Residency Project Year Two Guide

Return to Residency Project Guidelines

Some of the steps in Collecting Data are a review from Year 1.

Collecting Data

You will collect your data.

For UC Davis residents, the entire process to setup a REDCap database to collect data is described at: REDCap at UC Davis.

During Year One, you met with the database/informatics support personnel and reviewed the scope of your project . During Year Two, you will create a database and enter your study data.

Create a Data Dictionary

You will meet with your advisors and determine the data points that you will be collecting for your study.

Vanderbilt University, originators of REDCap, have produced a video describing how to create your data dictionary: on creating a data dictionary.

They have also provided a sample data dictionary which can changed as needed for your study: Sample CSV File with REDCap Data Fields.

Data Review Meeting

Resident will review the data dictionary with their mentor, a statistician, and an informatician to make sure the appropriate information is being collected and the data is structured appropriately for statistical analysis.

Create Database

You will need to create a database.

Specific site information: UC Davis

Documents needed to access REDCap at UC Davis. The CTSC Biomedical Informatics Program will review your data dictionary and create your database in REDCap. Final Database Review

You and your team will ensure that the formswill be collected in the appropriately.

Collect Data

You can now collect the data.

Specific site information: At UC Davis the CTSC Biomedical Informatics Program can extract data from the EMR and other data sources and import it into REDCap (provided that it is in the correct format). Residents can conduct chart abstractions for the data that cannot be obtained directly and capture it in REDCap.

Additional Data Collection

If you need additional data that is is not in the general database, e.g. Radiology data, you will need to collect this data from the Radiology database. With REDCap services, you will have a database seeded or structured to collect additional data.

Clean Database

• Complete and clean database that can be handled electronically. If you do not use REDCap, be sure that your Excel or Access data is clean.
• Data Preparation: Data needs to be prepared for analysis.
• Coding Scheme - Re-code your variables:
  • Be sure to code free text into variables.
  • You can do automatic re-coding for your entire database.
  • Be sure to check and fix all errors in the data.

Information for Data Collection

During Year 2, you will look more specifically at data types and study types for your question. You will want to make sure that the type of data, e.g. categorical, nominal, interval, etc you are collecting will help you to answer your question. You will want to identify the type of study you are doing: case-control, cohort, cross-sectional and whether the data that you are collecting can be used for that type of study. The following materials can help you answer specific questions about your study design.

* Browse Clinical Research Case Studies for step-by-step review of clinical research questions.

* Browse Educational Course Materials for slide and video presentations of important concepts in clinical research.

• Browse Articles or Content of Interest and Clinical Research Vocabulary to familiarize yourself with clinical research terms.

Information for Analysis

• Educational Materials contains a number of Stata Courses currently taught at CTSA institutions.
• You can request an R-Studio account for your analysis.
• In Statistical Tools you will find SAS macros, R-scripts, and Stata files for doing clinical research analysis.
• In Statistical Graphics you will find many graphs for discussing safety parameters on treatment studies and a best practices discussion for safety graphics. The graphs include SAS macros, R-scripts, and Stata code for reproducing these outputs. These graphs have been developed by FDA, industry, and CTSA biostatisticians.

What Could Go Wrong?

Reproducible Research provides templates for determining if your results are reproducible as well as discussion of what can go wrong on a clinical research study.

This topic: CTSpedia > WebHome > EducationalMaterialsBERD > ResidencyTrainingPrograms > ResidencyProjectGuidelines > ResidencyYearTwo
Topic revision: 06 Sep 2012, MaryBanach

---