Original Article

A Decision Support Model for the Necessity of Cardio-Angiography, A Result of Generalized Additive and logistic Regression Model


Introduction: Angiography is used as the gold standard for diagnosis of coronary artery disease (CAD). It is an invasive procedure and has several complications. Also, some patients refuse angiograms for reasons such as fear, high cost, and loss of trust in physician diagnosis. The negative results of this test is more than a third. Therefore, having a statistical predictive model for estimating the risk of CAD, as an evidence-based support system, can help the physician and patient decide on the necessity of angiography. 

Aims: In this study we aimed to find an evidence-based supportive model for decision making on the necessity of angiography in people who were candidates for angiography by the physician after initial tests. 

Methods: In this study, 1187 patients who had been referred to Ghaem Hospital of Mashhad University of Medical Sciences and diagnosed with physicians after initial tests were enrolled. Demographic data, lipid and blood sugar levels, and the history of underlying disorders were variables that were studied in the statistical model fitting. Initially, generalized additive models were used singularly for quantitative predictors, then by applying right transformations on the predictor variables we entered them simultaneously in logistic and count regression models. These two models were fitted to the data using R software and then compared in terms of predictive accuracy. 

Findings: Generalized additive models showed that the relationship between CAD with the hs-CRP level was not monotone. Exploratory analyzes showed that 62% of people with hs-CRP level <3 and 50% of people with hs-CRP levels between 3 and 6 were suffered from the CAD. The highest rate of CAD was seen in the range of 6-8 (93%) but with increasing the hs-CRP level to above 8 it decreased to 70%. The relationship between age and the risk of CAD was S-shaped. Risk of CAD in diabetic subjects with normal FBS was equal to that of nondiabetic subjects with normal fasting blood sugar. The age, gender, diabetes, FBS, and hs-CRP were significant in both models (p <0.05). The area under the ROC curve was upgraded to 81 for the logistic model. 

Conclusion: The most important finding of this exploratory study was that out of 232 patients with hs-CRP level between 6 to 8, 217 (93%) had coronary artery occlusion, for these subjects the probability of occluding a coronary artery was 0.93. The present study also showed that if the blood sugar is controlled in patients with diabetes, then this disease will not be a risk factor for patients with coronary artery occlusion. The logistic regression model presented in this study is recommended as the final model to support decision-making about the necessity of angiography.

1. Concannon TW, Nelson J, Goetz J, Griffith JL. A percutaneous coronary intervention lab in every hospital? Circulation: Cardiovascular Quality and Outcomes. 2012;5(1):14-20.
2. Bijani M, Valizadeh A, Sayari A, Samizadeh B. Surveying the Reasons for Refusing Coronary Angiography in Patients Referring to Cardiac Ward of Valiasr Hospital in Fasa. Journal of Fasa University of Medical Sciences. 2015;4(4):375-81.
3. Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden WB, et al. Heart disease and stroke statistics—2012 update. Circulation. 2012;125(1):e2-e220.
4. Ebrahimi M, Kazemi-Bajestani S, Ghayour-Mobarhan M, Ferns G. The status of coronary artery disease and its risk factors in Iran: A Review. Iranian Red Crescent Medical Journal. 2011;13(9):615-26.
5. Bidel Z, Nazarzadeh M. Evaluating the Associations between Some Cardiovascular Risk Factors and Coronary Artery Obstruction Based on Angiographic Findings. Scientific Journal of Ilam University of Medical Sciences. 2014;22(1):147-54.
6. Mahmoudi I, Moazzam M, Sadeghian S. Prediction model for coronary artery disease using neural networks and feature selection based on classification and regression tree. Journal of Shahrekord University of Medical Sciences. 2013;15(5):47-56.
7. Sedehi M, Mehrabi Y, Kazemnejad A, Hadaegh F. Comparison of artificial neural network, logistic regression and discriminant analysis methods in prediction of metabolic syndrome. Iranian journal of endocrinology and metabolism. 2010;11(6):638-731.
8. Agresti A. An Introduction to Categorical Data Analysis: Wiley; 2019 2019.
9. Azadeh Saki MRE, Hamed Tabesh. Patterns of Daily Duration and Frequency of Breastfeeding among Exclusively Breastfed Infants in Shiraz, Iran, A 6-Month Follow-up Study Using Bayesian Generalized Linear Mixed Models Global Journal of Health Science. 2013;5(2):123-33.
10. Azadeh Saki MRE, Kazem Mohammad, Abbas Rahimi Foroushani, Mohammad R Bordbar. A prospective study of the effect of delivery type on neonatal weight gain pattern in exclusively breastfed neonates born in Shiraz, Iran. International Breastfeeding Journal. 2010;5(1).
11. Birjandi M, Salehi-Marzijarani M, Ayatollahi SMT, Rashidi H, Hosseinzadeh A. Comparison of Several Count Regression Models on Modeling Decayed Missed Filled Teeth Dental Index in Dentistry. Journal of Sabzevar University of Medical Sciences. 2016;23(3):468-77.
12. Hilbe JM. Problems with zeros. Modeling count data. 2014:172-209.
13. Chiha J, Mitchell P, Gopinath B, Plant AJ, Kovoor P, Thiagalingam A. Gender differences in the severity and extent of coronary artery disease. IJC Heart & Vasculature. 2015;8:161-6.
14. Rott D, Benhorin J, Keren A, Baras M, Leibowitz D, Banai S. Prediction of coronary artery disease severity in patients referred for coronary angiography. Clinical cardiology. 2005;28(6):288-92.
15. Hosseini SA, Abdollahi AA, Behnampour N, Salehi A. Relationship Between Number Of Involved Coronary Artery With Some Risk Factors By Angiography. Journal of Payavard Salamat. 2013;6(5):383-91.
16. Mousavinasab N, Yazdani Cherat J, Bagheri B, Bakhti F-s, Bakhti Z. Identifying the risk factors for cardiovascular disease in individuals aged above 35 years using logistic regression model. Journal of Mazandaran University of Medical Sciences. 2017;26(144):50-6.
17. Assadpour Piranfar M, Biranvand MR, Fartoukzadeh S, Velayi N. The Correlation Between hsCRP Serum Levels With The Severity of Coronary Artery Aisease. Research in Medicine. 2012;36(3):139-42.
18. Azarkar Z, Jafarnejad M, Ebrahimzadeh A, Rezvani M, Birjand I. Association of quantitative C-reactive protein with acute myocardial infarction. J Qazvin Univ Med Sci. 2014;4:4-8.
19. Hatami S, Mousavi J, Ali S. C-reactive protein levels correlate with mortality in critically ill patients. Razi Journal of Medical Sciences. 2015;22(130):47-52.
20. Neshati Tanha A, Soleimani P. Prediction of Acute Heart Attack using Logistic Regression (Case Study: A Hospital in Iran). Advances in Industrial Engineering. 2016;50(1):109-19.
21. Sedghi Sabet M, Benvan G, Baghaie M, Atrkar-Roshan Z, Khalili M. Vascular Complications and its Related Factors after Coronary Angiography. Iran Journal of Nursing. 2015;27(92):13-22.
22. Guddattu V, Rao A, Rajkannan T. Comparison between count regression and binary logistic regression models in the analysis of adverse drug reaction data. ProbStat Forum. 2015;8:140-7.
IssueVol 7 No 2 (2021) QRcode
SectionOriginal Article(s)
DOI https://doi.org/10.18502/jbe.v7i2.6712
logistic regression Generalized additive model Decision support system Angiography hs-CRP

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
Rezaei Sharif F, Saki A, Taghipour A, Tajfard M. A Decision Support Model for the Necessity of Cardio-Angiography, A Result of Generalized Additive and logistic Regression Model. jbe. 2021;7(2):131-142.