This repository contains the code and analysis for a coursework project that investigates the role of Sodium-glucose Cotransporter-2 (SGLT-2) inhibitors in the secondary prevention of myocardial infarction (MI). The focus of this project is on the survival analysis of MI survivors over a five-year period, examining the effect of SGLT-2 inhibitors on reducing the risk of a secondary MI.
Myocardial infarction (MI) is a leading cause of mortality and morbidity worldwide. In the UK alone, over a million people live as MI survivors, many of whom are at significant risk of experiencing a second MI. Sodium-glucose Cotransporter-2 (SGLT-2) inhibitors, originally designed for managing blood glucose levels in diabetic patients, have shown promising results in reducing cardiovascular events, including secondary MI.
This project specifically aims to assess the efficacy of SGLT-2 inhibitors in reducing the risk of secondary MI, using survival analysis techniques such as Cox proportional hazards regression, Kaplan-Meier estimates, and time-split analyses. The main objective of this project is to determine whether SGLT-2 inhibitors can lower the incidence of secondary MI, focusing on differences in treatment effects across gender and time.
- Medication adherence: Logistic regression was used to identify factors influencing adherence to SGLT-2 inhibitors.
- Survival analysis: Cox regression models were employed to investigate the impact of SGLT-2 inhibitors on secondary MI risk, with a particular focus on time-to-event analysis.
- Blood pressure outcomes: The effect of SGLT-2 inhibitors on systolic blood pressure was evaluated through linear regression and t-tests.
- Survival Analysis: The survival analysis revealed that SGLT-2 inhibitors significantly reduced the risk of secondary MI in male participants, particularly after the first 1.5 years of treatment. The results suggest that the protective effect of SGLT-2 inhibitors becomes more pronounced over time.
- Gender Differences: While male participants showed a significant reduction in MI risk, the analysis did not find a statistically significant effect for female participants. This may be attributed to the smaller number of female participants in the study.
- Blood Pressure: Participants taking SGLT-2 inhibitors experienced a significant reduction in systolic blood pressure compared to those receiving usual care, further supporting the use of these inhibitors for comprehensive cardiovascular management.
Data were obtained from a randomised controlled trial involving 1,988 MI survivors across four regions in the UK. The trial included both a treatment group receiving SGLT-2 inhibitors and a control group receiving usual care, with follow-up data collected over five years. Data used in this project can be found in mi_sglt2.csv
The project utilised survival analysis methods to analyse time-to-event data:
- Cox Proportional Hazards Regression: To assess the effect of SGLT-2 inhibitors on the incidence of secondary MI over time, while controlling for baseline BMI and stratifying by gender.
- Kaplan-Meier Estimates: To visualise survival probabilities and compare the cumulative hazard functions between treatment groups. Time-Split Analysis: A time-split Cox regression was performed to evaluate how the effect of SGLT-2 inhibitors varied over different time periods.
Here is the R Scripts for all analysis and modelling.
This project demonstrates the potential of SGLT-2 inhibitors to reduce secondary MI risk in male survivors, particularly after long-term use. However, further research is required to confirm these benefits in female patients and to explore additional factors influencing medication adherence.
Full analysis report can be found in here.