tricuspid atresia, HLHS, Interrupted Aortic Arch, and Total anomalous pulmonary venous return. CVA was identified as any diagnosis of occlusion of cerebral arteries, stenosis of cerebral arteries, cerebral thrombosis, transient cerebral ischemia, generalized ischemic cerebrovascular disease, or post-operative stroke. Multivariate logistic regression models were used to estimate adjusted relative risk (aRR) and 95% confidence intervals (CIs).
   Results: Among 547 patients with CHD who had any stroke within the study period (3.9%), 20.7% were in the complex CHD group (mean age 25.4), 25.4% were in the shunt group (mean age 26.3), 46.6% were in the valve group (mean age 28.3), and 7.3% were in the shunt and valve group (p < 0.05). When age, insurance, geographic distribution, diabetes, hypertension, hyperlipidemia, endocarditis, heart failure and cyanosis were controlled for, CHD anatomic complexity did not affect the odds of developing a CVA.
   Conclusion: Patients with complex CHD are at an increased risk of CVA. Further studies are needed to explore this association in broader populations.

5. Kapera, O. Association of Heart Failure and Pulmonary Hypertension among Individuals with Coexisting Congenital Heart Defects and Down Syndrome (RSPH, Department of Epidemiology, 2021)

   Abstract

   Background: Congenital heart defects (CHD) are the most common congenital defect in the U.S., accounting for 1% of annual births. CHD complications include pulmonary hypertension (P-HTN) and heart failure (HF). Individuals with CHD are also more likely to have Down syndrome (DS). Despite medical advancements, P-HTN and HF present particular challenges among those affected by CHD, and this association has not been well examined by presence or absence of co-occurring DS. The proposed study investigates if individuals with CHD with or without coexisting DS are at increased risk for P-HTN and/or HF.
   Methods: This a retrospective secondary data analysis of 22,499 CHD patients aged 1-64 years identified using ICD-9-CM codes from healthcare encounters that occurred between 1/1/2008-12/31/2013 from 11 clinical and/or administrative data sources. Multivariate logistic regression was used to assess adjusted relative risks (aRR) and 95% confidence intervals (CIs) of P-HTN without HF, and HF-without P-HTN, among patients with CHD grouped by DS status.
   Results: Overall, 9.3% of individuals with CHD had DS in our analytic study sample. Among all CHD cases, 4.7% had P-HTN without HF, 21.2% had HF without P-HTN, and 0.6% had both P-HTN and HF. Among individuals with CHD and DS, 7.6% had P-HTN without HF, 9.9% had HF without P-HTN, and 5.7% had both P-HTN and HF. Our regression analysis showed a 23% lower risk of P-HTN without HF among those with CHD and DS compared to patients with CHD without DF (aRR=0.77; 95% CI: 0.59-1.00). On the contrary, we found a significantly increased risk of HF without P-HTN in patients with CHD and DS compared to those with CHD without DS (aRR=1.51; 95% CI: 1.04-2.18). We were unable to analyze the group with both P-HTN and HF due to small number of affected individuals.
   Conclusions: We found that co-occurring DS among individuals with CHD can impact the development of P-HTN and HF overtime. Comorbidities, like cyanosis and atrial arrythmia, and sleep apnea may affect risk of P-HTN and HF for patients with CHD and co-occurring DS, and should be further explored. Future studies should aim at improved assessment of clinical variables and potential confounders.

6. Graham, N. Association between Diabetes Mellitus and Hospitalization in the Adult Congenital Heart Disease Population. (School of Medicine, 2021).

   Abstract

   Background: For an aging congenital heart disease (CHD) population, the impact of noncardiac comorbidities, such as diabetes mellitus (DM) and other potentially modifiable risk factors, is recognized, but medical consequences are not well quantified. This study assessed the association between DM, in addition to covariates including socio-demographic and clinical factors, and hospitalizations among adults with congenital heart disease (ACHDs).
   Methods: This retrospective cohort study identified 4,886 ACHDs, 20-64 years old (average age 37.8 years, 61.6% female) with 2008-2013 healthcare encounters from linked administrative health records. DM, including types I and II, was defined by at least 1 of 64 ICD-9-CM codes. Other comorbidities, including obesity, hypertension, and hyperlipidemia, were also defined by the presence of at least one of their respective ICD-9-CM codes. Hospitalization was defined as a hospital admission requiring an overnight stay for any reason. Multivariate logistic regression was used to estimate the association.
   Results: 17.1% of ACHDs with at least 1 hospitalization had DM compared to 2.5% of their non-hospitalized counterparts (p<0.0001). When age, gender, race, insurance, CHD anatomic group, obesity, hypertension, hyperlipidemia, and neighborhood poverty were controlled, odds of hospitalization were 2.78 times (95% confidence interval (CI): 2.07-3.74) greater in ACHDs with DM, compared to those without DM. When controlling for these same factors, obesity (adjusted prevalence odds ratio (aPOR) 1.83; 95% CI: 1.46-2.29), hypertension (aPOR 2.64; 95% CI: 2.24-3.11), and hyperlipidemia (aPOR 1.23; 95% CI: 1.02-1.50) were also all associated with increased hospitalization odds.
   Conclusion: ACHDs with DM, in addition to those with obesity, hypertension, and hyperlipidemia, are at increased risk of hospitalization. These findings display the burden of potentially modifiable risk factors in this population. Further research is needed to determine the reasons for increased hospitalization among ACHDs with DM and other potentially modifiable risk factors to better understand and identify interventions that can improve outcomes for these patients.

7. Rink, D. Association between Combined Anatomic and Physiologic Classification of Adults with Congenital Heart Defects and Selected Healthcare Utilization and Clinical Outcomes. (RSPH, Department of Epidemiology and School of Medicine, 2021)

   Abstract

   Background: Classifying complexity of congenital heart disease in adults (ACHD) through native anatomy alone based on ICD codes may not identify those at risk of adverse outcomes. Incorporating physiologic comorbidities into classification may improve the ability to predict adverse outcomes using administrative data. The objective of this study is to examine the association between combined anatomic and physiologic classification of congenital heart disease (CHD) complexity with healthcare utilization and adverse clinical outcomes among adults.
   Methods: Data from Georgia Medicaid claims and Emory Healthcare electronic health records (eHR) were examined for adult patients aged 18-45 years with a CHD-related diagnosis with encounters from 2008 to 2013. Anatomic complexity was examined and categorized as complex anatomy or shunt and/or valve. ACHD guideline-based physiologic comorbidities captured at one year were used to determine physiologic classification, categorized as A/B or C/D. Healthcare utilization (i.e., hospitalizations and emergency department (ED) visits) and adverse clinical outcomes (i.e., transplantation and mortality) were examined for one year. Adjusted relative risks (aRR) and 95% confidence intervals (95% CI) were estimated using multivariable logistic regression.
   Results: Among 2,384 eligible patients, 34.4% had complex anatomy and 41.6% had C/D physiology. Overall, 10.2% had at least one hospitalization and 8.3% had at least one ED visit. There were 22 deaths and one transplant with no significant group differences by combined anatomic and physiologic classification status. The risk of any hospitalization for those with complex ACHD and C/D physiology was 31.2 (aRR 31.2, 95% CI: 11.9, 81.6) times higher than those with shunt and/or valve anatomy and A/B physiology over 1-year of follow-up. The risk of having any ED visit for those with complex ACHD and C/D physiology was 10.6 (aRR 10.6, 95% CI: 3.4, 33.5) times higher than those with shunt and/or valve anatomy and A/B physiology over 1-year of follow-up.
   Conclusions: Physiologic comorbidities provide additional information compared to native anatomy alone in assessing outcomes in adults using healthcare administrative databases. Future analyses should examine the associations noted in this study and apply alternative study designs that may better handle influential covariates and potential confounders that inform outcomes.

8. Reynolds, E. The Effect of Rurality on Utilization of Psychotherapy for Perinatal Mood Disorders among Georgia Congenital Heart Defect Medicaid Patients (RSPH, Department of Health Policy and Management, 2020)

   Abstract

   Background: As the number of people born with congenital heart defects (CHD) living into childbearing years increases, the need to assess possible health risks of pregnant women living with CHD is paramount. Perinatal mood disorders (PMD) affect 15-20% of women experiencing pregnancy and are the number one complication in pregnancy and childbirth in the U.S, yet remain undertreated. Literature suggests that a major barrier to the receipt of treatment, such as psychotherapy, is geographic access to mental health providers. There is no current research concerning the receipt of psychotherapy for PMD in women with CHD. Therefore, this study seeks to address this gap in knowledge utilizing the geographic distribution of Georgia’s CHD Medicaid beneficiaries.
   Methods: This study uses Medicaid claims data from the Medicaid Analytic Extract (MAX) files spanning the years 1999-2013 to assess the effect of rurality on the receipt of psychotherapy for PMD in a CHD population. A two-part model using logistic regression analysis will be used to evaluate the receipt of any treatment and adequate treatment, given any treatment, across the urban to rural continuum.
   Results: Results indicate that of the 5,235 women with CHD who had a delivery, 931 (17.8%) had a diagnosis of PMD. Only approximately 15% of these women received any psychotherapy. Those living in rural areas were significantly less likely to receive any psychotherapy within 12 weeks of diagnosis of PMD as compared to their urban counterparts, even after accounting for differences in geographic access.
   Conclusion: This study suggests that pregnant and postpartum women with CHD with a PMD diagnosis who live in rural areas are less likely to receive psychotherapy. These findings provide support for adopting a prescriptive approach to perinatal mental health services for women with chronic disease and encourage the formation of public health policies that address barriers to mental health treatment in the perinatal period.

9. Gray, K., Hypertension and Heart Failure as Predictors of Mortality in the Adult Congenital Heart Defect (ACHD) Population. (Mercer University College of Health Professions Department of Public Health, 2019)

   Abstract

   Early intervention to prevent premature mortality is vital for adults with congenital heart defects (CHD). Anatomic complexity and comorbid conditions are thought to contribute to CHD mortality. Since hypertension (HTN) and heart failure (HF), comorbid conditions among the most prevalent causes of death in the United States, commonly accompany CHD, it is crucial to evaluate whether they are reliable predictors of mortality for adults with CHD (ACHD) independent of anatomic CHD complexity. A retrospective cross-sectional analysis of ACHD, aged 18-64, with concomitant HTN and/or HF and at least one health care encounter during 2008-2010 were assessed. Of 7,061 ACHD patients (18.3% HTN without HF, 5.6% HF without HTN, 12.4% with both), 3.5% died (n=244) during the study period. Overall, the sample was 50.6% white, 55.2% female, and 20.3% had a severe CHD. Among those who died, 20.1% had HTN without HF, 17.6% had HF without HTN, and 46.7% had both. Crude analyses revealed that older age and black race were associated with increased mortality during the three-year study period compared to those of younger age and white race. HF with or without HTN was also associated with increased mortality. ACHD patients diagnosed with shunt or valve CHDs, were also more likely to die compared to those with a severe CHD. After adjustment for age, race and CHD complexity, ACHD patients with HF with or without HTN were equally likely to die during the study period. However, ACHD patients with shunt and valve defects, also diagnosed with HF without HTN, were more likely to die during the three-year study period compared to patients with complex CHDs.

10. Lowe, J. Severity of Congenital Heart Disease as a Predictor for Preterm Birth. (RSPH, Department of Epidemiology, 2019)

    Abstract

    Purpose: To examine if preterm birth risk varies by congenital heart defect (CHD) severity.
    Methods: This study is a retrospective cohort design analyzing pregnant and non-pregnant female patients with CHD who were identified by encounters occurring between 1/1/2011-12/31/2013 in an existing Emory CHD surveillance repository. Women were linked to Georgia birth certificates during this time to examine the association between severity of CHD and preterm birth.
    Results: Among the initial cohort of 2,523 women aged 12-55, 1,525 (60.4%) had at least one pregnancy diagnosis code in their administrative record, but did not match to a birth certificate; 129 (5.1%) women matched to a birth certificate, but had no pregnancy diagnosis codes in their record; and 869 (34.4%) women had both pregnancy diagnosis codes and a matched birth certificate. After excluding women without a birth certificate match, without a pregnancy diagnosis code, or who only had a 745.5 code in isolation, we retained 823 women for further analyses. Overall, 23.9% (197/823) births were preterm and 43.4% (357/823) had a severe CHD. Both crude and adjusted analyses revealed that preterm birth was not significantly different for women who had a severe compared to those who had a not severe CHD.
    Conclusion: CHD severity may not be associated with preterm birth risk. However, failure to match a large segment of this sample with their birth outcomes may have biased the results towards the null. Further, a real difference in preterm birth risk may have been masked with differential misclassification of exposures because of issues with either the Marelli severity classification schema not sufficiently categorizing important CHD diagnoses for adverse pregnancy outcomes, or with administrative data using ICD-9-CM codes that comprise certain CCS categories that may not adequately differentiate obstetrical complications and comorbidities such as hypertension. To explore these hypotheses, integrated records are vital for patients with CHD, especially for women with CHD who are of reproductive age, to better manage their care and understand their risks during pregnancy.

11. Everitt, I. Predictors of 30-Day Readmission Following Congenital Heart Surgery across the Lifespan. (School of Medicine, 2019).

    Abstract

    Background: Hospital readmission is an important driver of costs among patients with congenital heart disease (CHD). We assessed predictors of 30-day rehospitalization following cardiac surgery in CHD patients across the lifespan.
    Methods: This was a retrospective analysis of 981 patients with CHD who had cardiac surgery between January 2011 and December 2012. A multivariate logistic regression model was used to identify demographic, clinical, and surgical predictors of 30-day readmission. Receiver operating curves (ROC) derived from multivariate logistic modeling were utilized to discriminate between patients who were readmitted and not-readmitted at 30 days. Model goodness of fit (GOF) was assessed using the Hosmer-Lemeshow (HL) test statistic.
    Results: This was a retrospective analysis of 981 patients with CHD who had cardiac surgery between January 2011 and December 2012. A multivariate logistic regression model was used to identify demographic, clinical, and surgical predictors of 30-day readmission. Receiver operating curves (ROC) derived from multivariate logistic modeling were utilized to discriminate between patients who were readmitted and not-readmitted at 30 days. Model goodness of fit (GOF) was assessed using the Hosmer-Lemeshow (HL) test statistic.
    Conclusion: Readmission following congenital heart surgery is common. Risk factors for readmission include a history of cardiac surgery, longer length of stay, and comorbid conditions. This information may serve to guide efforts to prevent readmission and inform resource allocation in the transition of care to the outpatient setting. This study also demonstrates the feasibility of linking a national subspecialty registry to a clinical and administrative data repository to follow longitudinal outcomes of interest.

12. Licitra, G. Predictors associated with 30-day Readmission among a Cohort of Adult Congenital Heart Disease Patients, Medicaid Claims Data 2010-2013. (RSPH, Department of Epidemiology, 2018).

    Abstract

    Purpose: Advancements in medical technology and treatment of CHD has led to improved survival. Adults with CHD (ACHD) have been largely understudied, especially with respect to hospital readmissions - a quality indicator commonly used to assess healthcare utilization. This study aims to identify common diagnoses at index admission associated with readmissions within 30-days of a cardiac-related hospital admission among ACHD patients. We also analyze socio-contextual factors associated with 30-day readmission including race, sex, and distance from patient zip code to a facility providing specialized cardiology care.
    Methods: Advancements in medical technology and treatment of CHD has led to improved survival. Adults with CHD (ACHD) have been largely understudied, especially with respect to hospital readmissions - a quality indicator commonly used to assess healthcare utilization. This study aims to identify common diagnoses at index admission associated with readmissions within 30-days of a cardiac-related hospital admission among ACHD patients. We also analyze socio-contextual factors associated with 30-day readmission including race, sex, and distance from patient zip code to a facility providing specialized cardiology care.
    Results: Of the 1,697 patients included in the analysis, about 22% had a 30-day hospital readmission. The remaining 78% had no recorded hospital readmissions within 30 days. Readmission was higher for those at initial admission with comorbid congestive heart failure (CHF) non-hypertensive (13% higher in readmitted patients compared to non-readmitted patients), fluid and electrolyte disease (15% higher), and diabetes mellitus (DM) with complications (~7% higher). Readmission was also higher for black patients compared to white (~6% higher). After adjustment for potential confounding variables, there was no significant increase in hazard found for greater distance between patient’s resident zip code and healthcare facilities providing cardiology care.
    Conclusion: Specific comorbid diagnoses at index admission and black race identify subpopulations at increased risk for 30-day hospital readmission. Increased attention to comorbid diagnoses during initial hospital stay may reduce 30-day hospital readmission rates and improve health outcomes for the ACHD population.

13. Josias Sejour, D. Prevalence of Pregnancy-related Complications among CHD Patients Enrolled in Medicaid during 1998-2013 (RSPH, Department of Behavioral Sciences and Health Education, 2017)

    Abstract

    Purpose: Those with congenital heart defects (CHD) require on-going care and lifelong cardiac surveillance. However, individuals with CHD are often lost to follow up as they transition from childhood to adulthood. While reasons for lapses in medical care vary, studies have found that a change in or a loss of medical insurance are two major causes of lapses in care for individuals with CHD. Due to the increased risk of pregnancy related complications, it is imperative that women with CHD receive adequate medical care prior to and during pregnancy. The current study determined whether prevalence of pregnancy-related complications differed by history of Medicaid enrollment among pregnant women with CHD.
    Methods: Medicaid claims were limited to female patients who were coded as having at least one CHD diagnosis in the years 1999-2007 with at least one pregnancy-related diagnosis in 2008-2013. Using multivariable logistic regression, odds ratios were calculated between Medicaid enrollment history and pregnancy-related complications.
    Results: The analytic sample retained was 1,799 women. Of those, 557 (31%) were continuously enrolled in Medicaid from 1999-2007, while 1,242 (69.0%) were occasionally enrolled in Medicaid from 1999-2007. With respect to pregnancy-related complications, 206 (11.5%) had cardiovascular complications, 476 (26.5%) experienced neonatal/fetal loss, 1,426 (79.3%) had maternal complications and 419 (23.3%) experienced complications in pregnancy. While history of Medicaid enrollment was not a significant predictor of cardiovascular complications, complications during pregnancy or neonatal/fetal loss, it was a significant predictor of complications during delivery for women aged 19 or older. Pregnant women > 19 with CHD who were only occasionally enrolled in Medicaid were more likely to have complications during delivery than those who were continuously enrolled in Medicaid.
    Conclusion: Results suggest an association between history of enrollment in Medicaid and certain pregnancy-related complications among pregnant women with CHD. More research is needed to further examine this relationship, especially with the inclusion of previously uninsured women with CHD who only become eligible for Medicaid because of their pregnancy. Subsequently, to assess this relationship, there is a need for additional data sources that provide more accurate reporting of medical histories for Medicaid patients with CHD. Given that the majority of this CHD sample were occasionally enrolled in Medicaid, and given that there is an ever-growing number of individuals with CHD surviving into adulthood, these findings indicate the need for a re-assessment of Medicaid's eligibility requirement for adult disability status.

14. Jennifer Gerardin, MD Lost in the system? Transfer to adult congenital heart disease care – challenges and solutions (CHD Fellow, Emory School of Medicine, 2015-2017)

    Abstract

    Objective: Transfer of congenital heart disease care from the pediatric to adult setting has been identified as a priority and is associated with better outcomes. Our objective is to determine what percentage of patients with congenital heart disease transferred to adult congenital cardiac care.
    Design: A retrospective cohort study.
    Setting: Referrals to a tertiary referral center for adult congenital heart disease patients from its pediatric referral base.
    Patients: This resulted in 1514 patients age 16-30, seen at least once in three pediatric Georgia health care systems during 2008-2010.
    Interventions: We analyzed for protective factors associated with age-appropriate care, including distance from referral center, age, timing of transfer, gender, severity of adult congenital heart disease, and comorbidities.
    Outcome measures: We analyzed initial care by age among patients under pediatric care from 2008 to 2010 and if patients under pediatric care subsequently transferred to an adult congenital cardiologist in this separate pediatric and adult health system during 2008-2015.
    Results: Among 1514 initial patients (39% severe complexity), 24% were beyond the recommended transfer age of 21 years. Overall, only 12.1% transferred care to the referral affiliated adult hospital. 90% of these adults that successfully transferred were seen by an adult congenital cardiologist, with an average of 33.9 months between last pediatric visit and first adult visit. Distance to referral center contributed to delayed transfer to adult care. Those with severe congenital heart disease were more likely to transfer (18.7% vs 6.2% for not severe).
    Conclusion: Patients with severe disease are more likely to transfer to adult congenital heart disease care than non-severe disease. Most congenital heart disease patients do not transfer to adult congenital cardiology care with distance to referral center being a contributing factor. Both pediatric and adult care providers need to understand and address barriers in order to improve successful transfer.

15. Claxton, J. Estimate of Care Non-Continuity among Medicaid Beneficiaries Diagnosed with Congenital Heart Defects in Five Metropolitan Georgia Counties: 1999-2010 (RSPH, Department of Epidemiology, 2016)

    Abstract

    Purpose: Continuity in healthcare for individuals with a congenital heart defect (CHD) is an important public health issue. The aim of this study was to estimate the percentage of Georgia Medicaid beneficiaries diagnosed with a CHD sometime during 1999-2007 who also had a Medicaid-paid claim during 2008-2010. Medicaid claims paid during 2008-2010 were analyzed to determine the extent to which age, gender, and disease severity explain the likelihood of healthcare indicating their CHD.
    Methods: Medicaid data were used to identify a CHD cohort, ages 9-62 years. Using multivariable logistic regression, odds ratios were calculated between age and having a Medicaid-paid claim during 2008-2010 and, among those with a claim, between age and having a CHD-related diagnosis on the claim.
    Results: 5,944 patients had a CHD-related diagnosis on their Medicaid claim during 1999-2007; only 46% also had a Medicaid-paid claim during 2008-2010. After excluding those known to have left the catchment area, 52% had at least one Medicaid-paid claim; only 10% (522 of the 5,285) had a CHD-related diagnosis. Among 214 females less than 18 with a severe CHD classification, 115 (53%) had a Medicaid claim during 2008-2010, with 60% having a claim that included a CHD diagnosis. They were the most likely to have a Medicaid claim with a CHD diagnosis. With them as referent, males over 18 years regardless of CHD severity were less likely to have any Medicaid-paid claims during 2008-2010; further, among those with claims, almost every combination of age, sex, and CHD severity was less likely to have a Medicaid claim with a CHD diagnosis.
    Conclusion: Among Medicaid patients in Georgia known to have CHD, during a three-year period surveillance of claims for CHD adolescents and adults, only 10% were identified by a Medicaid claim indicating CHD. As adolescents transition into adulthood, many no longer meet the requirements for Medicaid coverage in Georgia unless they are pregnant. Pregnant women with CHD need to be identified and referred for specialty care. Georgia needs to address implementing Medicaid expansion to cover individuals who otherwise may not be able to afford health coverage.

16. John, K. Congenital Heart Defect Prevalence and Capture-Recapture Methodology (RSPH, Department of Epidemiology, 2016)

    Abstract

    Purpose: : To determine the congenital heart defect (CHD) prevalence in five metropolitan counties (Clayton, Cobb, DeKalb, Fulton, and Gwinnett) in Georgia from 2008-2010 using capture-recapture methodology.
    Method: Using data from Children's Hospital of Atlanta (CHOA), Sibley Heart Center Cardiology, Pediatric Cardiology Services (PCS), Grady Health, Emory Healthcare including St. Joseph's Hospital, and Georgia Medicaid claims, capture-recapture (CR) methodology and logistic regression were employed to estimate the prevalence of CHD for both adolescents, aged 11-20, and adults aged 21-64, in five metropolitan Atlanta, Georgia counties. From this, the number of CHD cases that were missed were estimated by these data sources from January 1, 2008 to December 31, 2010.
    Results: Altogether 1,858 adolescent cases were captured from at least one "adolescent" database (CHOA, Sibley, PCS, and Medicaid), and 3,183 adult cases were captured from at least one "adult" database (Emory Healthcare, St. Joseph's Hospital, Grady Health, and Medicaid). The estimated number of adolescents (aged 11-20 years) with CHD and living in the 5 metropolitan Atlanta counties in Georgia was 3,718 (95%CI: 3,471 - 4,004) for a prevalence estimate of 7.85 per 1,000 population aged 10-19 in 2010. The number of adults with CHD aged 21-64 years was estimated to be 12,969 (95%CI: 13,873 -18,915) for a prevalence estimate of 6.08 per 1,000 population aged 20-64 in 2010.
    Conclusion: Despite the need for lifelong care, adults with CHDs are being lost within the healthcare system. Public health initiatives should focus on the high proportion of adult CHDs retained in adolescent care. Lack of referrals and patient retention in adolescent care provides context for the need of more specialized adult congenital care units and to mandate policies, such as patient referrals, to assist physicians with coordinated transfer of patients to adult care.

17. Deshmukh, A. In or Out of Care: Congenital Heart Defects and the Impact on Health Outcomes (RSPH, Department of Epidemiology, 2015)

    Abstract

    Purpose: To determine 1) how many Georgia adolescents with a Congenital Heart Defect (CHD) received continuous care from 2008-2010; 2) how many of them successfully transitioned into adult congenital care; and 3) a predictive model of risk factors for loss to follow-up and successful transition.
    Method: Data from an ongoing pilot CHD surveillance project were used to identify a cohort of adolescent patients, 16-21 years old seen at Sibley Heart Center, Pediatric Cardiology Services (PCS), or Children's Healthcare of Atlanta (CHOA) during 2008 or 2009. Evidence of transitioning into adult care was searched for in Emory Healthcare, St. Joseph's Hospital, Grady Health, or Georgia Medicaid data during 2008-2010. Odds ratios were calculated using multivariable logistic regression.
    Results: After controlling for age, sex, insurance, proximity, CHD severity, number of procedures, and comorbidities, more than half (53.6%) of the adolescents were lost to follow-up and only about 20% successfully transitioned into adult congenital care. Being older and female predicted loss to follow-up, while severity, procedure history and having a comorbidity were protective. Being older, and having public insurance, a severe CHD, a non-CHD birth defect, and a respiratory/pulmonary comorbidity predicted successful transitioning.
    Conclusion: As adolescent patients age, follow-up care and proper transitioning into an adult congenital heart defect practice must be reinforced. Implementing a national CHD surveillance program and continuing research on the factors affecting loss to follow-up and successful transitioning can help increase specialized healthcare utilization for those living with a CHD.

18. Oandasan, P. Evaluating the Cardiovascular Risk Factors of Adult Congenital Heart Defects (RSPH, Career MPH, Epidemiology, 2015)

    Abstract

    Context: Due to advances in medical technology and treatment, patients born with congenital heart defects (CHDs) are living longer, and their longer life expectancy puts them at risk for developing other acquired diseases like Diabetes Mellitus (DM). DM patients and Adult Congenital Heart Defect (ACHD) patients are both high utilizers of hospital resources, specifically through their use of hospitalization. However, there has been no research assessing the risk of increased hospitalization among diabetic ACHD patients.
    Purpose: The objective of this study is to examine the relationship between hospitalization and DM among the ACHD population.
    Methods: A cross-sectional study of ACHD patients was conducted using 2008-2010 data from Emory University Hospital (EUH), The Emory Clinic (TEC), and Emory University Hospital-Midtown (EUH-M) in Atlanta, Georgia. Using logistic regression, odds ratios were calculated to assess the association between DM and hospitalization.
    Results: An association between DM and hospitalization among ACHD patients was found (OR=7.8; 95%CI: 5.41, 11.3). After controlling for CHD severity, hypertension, and hyperlipidemia, the adjusted odds of being hospitalized among diabetic ACHD patients was four times greater compared to non-DM ACHD patients (aOR=4.2, 95%CI: 2.9, 6.2).
    Conclusion: DM may play a role in an increased risk of hospitalization among the ACHD population. Further exploration of the main reasons for hospitalization among diabetic ACHD patients is needed to develop strategic care management to prevent such hospitalizations.

> CHD Education Module

We have partnered with The Georgia Department of Public Health, The Centers for Disease Control and Prevention, community organizations, healthcare systems, patient advocacy groups, patients and parents across Georgia.

Funding/Support: Funding/Support received by Centers for Disease Control and Prevention, Grant/Award Number: CDC‐RFA‐DD12-1207

Surveillance period
2008 - 2010

Collaborators (CDC and 3 sites: GA, MA & NY)
Centers for Disease Control and Prevention (CDC): National Center on Birth Defects and Developmental Disabilities (NCBDDD), Atlanta, GA

1. Emory University School of Medicine and Rollins School of Public Health, Atlanta, GA collaborated with:
   • Metropolitan Atlanta Congenital Defects Program (MACDP) - pediatric
   • Georgia Department of Public Health (GADPH) - pediatric and adult
   • Sibley Heart Center - pediatric
   • Pediatric Cardiology Services (now part of Sibley Heart Center) - pediatric
   • Children’s Healthcare of Atlanta (CHOA) - pediatric
   • Emory Healthcare (EHC) - adult
   • St. Joseph’s Hospital (now part of Emory Healthcare) - adult
   • Grady Health - adult
   • Centers for Medicare and Medicaid Services (CMS) (via Research Data Assistance Center (ResDAC) - pediatric and adult
2. Massachusetts Department of Health, Boston, MA
3. New York State Department of Health, Albany, NY

Priorities

• To determine feasibility of population-based CHD surveillance across multiple sites
• To estimate prevalence of CHD among adolescents and adults
• To examine survival, healthcare utilization, longer term outcomes among adolescents and adults with CHD

What we Learned

• Many U.S. children, teens, and adults live with and receive care for CHD
• 20-33% of adolescents and adults with CHD have been diagnosed with a mental health condition
• Many women with CHD become pregnant and experience pregnancy-related health complications
• Severe CHD ranged from 11% (NY) to 20% (GA)
• Medicaid Coverage: > 75% (MA), 31% (GA), 28% (NY)
• Capture of comorbid conditions varied across sites
• Older patients and those with non-complex CHD contribute to health burden
• Cardiac complications and comorbid conditions are common

Estimates of Adolescent and Adult Congenital Heart Defect Prevalence in Metropolitan Atlanta, 2010, Using Capture-Recapture Applied to Administrative Records (Raskind-Hood et al. Annuals of Epidemiology. 2019 Apr;32:72-77.e2)

The 745.5 Issue in Code-based, Adult Congenital Heart Disease Population Studies: Relevance to Current and Future ICD-9-CM and ICD-10-CM Studies (Rodriguez et al. Congenital Heart Disease. 2018 January; 13(1): 59-64).

Assessing Pregnancy, Gestational Complications, and Comorbidities in Women with Congenital Heart Defects (Data from ICD-9-CM Codes in Three U.S. Surveillance Sites). (Raskind-Hood et al. American Journal of Cardiology. 2020. Marv1; 125(5):812-819.

Funding/Support: Funding/Support received by Centers for Disease Control and Prevention, Grant/Award Number: CDC‐RFA‐DD15-1506

Surveillance period
2011 - 2013

Collaborators (CDC and 5 sites: CO, GA, NC, NY & UT)
Centers for Disease Control and Prevention (CDC): National Center on Birth Defects and Developmental Disabilities (NCBDDD), Atlanta, GA

1. Emory University School of Medicine and Rollins School of Public Health, Atlanta, GA collaborated with:
   1. Data Sources:
      • Metropolitan Atlanta Congenital Defects Program (MACDP) - pediatric
      • Georgia Department of Public Health (GADPH) – pediatric and adult
      • Children’s Healthcare of Atlanta (CHOA) - pediatric
      • Sibley Heart Center (now subsumed under CHOA) - pediatric
      • CHOA Society of Thoracic Surgeons (STS) – pediatric
      • Emory Healthcare (EHC) – adult
      • EHC Society of Thoracic Surgeons – adult
      • Grady Health System - adult
      • Piedmont Healthcare – pediatric and adult
      • Wellstar Health System - adult
      • Centers for Medicare and Medicaid Services (CMS) (via Research Data Assistance Center (ResDAC) - pediatric and adult
   2. Community Advisory Committee:
      • St. Joseph’s Hospital (now part of Emory Healthcare)
      • Georgia Department of Public Health (GA DPH)
      • Sibley Heart Center
      • The Children's Heart Foundation
      • Georgia Center for Learning Disabilities – GA State University
      • Adult Congenital Hearts of Greater Atlanta (patient advocacy)
      • Grady Healthcare
      • Wellstar Healthcare
      • Piedmont Healthcare
      • Children’s Healthcare of Atlanta
      • The Urban Health Initiative
      • CHD Healthcare providers
      • Parents of teens with CHD
2. Duke University School of Medicine, Duke Clinical Research Institute, Durham, NC
3. New York State Department of Health, Albany, NY
4. University of Colorado | Anschutz Medical Campus, Departments of Internal Medicine and Pediatrics, Divisions of Cardiology, Aurora, CO
5. University of Utah, Salt Lake City, UT

Priorities

• Prevalence of CHD across the lifespan, healthcare utilization among adolescents and adults, comorbidities, heart failure
• Assessment of barriers to appropriate transition of adolescents to adult cardiac care (NY & GA)
• Education Modules (GA)

What we Learned so Far

• 24% complex anatomy, 37% shunt, 53% valve
• 69% children
• 19% black or multi-racial, 15% Hispanic
• 48% have public insurance, more common for children
• Overall retention in care was low when transitioning from pediatric to adult care
• Severe CHD were more likely to transfer to ACHD care than those with shunt and/or valve lesion
• Transfer to adult care typically was to an ACHD care setting
• Barriers to care transition included concern with replacing a strong relationship with pediatric providers, concern with obtaining insurance coverage into adulthood, and inability to locate an appropriate adult provider.

Paper Topics/Projects in Progress

• Racial and socioeconomic differences in healthcare outcomes
• Comorbidities
• ICD-9-CM CHD code validation
• Pregnancy outcomes
• Healthcare utilization

Individuals aged 1-64 years with documented congenital heart defects at healthcare encounters, five U.S. surveillance sites, 2011-2013 (Glidewell et al. American Heart Journal, 2021. May 2;S0002-8703(21):00109-5.)

Percentage of CHD cases decreased with age until 20 to 44 years, and for non-severe CHD, the percentage increased slightly for those 45 to 64 years. These findings will inform planning for the needs of this growing population.

Health Care Transition Perceptions among Parents of Adolescents with Congenital Heart Defects in Georgia and New York (Gaydos et al. Pediatric Cardiology. 2020.41:1220–1230).

These findings suggest that there are significant concerns about the realities of the transition process. Future investment in adolescent transition should include educational interventions and training of maternal and child healthcare professionals who can assist with allaying fears about transition and ensuring that barriers to care remain low.

Hypertension and Heart Failure as Predictors of Mortality in an Adult Congenital Heart Defect Population. (Raskind-Hood et al. 2021. Congenital Heart Disease. 16(4):333–355).

To evaluate the impact of hypertension (HTN) and heart failure (HF) on mortality in adults with CHD, a retrospective cross-sectional analysis of 5,397 adults with CHD (ACHD), aged 18-64, was conducted. 18.3% of the sample had HTN without HF, 5.6% had HF without HTN, 12.4% had both. While overall mortality was 3.5%, older age and black race were associated with increased mortality during the three-year study period compared to those of younger age and white race. Also, valve lesion patients with HF, with or without HTN, were at increased risk of death. Findings suggest that vigilance for signs and symptoms of HF in adults with all heart defects, and treatment to mitigate comorbid HF and HTN may lessen mortality in ACHD regardless of the anatomic complexity of the heart defect, emphasizing the need for all ACHD to stay in specialty care.

Funding/Support: Funding/Support received by Centers for Disease Control and Prevention, Grant/Award Number: CDC‐RFA‐DD19-1902A and 1902B

Component A (in progress)

Surveillance period
2010 - 2019

Collaborators (CDC and 7 sites: AZ, GA, IA, NC, NY, SC & UT)
Centers for Disease Control and Prevention (CDC): National Center on Birth Defects and Developmental Disabilities (NCBDDD), Atlanta, GA

1. Emory University School of Medicine and Rollins School of Public Health, Atlanta, GA collaborates with:
   1. Data Sources:
      • Children’s Healthcare of Atlanta (CHOA) – pediatric [case finding & encounters]
      • CHOA Society of Thoracic Surgeons (STS) – pediatric [gold standard case finding source]
      • Emory Healthcare (EHC) – adult
      • EHC Society of Thoracic Surgeons – adult [gold standard case finding source]
      • Emory ‘Archival’ CHD databases – adult
      • Centers for Medicare and Medicaid Services (CMS) (via Research Data Assistance Center (ResDAC) and/or GA Dept Community Health (GADCH) - pediatric and adult
      • Metropolitan Atlanta Congenital Defects Program (MACDP) - pediatric
      • Piedmont Healthcare -pediatric and adult
      • Wellstar Health System – adult
      • Northside Hospital - adult
      • GA Dept of Public Health (GADPH)
        • Vital records (birth and death certificates) – pediatric and adult
        • Critical Congenital Heart Disease (CCHD) Newborn Screening - pediatric
   2. Community Advisory Committee:
      • GA Dept of Public Health (GADPH)
      • GA Dept of Community Health (GADCH)
      • Children’s Healthcare of Atlanta (CHOA) & Sibley Heart Center
      • The Children's Heart Foundation
      • Georgia Center for Learning Disabilities – GA State University
      • Adult Congenital Hearts of Greater Atlanta (patient advocacy)
      • Grady Health System
      • Wellstar Health System
      • Piedmont Healthcare
      • CHD Healthcare providers
      • Parents of teens with CHD
2. University of Arizona, Department of Pediatrics, College of Medicine, Tucson, AZ
3. Duke University School of Medicine, Duke Clinical Research Institute, Durham, NC
4. University of Iowa, College of Public Health, Department of Epidemiology, Iowa City, IA
5. New York State Department of Health, Albany, NY
6. South Carolina Department of Health and Environmental Control, Columbia, SC
7. University of Utah, Salt Lake City, UT

Priorities

• Assess and conduct a population-based surveillance of individuals with CHD, ages 1-45 years
• Examine descriptive epidemiology of CHD, survival, healthcare utilization, and comorbidities over time
• Improve understanding of health outcomes of CHD among public health practitioners
• Estimate age-specific mortality

Improved treatment of congenital heart defects (CHD) in the recent decades has increased survival; however, little data exist on healthcare utilization, comorbidities, long-term health and non-health outcomes and mortality to inform quality of life and develop effective secondary prevention strategies to address health complications among those affected. Through expansion of an integrated surveillance system for CHD, using individual encounter level data for 2010-2019, we aim to:

1. Improve understanding of age-specific mortality, healthcare utilization, comorbidities, survival and other outcomes over time;
2. Improve understanding of racial/ethnic and socioeconomic patterns in healthcare usage, and their impact on long-term outcomes over time;
3. Gain a greater understanding of the strengths and limitations of databases used for CHD surveillance; and
4. Increase CHD awareness among the public and stakeholders

Our source population is the state of Georgia based on 2010 Census [9.7 million (30.5% black, 8.8% Hispanic)], and the 2017 American Community Survey (ACS) estimate [10.2 million (31.3% black, 9.3% Hispanic)].

• a case born between 1/1/1965 - 12/31/2019 with at least one of the included CHD codes found in an approved data source or cardiac encounter type, at any time, and with at least one encounter between 2010-2019, with or without an associated CHD code, and with residence in the state of Georgia at some point in the decade 2010-2019.

Regional and community partners enhance our surveillance system, guide education efforts, and ensure data quality and consistency. Through our previous work, we have already demonstrated success in conducting CHD surveillance in the proposed region, and have developed processes to assess data quality and performance measures applicable to the proposed project. Through this project, we will gain a greater understanding of the strengths and limitations of integrated CHD surveillance, increase awareness among stakeholders, and improve ability to estimate life expectancies. Achieving these aims should assist affected individuals and families in their expectations and healthcare decisions, improve decision-making among stakeholders, and ultimately lead to more effective secondary prevention strategies to reduce the public health impact of CHDs.

Component B (in progress)

Surveillance period
2010 - 2019

Collaborators (CDC and Emory)

1. Emory University School of Medicine and Rollins School of Public Health, Atlanta, GA collaborates with:
2. Emory University School of Medicine, Dept of Biomedical Informatics, Atlanta, GA whose team includes: Gari Clifford, Rishi Kamaleswaran, Chad Robichaux, Abeed Sarker, Reza Sameni, Qiao Li and Mary Bell (GA Tech student)

Priorities

• To validate billing codes for CHDs in healthcare claims data
• Determine best ways to use healthcare claims for CHD surveillance across the lifespan

Building on existing CHD surveillance infrastructure and using individual encounter level integrated electronic health record (eHR) and administrative data for 2010-2019, and through validation of healthcare claims data in an integrated surveillance system for CHDs, this project aims to:

1. Improve understanding of the validity and utility of healthcare claims data for surveillance of CHDs;
2. Improve the accuracy of identification and surveillance of CHDs; and
3. increase awareness among the public and stakeholders.

Achieving these aims should improve decision-making among stakeholders, and ultimately lead to more effective secondary prevention strategies to reduce the public health impact of CHDs. Improved accuracy in CHD surveillance will lead to improved understanding of healthcare outcomes over time for patients with CHD; racial/ethnic and socioeconomic patterns in healthcare usage, and their impact on outcomes over time; and age-specific mortality.

A minimum of 1500 cases identified by CHD ICD codes have already been validated via medical record review. A machine learning (ML) approach will be utilized to determine PPV of CHD ICD codes by CHD anatomic group, age, CHD type, and individual and healthcare characteristics. From these results, further development of algorithms to maximize PPV and sensitivity will be proposed. ML techniques developed together with colleagues in the Emory Biomedical Informatics Dept. will be tested to optimize algorithms for accurate detection of CHDs in healthcare claims data. In addition, algorithms will be tested in additional data sources such as GA Medicaid and Kaiser Permanente. In addition, we are exploring the role of natural language processing (NLP) for identifying congenital heart conditions that cannot be accurately characterized by claims data. Through this project, we will gain a greater understanding of the strengths and limitations of integrated CHD surveillance, increase awareness among stakeholders, and improve ability to calculate life expectancies. Regional and community partners enhance our surveillance system, guide education efforts, and ensure data quality and consistency.