Virtual StatsPD@Waite meeting

Every month, the professional development meetings of statisticians and data scientists at Waite, known as StatsPD@Waite, bring together specialists in various aspects of data sciences in agriculture from Waite, Roseworthy and Adelaide.

Please join us for the next virtual StatsPD@Waite meeting where James Brown from Adelaide Medical School, Department of Women’s and Children’s Health will present his Biostatistics Masters thesis.

Please note that the StatsPD@Waite meetings are recorded. If you have a question to the speaker but would rather not be recorded, please send me your question via chat during the meeting and I will ask it on your behalf. 

Please email Beata Sznajder for details of the Zoom meeting.


Factors affecting survival and readmission rate among heart failure patients 

James Brown (Adelaide Medical School, Department of Women’s and Children’s Health)

From a cohort of patients at the RAH diagnosed with either heart failure with preserved ejection fraction (HFpEF) or heart failure with reduced (HFrEF) ejection fraction, the Cox proportional hazards model was used to determine which patient characteristics and medications were associated with survival time.  Extensions of the Cox model to multiple events including the Andersen-Gill model, the Prentice-Williams Peterson gap time model and shared frailty models were then used to determine the factors that were associated with an increased rate of readmissions to hospital. Multiple imputation by chained equations was explored as a way to flexibly impute missing data as there were a number of missing observations in the data. Restricted cubic splines were used to allow non-linear relationships to be fit, whilst avoiding the potential pitfalls that are associated with unnecessary categorization of continuous variables. We found that the commonly used medications, known as Beta Blockers and ACE Inhibitors, exhibited a dose dependant protective effect on patients with HFrEF, whilst none of the medications examined had any effect on survival or readmission rate for patients with HFpEF. Additionally, an interesting non-linear relationship was found between blood pressure and survival for both HFpEF and HFrEF patients, where lowered blood pressure up to a point was associated with longer survival time, but when lowered too much, was associated with lower survival times. For patients with HFpEF, which to date has had no medication approved for use in treating it, we found that a strong predictor of both mortality and readmission to hospital was burden of comorbid disease, measured using the Charlson Comorbidity index. These results suggest that, in the absence of a proven therapy, management of these patients with HFpEF should focus on treating the comorbidities which frequently accompany the disease. 

Tagged in Stats PD, Waite, Statistics, Biometry Hub