Association of Unobserved Heterogeneity & Observed Factors with Survival Time of Pediatric Cardiac Patients: A Frailty Model Analysis

Abstract

Introduction: Cardiac disease is a major cause of mortality and disability worldwide and represents an important contributor to morbidity among children. Survival outcomes in pediatric cardiac disease may vary substantially between individuals because of differences in clinical characteristics, nutritional status, and other factors that may not be directly measurable. Accounting for such unobserved heterogeneity is therefore essential for accurately identifying determinants of survival.

Aim & Objective: The present study aimed to assess the association of observed clinical and nutritional factors with survival among children with cardiac disease and to determine the contribution of unobserved heterogeneity to survival outcomes.

Material & Methods: A retrospective cohort study was conducted among 323 children under follow-up at the Cardiac Center–Ethiopia. Clinical and demographic information was analyzed to identify factors associated with mortality. Shared frailty survival models were employed to account for unobserved individual-level heterogeneity. Model parameters were estimated using the Expectation-Maximization (EM) algorithm, and alternative frailty distributions were compared to determine the model providing the best fit.

Results: The gamma frailty model demonstrated the best fit for the data. The estimated gamma frailty variance was 3.817 (95% CI: 1.024–6.620) and was significantly different from zero, indicating substantial unobserved heterogeneity in survival among children with cardiac disease. In the gamma frailty model, congenital cardiac disease was significantly associated with increased mortality risk (HR = 1.804, p = 0.033). Similarly, pulmonary hypertension (HR = 1.941, p = 0.019) and anemia (HR = 1.509, p = 0.025) were significant predictors of increased mortality risk.

Conclusions: Survival among children with cardiac disease is influenced by important clinical and nutritional factors. The significant frailty variance further indicates substantial individual-level unobserved heterogeneity, highlighting the importance of incorporating frailty models into pediatric survival analyses.

REFERENCES

1. Shaddy RE, et al. Systematic literature review on the incidence and prevalence of heart failure in children and adolescents. Pediatr Cardiol. 2018;39:415-436.
2. Madriago E, Silberbach M. Heart failure in infants and children. Pediatr Rev. 2010;31(1):4-12.
3. Kaptoge S, et al. World Health Organization cardiovascular disease risk charts: revised models to estimate risk in 21 global regions. Lancet Glob Health. 2019;7(10):1332–1345.
4. Zhao Q, Ma X, Jia B, Huang G. Prevalence of congenital heart disease at live birth: an accurate assessment by echocardiographic screening. Acta Paediatr. 2013;102(4):397-402. https://doi.org/10.1111/apa.12170
5. Keates AK, Mocumbi AO, Ntsekhe M, Sliwa K, Stewart S. Cardiovascular disease in Africa: epidemiological profile and challenges. Nat Rev Cardiol. 2017;14(5):273-293. https://doi.org/10.1038/nrcardio.2017.19
6. Nowbar AN, Gitto M, Howard JP, Francis DP, Al-Lamee R. Mortality from ischemic heart disease: analysis of data from the World Health Organization and coronary artery disease risk factors from NCD Risk Factor Collaboration. Circ Cardiovasc Qual Outcomes. 2019;12(6):1-11. https://doi.org/10.1161/circoutcomes.118.005375
7. Nigussie B, Tadele H. Heart failure in Ethiopian children: mirroring the unmet cardiac services. Ethiop J Health Sci. 2019;29(1):811–818. https://doi.org/10.4314/ejhs.v29i1.2
8. Gebremariam S, Moges T. Pediatric heart failure, lagging, and sagging of care in low income settings: a hospital based review of cases in Ethiopia. Cardiol Res Pract. 2016;2016:1-7. https://doi.org/10.1155/2016/7147234
9. World Health Organization. Health in 2015: from MDGs, millennium development goals to SDGs, sustainable development goals. Geneva: World Health Organization; 2015:1-204. https://doi.org/10.2471/blt.15.165027
10. Mulatu HA, Woldemichael M, Aberra M. Prevalence of rheumatic heart disease among primary school students in Mid-Eastern Ethiopia. Biol Syst Open Access. 2015;5:1-4. https://doi.org/10.4172/2329-6577.1000149
11. Tesfaye S, Shifeta M, Hirigo AT. Pattern of cardiac diseases and co-existing morbidities among newly registered cardiac patients in an adult cardiac referral clinic of Hawassa University Comprehensive Specialized Hospital, Southern-Ethiopia. Vasc Health Risk Manag. 2020:379-387. https://doi.org/10.2147/vhrm.s266582
12. Hurisa T, Megersa H, Tsegaye T. Delayed in diagnosis of congenital heart disease and associated factors among pediatric patients in Cardiac Center Addis Ababa, Ethiopia, 2021/2022 GC. Am J Health Res. 2022;7(1):51-62 .https://doi.org/10.21203/rs.3.rs-778324/v1
13. Opotowsky AR, et al. Pediatric and congenital cardiovascular disease research challenges and opportunities: JACC Review Topic of the Week. J Am Coll Cardiol. 2022;80(23):2239-2250.
14. Ehwerhemuepha L, et al. Association of congenital and acquired cardiovascular conditions with COVID-19 severity among pediatric patients in the US. JAMA Netw Open. 2022;5(5):1-10. https://doi.org/10.1001/jamanetworkopen.2022.11967
15. Cox DR. Regression models and life-tables. J R Stat Soc Series B Stat Methodol. 1972;34(2):187-202.
16. Hougaard P. Analysis of multivariate survival data. New York: Springer; 2000:1-542.
17. Gutierrez RG. Parametric frailty and shared frailty survival models. Stata J. 2002;2(1):22-44. https://doi.org/10.1177/1536867x0200200102
18. Woodward M. Epidemiology: study design and data analysis. 3rd ed. Boca Raton: CRC Press; 2013:1-854.
19. Cochran WG. Sampling techniques. 3rd ed. New York: John Wiley & Sons; 1977:1-428.
20. Riley RD, et al. Minimum sample size for developing a multivariable prediction model: Part II—binary and time-to-event outcomes. Stat Med. 2019;38(7):1276-1296. https://doi.org/10.1002/sim.8409
21. Chen LM, Ibrahim JG, Chu H. Sample size and power determination in joint modeling of longitudinal and survival data. Stat Med. 2011;30(18):2295-2309. https://doi.org/10.1002/sim.4263
22. Duchateau L. The frailty model. New York: Springer-Verlag; 2008.
23. Balan TA, Putter H. A tutorial on frailty models. Stat Methods Med Res. 2020;29(11):3424-3454. https://doi.org/10.1177/0962280220921889
24. Moon TK. The expectation-maximization algorithm. IEEE Signal Process Mag. 1996;13(6):47-60. https://doi.org/10.1109/79.543975
25. Schwarz G. Estimating the dimension of a model. Ann Stat. 1978;6(2):461-464.
26. Shamliyan T, Talley KMC, Ramakrishnan R, Kane RL. Association of frailty with survival: a systematic literature review. Ageing Res Rev. 2013;12(2):719-736. https://doi.org/10.1016/j.arr.2012.03.001
27. Phipson B. Analysis of time-to-event data including frailty modeling. 2006:1-147.
28. Tadege M, Tegegne AS, Dessie ZG. Cardiac patients' surgery outcome and associated factors in Ethiopia: application of machine learning. BMC Pediatr. 2024;24(1):1-11. https://doi.org/10.1186/s12887-024-04870-4
29. Lang'o MO, Gitchan'a JN, Yuko-Jowi CA. Prevalence of iron deficiency in children with cyanotic heart disease seen at Kenyatta National Hospital and Mater Hospital, Nairobi. East Afr Med J. 2009;86(12):47-51. https://doi.org/10.4314/eamj.v86i12.62901
30. Abdulkadir MB, Ibraheem RM, Issa A, Suberu HD, Owolabi BI, Alabi K. Prevalence, etiology, and fatality rates of childhood heart failure in Nigeria: a systematic review and meta-analysis. Int J Health Sci (Qassim). 2023;17(1):53-60.
31. Tsega T, Tesfaye T, Dessie A, Teshome T. Nutritional assessment and associated factors in children with congenital heart disease-Ethiopia. PLoS One. 2022;17(9):1-12. https://doi.org/10.1371/journal.pone.0269518