• the protocol for each trial specifies what groups of patient data will be analyzed
• ITT - intent to treat population: includes all patients who were randomized to atreatment group, regarless of their compliance with therapy or with the study rules (the protocol)
• per protocol population (PP) - includes only those individuals assigned to a treatment group who actually complied with the protocol
• ITT usually mirrors actual clinical practice, reflecting the compliance issues.

What is the common definition of major bleeding utlized by most anticoagulation trials?

• developed by the International Society on Thrombosis and Hemostasis (ISTH)
• Clinically overt bleeding that is accompanied by one or more of the following:
• decrease in hemoglobin of ≥2 g/dl over a 24 hour period
• transfusion of ≥ 2 units of packed red blood cells
• bleeding that occurs in a critical site (intracranial, intraspinal, intraocular, pericardial, intraarticular, intramuscular with compartment syndrome, retroperitoneal)
• bleeding that is FATAL

• all acute clinically overt bleeding events not meeting criteria for either major or clinically relevant non major bleeding

GUSTO severe or life threatening bleeding definition?

• either intracranial hemorrhage or bleeding that causes hemodynamic compromise and requires intervention

GUSTO MILD BLEEDING ?

• the risk of an endpoint in one group compared to the risk of the endpoint in another group at a fixed time point, such as the end
• rr = 1 means that risk is equivalent in 2 groups rr>1 means higher risk in the investigational or 1st group
• example 1.5% pts receive drug had event vs 2.2% patients in control risk is 1.5/2.2 or .68  

• RRR = 1- relative risk
• measures how much the risk is reduced in one group compared to another

• is a measure of relative risk over time vs. risk at a fixed point in time as with relative risk.
• represents the weighted relative risk over the duration of the study and the first step in its calculation is to derive the events per person-years.

• probability that an observed outcome of an experiment or study is unlikely to have occurred by chance alone.

• a measure that reflects the likelihood that the results are due to chance rather than real differences.
• usually accepted at p<.05 level which means that the probability that the reported results occurred by chance 

• an estimated range of values that is likely to include the true measure of treatment effect
• they are defined by upper and lower confidence limits. for example purposes, the 95% CI is shown (0.66, 0.95) means that the given confidence interval is the range of values within which the study investigator is 95% confident that the true value lies. 
• wide confidence intervals indicate greater uncertainty about the true value of a result.

Forest Plot

• shows how the results of subgroups compare to the results of individual trials or how results of individual trials compare to the rsults of all the trials pooled in a meta analysis.