Toolbox » Data Management Tips

Data Management Tips

  Print This Page

These data management tips apply to multi-center studies. Investigators of single-site studies are encouraged to review them and adapt them as relevant.


Screening Log - captures data on individuals screened, including those who are not eligible.

Study schedule – identifies visits and telephone calls in a matrix format.

Written procedures – define data and their collection at each subject contact.

Case Report Forms [CRFs] - capture all data required by the study protocol. Use PPRU templates wherever possible.

  • Utilize numerical response categories in data elements to allow summarization and analysis.
  • Avoid redundant data elements and unnecessary data.

Adverse event (AE) form - ensures collection of the necessary data to generate safety reports.

  • Record only one diagnosis, sign or symptom per event.
  • Death should not be recorded as an event but should be recorded as the outcome of the event.
  • Do not use abbreviations.

Data Collection Procedures - for completing CRFs.

  • Print neatly and legibly.
  • Complete all data fields on the CRF.
  • Write in black ink and press firmly.
  • Use only subjectID and initials and not the full name and chart number of the subject.
  • Use conventions for missing data (e.g., NA= not available/not applicable; ND = not done; UNK= unknown or 97= not available/not applicable; 98= not done; 99=unknown).
  • Provide clear, concise comments and only when necessary.
  • Provide clear, concise comments and only when necessary.
  • Record dates as specified.

Data Handling Procedures –describe how data are entered into a computer system (e.g. at study site or transmitted to Coordinating Center for central entry) and tracked from collection through entry, data clarification and study closure, ensure enough specificity to reproduce the analysis datasets from the source documentation.

Clinical data management system (CDMS) – typically a computer system for data entry, editing, storage, and transmission to an analysis package. Include:

  • Edit Processes – check for missing, inconsistent, and erroneous entries; describe data correction processes and tracking changes.
  • Status – method or codes that track the status of each subject throughout the study.
  • Laboratory Data - describe process for obtaining, transmitting, and reconciling laboratory data.
  • Data Management Reports– routinely describe study progress and data quality.
  • Final, Cleaned Data Set - mechanism to ensure it is created and archived when the study has completed.  All clinical data, metadata, administrative data and reference data and documents should be maintained.

Data Sharing Policy-identify considerations for sharing the final data set.

Confidentiality Mechanisms – to ensure that the Health Insurance Portability and Accountability Act (HIPPA)guidelines are followed to protect the confidentially of participants’ data and to establish a convention for unique participant identifiers (PIDs).

Follow-up Procedures – describe procedures for following subjects from enrollment through study completion.