-- BenSaville - 06 Nov 2009

Actionable Items for SIT trail

  • Add event ASCIE (acute silent cerebral ischemic event) in the descriptive report and stats report.
  • Provide descriptive stats for the person year from "date of informed consent" to "the date of their last well visit height, weight and blood pressure recorded"
  • Provide frequency and percentage of patients seen within one year and two year by site and overall (20 sites in total)
  • Wait for the demographic data to be cleaned by WashU and rerun the analysis then
  • Find out how many patients have RPMRI? and how many patients are tested positive or negative
  • Choose three predictors of most interest (e.g., age, gender, age times gender) and just test the linear relationship
  • Stratified the survival curves by blood pressure levels as Pegelow's paper
  • Update the analysis plan.
  • Background reading
    • Reading background paper
    • Export data from redcap and generate descriptive stats
    • Get familiar with the dataset
  • Statistical analysis
    • Work on the analysis plan
    • Reshape the dataset from wide to long (Note: also including spirometry data, but the visit date has a lot of missing and messy input)
    • One more issues on the well visit date --- they are ordered in decreasing date (from latest to oldest)
    • Check the date of spirometry data

Completed Items

    • Include the interaction of gender with all other predictors
    • Fit a marginal model on blood pressure with age, gender and their interaction.
    • Provide predicted means plot of blood pressure vs. age by gender
    • Provide correlation matrix for all the predictors used in the cox PH model
    • Provide a dataset containing predicted FEV and FEV%, actual FEV, with age, height, weight, bmi
    • Provide p values for table 2-5 in report 10
    • Use black equations for other group
    • Figure out the tick marks in the survival curves
    • Provide HR estimate for the comparison of two groups in the KM plot
    • Report the missing values in event date and outliers in height, age in the spirometry dataset.
    • Create FEV% and FVC variables in the spirometry dataset.
    • Perform log rank test for comparisons of positive MRI group versus negative MRI group.
    • Provide survival curves for the endpoint--with any type of event.
    • Provide boxplot with labeled outliers
    • Provide the total number of patient years
    • Provide number of patients in the model
    • Recode the missing in the stroke event as no and rerun the descriptive tables and KM plot.
    • Add a legend in the KM plot annotating the confidence intervals.
    • Check why there are mismatches in weight, height and bmi.
    • Check 211 missing records in history of silent new infarct
    • Provide number of records par patient
    • Look at the missing values in age in the demographic dataset
    • Categorize age into intervals and provide descriptive stats for each variable within the interval
    • Total patient years (overall, by those with and without events)
    • Boxplot
    • Frequencies for 3 types of events (silent infarct, CIA, over stroke)
    • Give Dionna first 25 and last 25 patients IDs in blood pressure, age, height, weight, hemoglobin.
    • Remove 9 patients with TIA/stroke from the analysis
    • Provide descriptive stats for all the variables in the model
    • Provide KM curves for the non-missing data
    • Determine the type of events, included in the table 1
    • Total patient years, by with events and without events
Topic revision: r26 - 08 Jun 2012, WenliWang
 

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback