Advanced search
Publication Cover
Degenerative Neurological and Neuromuscular Disease Volume 13, 2023 - Issue
Submit an article Journal homepage
257
Views
0
CrossRef citations to date
0
Altmetric
ORIGINAL RESEARCH

Development and Validation of a Risk Prediction Model to Estimate the Risk of Stroke Among Hypertensive Patients in University of Gondar Comprehensive Specialized Hospital, Gondar, 2012 to 2022

Yazachew Moges Chekol1 Department of Health Information Technician, Mizan Aman College of Health Science, Mizan Aman, EthiopiaCorrespondence[email protected]
ORCID Iconhttps://orcid.org/0000-0002-2616-3823
ORCID Icon,
Mehari Woldemariam Merid2 Department of Epidemiology and Biostatistics, University of Gondar, Gondar, Ethiopia
,
Getayeneh Antehunegn Tesema2 Department of Epidemiology and Biostatistics, University of Gondar, Gondar, EthiopiaORCID Iconhttps://orcid.org/0000-0001-6812-1659
ORCID Icon,
Tigabu Kidie Tesfie3 Department of Public Health, Debre Markos University, Debre Markos, Ethiopia
,
Tsion Mulat Tebeje4 Department of Epidemiology and Biostatistics, Dilla University, Dilla, EthiopiaORCID Iconhttps://orcid.org/0000-0002-2911-2095
ORCID Icon,
Negalegn Byadgie Gelaw5 Department of Public Health, Mizan Aman College of Health Science, Mizan Aman, Ethiopia
,
Nebiyu Bekele Gebi6 Department of Medicine, University of Gondar, Gondar, EthiopiaORCID Iconhttps://orcid.org/0000-0002-7113-3324
ORCID Icon &
Wullo Sisay Seretew2 Department of Epidemiology and Biostatistics, University of Gondar, Gondar, Ethiopia
 show all
Pages 89-110 | Received 13 Sep 2023, Accepted 07 Dec 2023, Published online: 13 Dec 2023

Figures & data

Figure 1 Location of study area (Gondar town), 2022.

Figure 1 Location of study area (Gondar town), 2022.

Figure 2 Flowchart of participant’s selection to estimate the risk of stroke among hypertensive patients in UOGCSH, 2012 to 2022.

Abbreviation: UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 2 Flowchart of participant’s selection to estimate the risk of stroke among hypertensive patients in UOGCSH, 2012 to 2022.

Table 1 Baseline Socio-Demographic and Behavioral Characteristics of Hypertensive Patients at UoGCSH from 2012 to 2022 (n=743)

Table 2 Baseline Clinical Characteristics of Hypertensive Patients at UoGCSH from 2012 to 2022 (n=743)

Table 3 Multivariable Logistic Regression Coefficients of Each Predictor Included in the Final Reduced Model to Predict Stroke Among Hypertensive Patients (n = 743)

Figure 3 The ROC curve represents the probability of risk for stroke among hypertensive patients at UoGCSH, 2012–2022.

Abbreviations: Un_HTN, Uncontrolled Hypertension; B_DBP, Baseline Diastolic Blood Pressure, DM, Diabetes Mellitus; ROC, Receiver Operator Characteristics.
Figure 3 The ROC curve represents the probability of risk for stroke among hypertensive patients at UoGCSH, 2012–2022.

Figure 4 Calibration plot for developed model based on original beta coefficient for stroke prediction model among hypertensive patients at UoGCSH, 2012–2022.

Notes: 95% CI calibration belts are plotted in dark gray. The black diagonal line is the reference line indicating perfect calibration.
Abbreviation: UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 4 Calibration plot for developed model based on original beta coefficient for stroke prediction model among hypertensive patients at UoGCSH, 2012–2022.

Figure 5 The ROC curve represents the probability of risk of stroke among hypertensive patients after internal validation at UoGCSH, 2012–2022.

Abbreviations: BV, Before Validation; AV, After Validation; UOGCSH, University of Gondar Comprehensive Specialized Hospital; ROC, Receiver Operator Characteristics.
Figure 5 The ROC curve represents the probability of risk of stroke among hypertensive patients after internal validation at UoGCSH, 2012–2022.

Figure 6 Calibration plot for a risk prediction model of stroke among hypertensive patients after internal validation at UoGCSH, 2012–2022.

Abbreviation: UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 6 Calibration plot for a risk prediction model of stroke among hypertensive patients after internal validation at UoGCSH, 2012–2022.

Figure 7 Prediction density plot for developed model using original beta coefficients at UoGCSH, 2012–2022.

Abbreviation: UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 7 Prediction density plot for developed model using original beta coefficients at UoGCSH, 2012–2022.

Figure 8 Nomogram for predicting the risk of stroke in hypertensive patients at UoGCSH, 2012–2022.

Notes: Stroke risk prediction with nomogram = sex (1.52) + residence (1.79) + baseline DBP (3.27) + DM (1.47) + comorbidity (2.02) + uncontrolled HTN (2.78).
Abbreviations: UOGCSH, University of Gondar Comprehensive Specialized Hospital; HTN, Hypertension; DM, Diabetes Mellitus; DBP, Diastolic Blood Pressure.
Figure 8 Nomogram for predicting the risk of stroke in hypertensive patients at UoGCSH, 2012–2022.

Figure 9 Mobile web-based stroke risk prediction model at UoGCSH, 2012–2022.

Abbreviation: UOGCSH University of Gondar Comprehensive Specialized Hospital.
Figure 9 Mobile web-based stroke risk prediction model at UoGCSH, 2012–2022.

Table 4 Risk Classification of Stroke Among Hypertensive Patients Using a Nomogram (n=743)

Figure 10 Decision curve analysis for the nomogram plotting net benefit of the model against threshold probability at UoGCSH, 2012–2022.

Notes: The y-axis represents a standardized net benefit. The thick red solid line is a nomogram used to predict stroke risk. The thin red solid line represents the 95% CI. The black solid line represents the assumption that all patients had no stroke. The gray solid line represented the assumption that all patients had stroke.
Abbreviation: UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 10 Decision curve analysis for the nomogram plotting net benefit of the model against threshold probability at UoGCSH, 2012–2022.

Figure 11 The ANN estimated importance of final selected predictors in classifying stroke at UoGCSH, 2012–2022.

Abbreviations: ANN, Artificial Neural Network; UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 11 The ANN estimated importance of final selected predictors in classifying stroke at UoGCSH, 2012–2022.

Table 5 Overall Model Performance Evaluation Using the Brier Score for Machine Learning and Multivariable Logistic Regression Model at UoGCSH, 2012–2012

Figure 12 AUROC curves of the machine learning algorithm and multivariable logistic regression at UoGCSH, 2012–2022.

Abbreviations: LR, logistic regression; ANN, artificial neural network; DT, decision tree; RF, Random Forest; UOGCSH, University of Gondar Comprehensive Specialized Hospital.
Figure 12 AUROC curves of the machine learning algorithm and multivariable logistic regression at UoGCSH, 2012–2022.

Related research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.