Statistical Methods:

As one can imagine, this project relies heavily on various statistical methods to determine the outcome of patients with diabetes. A brief, simplified summary of the methods being employed for the project is as follows.

There are two parts to the Statistical Method: estimation and simulation . Estimation is the process of determining the probability of moving from one state to another in a subprocess of the disease model. Simulation is the process of using the probabilities determined in estimation in order to simulate what will happen to a single patient or sample population as they progress through the disease model.

 

Estimation:

View Estimation Movie

• Estimates (statistics) describing progression rates from one state to another are gathered from various research journals.
• These medical studies are analyzed in order to generate accurate progression rates between states.
• Sometimes there are many studies conducted for the transition between certain states, and this makes generating progression rates fairly simple because all of the necessary information is readily available.
• However, some studies may skip certain states or there may not be any studies done for some of the transitions in the model. This makes determining progression rates much more complicated.
• If a study skips an intermediary state, then the estimated transition rates must be adjusted for the multiple stages progressed in the study.
• Some studies may not be completely representative of the US population, and this makes it more challenging to find progression rates that will be accurate when simulation is done on individuals with many different backgrounds and medical histories.
• Because simulation accuracy relies on the progression rates generated during estimation, it is very important that these values are as accurate as possible.

Simulation:

View Simulation Movie

• Simulation uses the probabilities generated in estimation in order to simulate what will happen to a single individual or sample population as they progress through the disease model.
• These represent the sample data being used to predict the outcome of patients with diabetes in the US.
• Subjects are identified as having various complications due to diabetes (as seen in the pictorial model ).
• For every year of the patient's simulated life, the patient's complication may progress as indicated in the pictorial model at the rates determined by the estimation.
• As a person progresses, the probability of various events can change.
• For instance, someone with non-proliferative retinopathy will have a much higher probability of developing blindness compared to someone who simply has diabetes and shows no other problems. Similarly, a person's risk of stroke can increase when the individual has a history of heart disease.

Ultimately, the estimation and simulation can be used together to predict the outcome of an unknown individual or sample population based on various symptoms and medical histories.