Grant Writing

Data Collection and Analysis Procedures for Grant Writing

Data collected at the beginning of a clinical study for all participants and for each arm or comparison group. These data include demographics, such as age and gender, and study-specific measures (for example, systolic blood pressure, prior antidepressant treatment). (See also Baseline Characteristics results data element on

  • Within the data collection procedures, the principal investigator should describe the method for identifying candidates for the study. This includes:
    • a thorough description of the procedure for obtaining data,
    •  the type of data to be collected and the timeframe, i.e., lab tests, procedure outcome, length of stay, etc.
  • The principal investigator should attach a sample of the database elements to be collected, often called a Data Collection Form.
  • All study forms contain page numbers and a place for the person collecting the information to sign and date. 
  • An example of data collection procedures could be:
    • Investigators will ask Medical Records department to run a query to generate a list of potential candidates based on the following criteria: [date range, ICD-10 code, etc.]
    • The list of medical records generated in that report will be reviewed for the following: [demographics, patient history, course of disease, etc.]. Those charts not meeting the more specific inclusion/exclusion criteria will be removed from the list and no data recorded. Those charts meeting all inclusion/exclusion criteria will have data recorded in the data collection form.
  • When the principal investigator begins considering the data analysis plan, it should be noted that some organizations have biostatisticians that may be available for consultation in developing a statistical plan.
  •  The data analysis plan should identify the measures the principal investigator proposes to use [i.e., response (measure of interest) explanatory (factor potentially associated with measure of interest), other (demographics etc.)].
  • There are both primary measures (including definition, data type and range of values) and secondary measure(s) to consider. This section should list the statistical methods to be used to address the primary and secondary objectives.
  • The principal investigator should specify any confounding variables for which it is anticipated adjustment will be made, and explain how missing data and outliers will be handled in the analyses. 

Further Reading

%d bloggers like this: