BACKGROUND: A key action outlined within Heat-Health Action Plans is to identify at-risk individuals for targeted interventions, but this is rarely done, partially due to limited understanding of specific individual-level risk factors. Identifying those most vulnerable to heat is challenging due to the complex nature of the risks involved. Previous efforts to map population vulnerability have been of limited use for public health. In England, electronic health records have been successfully used in other health areas to develop risk prediction tools. Aim of this thesis was to assess the feasibility of developing a risk stratification tool to identify those at high risk that could be used within primary care settings. METHOD: A time-stratified case-crossover analysis using conditional logistic regression was performed on 37 clinical risk factors and 9 socio-environmental factors using data from the Clinical Practice Research Datalink. Two Random Forest (RF) models were then developed in an attempt to identify individuals at risk. RESULTS: Results indicate that heat mortality risk is significantly affected by various chronic conditions and medications. A range of socio-environmental factors further influenced risk. The RF models, however, performed poorly overall, though they identified age and circulatory diseases as the most important predictors. CONCLUSION: The study demonstrates that clinical records alone are insufficient for accurately predicting individuals at risk of death during heatwaves. The poor performance of the RF models reflects the limitations of existing tools. Despite this, the study is significant as it is the first to comprehensively explore individual-level clinical and socio-environmental factors in heat risk using primary care records in England, highlight the importance of specific conditions and medications which need to be considered in patient management, suggest heat-risk should be considered in broader health policies, and highlight the role of socioeconomic disadvantage in the unequal distribution of climate impacts.