Faculty of Health Sciences (ETDs)
Permanent URI for this communityhttps://hdl.handle.net/10539/37925
Browse
4 results
Search Results
Item Predicting in-hospital mortality in heart failure patients using machine learning(University of the Witwatersrand, Johannesburg, 2023-05) Mpanya, Dineo; Ntsinjana, HopewellThe age of onset and causes of heart failure differ between high-income and low-and-middle-income countries (LMIC). Heart failure patients in LMIC also experience a higher mortality rate. Innovative ways that can risk stratify heart failure patients in this region are needed. The aim of this study was to demonstrate the utility of machine learning in predicting all-cause mortality in heart failure patients hospitalised in a tertiary academic centre. Six supervised machine learning algorithms were trained to predict in-hospital all-cause mortality using data from 500 consecutive heart failure patients with a left ventricular ejection fraction (LVEF) less than 50%. The mean age was 55.2 ± 16.8 years. There were 271 (54.2%) males, and the mean LVEF was 29 ± 9.2%. The median duration of hospitalisation was 7 days (interquartile range: 4–11), and it did not differ between patients discharged alive and those who died. After a prediction window of 4 years (interquartile range: 2–6), 84 (16.8%) patients died before discharge from the hospital. The area under the receiver operating characteristic curve was 0.82, 0.78, 0.77, 0.76, 0.75, and 0.62 for random forest, logistic regression, support vector machines (SVM), extreme gradient boosting, multilayer perceptron (MLP), and decision trees, and the accuracy during the test phase was 88, 87, 86, 82, 78, and 76% for random forest, MLP, SVM, extreme gradient boosting, decision trees, and logistic regression. The support vector machines were the best performing algorithm, and furosemide, beta-blockers, spironolactone, early diastolic murmur, and a parasternal heave had a positive coefficient with the target feature, whereas coronary artery disease, potassium, oedema grade, ischaemic cardiomyopathy, and right bundle branch block on electrocardiogram had negative coefficients. Despite a small sample size, supervised machine learning algorithms successfully predicted all-cause mortality with modest accuracy. The SVM model will be externally validated using data from multiple cardiology centres in South Africa before developing a uniquely African risk prediction tool that can potentially transform heart failure management through precision medicine.Item HbA1c Control in Type 2 Diabetic Patients with Coronary Artery Disease(University of the Witwatersrand, Johannesburg, 2023-10) Mhlaba, Lona; Tsabedze, Nqoba; Mpanya, DineoBackground: Type 2 diabetes mellitus (T2DM) patients with coronary artery disease (CAD) have an increased risk of recurrent cardiovascular events. These patients require optimal glucose control to prevent the progression of atherosclerotic cardiovascular disease (ASCVD). Current guideline recommendations target an HbA1c ≤7% to mitigate this risk. This study evaluated the level of HbA1c control in T2DM patients with CAD. Methods: This retrospective study assessed consecutive patients who presented with CAD to the Charlotte Maxeke Johannesburg Academic Hospital (CMJAH) between April 2017 and December 2019. The study included T2DM patients on anti-diabetic medication with angiographically confirmed CAD. HbA1c control was assessed using the HbA1c level measured at the index presentation and during the most recent follow-up visit. Results: The study population comprised 262 T2DM patients with a mean age was 61.3 ±10.4 years. Among the T2DM patients, 188 (71.8%) were males. At index presentation, 110 (42.1%) T2DM patients presented with ST-segment elevation myocardial infarction, 69 (26.4%) had non-ST-segment elevation myocardial infarction, 43 (16.5%) had unstable angina, and 39 (14.9%) had stable angina. The baseline median systolic blood pressure was higher in patients with an HbA1c ≤7% [136 mmHg (Interquartile range (IQR): 117-151) vs 124 mmHg (IQR: 112-142), p= 0.0121], compared to those with an HbA1c level above 7%. Furthermore, T2DM with an HbA1c ≤7% also had a higher median diastolic blood pressure [85 mmHg (IQR: 75.5-97) vs 78 mmHg (IQR: 71-88), p=0.0205]. After a median follow-up of 16.5 months (IQR: 7-29), 28.7% of the study participants had an HbA1c ≤7%. On multivariable regression analysis, patients with ST-segment depression on the resting electrocardiogram and index presentation had optimal glycaemic control (Odds ratio: 0.27, CI: 0.12-0.59, p= 0.001). Conclusion: After a median follow-up duration of 16.5 months, only 28.7% of T2DM patients with CAD had optimal glycaemic control. This finding underscores the substantial unmet need for optimal diabetes control in this very high-risk group.Item Attitudes and perceptions of caregivers regarding their presence at induction of anaesthesia(University of the Witwatersrand, Johannesburg, 2021) Le Roux, Johannes Jacobus; Redelinghuys, CaraBackground Caregiver presence at their children’s induction of anaesthesia is practiced daily around the world. International studies demonstrated conflicting emotions in caregivers present at induction of anesthesia of their children. These positive and negative emotions ranged from comforting and reassuring, to traumatising and disturbing. Research exploring the attitudes and perceptions of caregivers regarding this practice is limited within the African context. Aims The aim of this study was to describe caregivers’ attitudes and perceptions regarding their presence at induction of their children’s anaesthesia. Methods This descriptive, phenomenological, qualitative study was conducted in 2020 at Chris Hani Baragwanath Academic Hospital, a 3200-bed facility in South Africa. Twenty caregivers of children (aged 2 to 8 years) undergoing elective surgery were recruited. Data was collected through face-to-face, in-depth, semi-structured individual interviews using purposive sampling. Interviews ranged between 11 and 55 minutes in duration and were conducted within 24 hours of induction of anaesthesia. The audio recorded interviews were transcribed and subjected to inductive reflexive thematic analysis. Results Six themes were developed: Fulfilment of caregiver role, A positive experience, A traumatic experience, Not prepared for the experience, My world is my reality, and Your world is a place different to mine. Conclusion A caregiver’s perception of the induction process is influenced by multiple factors. A finding specific to our cohort is the interplay between complex multifaceted cultural beliefs and anaesthesia of their children. By acknowledging and addressing these beliefs, a caregiver’s presence can be tailored to ensure a positive experience for all involved at inductionItem Comorbidities in a cohort of privately insured South Africans with systemic lupus erythematosus(2024) Ntumba, Mbombo Henriette NganduBackground: Comorbidities in systemic lupus erythematosus (SLE) impact negatively health related quality of life and life expectancy. We undertook a retrospective study of the burden of comorbidities in privately insured South Africans with SLE. Methods: Data review of patients insured with Discovery Health Medical Scheme (DHMS), ≥16years at diagnosis, ≥6months follow-up and diagnosed with SLE based on ICD 10 codes. Demographics, drug therapy and comorbidities listed in the Charlson Comorbidity Index (CCI) and other comorbidities occurring commonly in SLE patients were documented. Results: Of 520 patients with SLE ICD 10 codes, only 207 met the other inclusion/exclusion criteria for data analysis. Most were women (90.8%), median (IQR) age and follow-up duration of 39 (30.3-53.0) and 6.1 (3.7-8.1) years, respectively. All patients had at least one comorbidity, the most frequent CCI comorbidities being pulmonary disease (30.9%), congestive heart failure (CHF) (15%) and renal disease (14.5%). Common CCI comorbidities were hypertension (53.1%), mood and anxiety disorders (46.9%), infections (urinary tract infections (UTI) (37.7%) and pneumonia (33.8%)). Independent predictors of 1) CHF were renal disease (OR=855), dyslipidaemia (OR=15.3) and male gender (OR=43.0); 2) hypertension were age at diagnosis (OR=1.03), type 2 diabetes (OR=4.45) and renal disease (OR=4.34); and 3) mood and anxiety disorders were female gender (OR=3.98), cerebrovascular accident (OR=3.18), UTI (OR=2.39) and chloroquine use (OR=1.94). Conclusion: Comorbidities in this cohort of privately insured South Africans with SLE were common, with all patients having at least one comorbidity. Hypertension, infections and mood and anxiety disorders were the leading comorbidities.