Completion↦Statistic Methodology↦Data Preparation↦Procedures
What is it? Why is it important?
Study data is ready for analysis and ready for export to a statistical software system, once data collection is complete and all open issues are resolved (e.g. resolution of data queries during data monitoring).
Prior to data analysis the study data must be prepared (i.e. by a statistician). Preparatory tasks may include the appropriate:
- Recording of variables (e.g. the labelling of missing data as NA, substitute text with numbers: Male=1, Female=2)
- Formatting of time and date (e.g. day comes before month)
- Generation of variables (e.g. compute participant age based on birth-year and date at study inclusion, or BMI based on height and weight)
- Configuration of data. Example: a variable is collected at several time points, and data can either be documented in a:
- Wide format: one line per participant with several columns showing values collected at time-point 0, 1, 2…
- Long format: each participant occupies several lines, with each time point represented by an added line
What do I need to do?
As a SP-INV:
- Ensure study data is ready for analysis, and instruct the data manger to export the data form the study database in order to make it accessible the study statistician
- Ensure the esported data does not contain any variables allowing for the identification of study participants
- In the event of a blinded study, ensure procedures are in place that protect the blind (e.g. exported data does not contain variables able to disclose the treatment received by study participants (e.g. intervention or control group?))
The statistician can start to prepare the data for analysis, once data is accessible in a statistical software system.
More
In order for the data analyses to be reproducible, it is crucial that staff working with the study data (e.g. data manager, statistician, SP-INV) have access to the same data-set. For this reason, corrections to study data should only be made in the study database. Thus, corrections on exported data during the data preparation process should be avoided.
In the event corrections in the database is no longer possible or permissible, the statistician should:
- Seek prior approval by the SP-INV before changing or correcting any data
- Document implemented corrections to the exported study data
Where can I get help?
Your local CTU↧ can support you with experienced staff regarding this topic
Basel, Departement Klinische Forschung, CTU, dkf.unibas.ch
Lugano, Clinical Trials Unit, CTU-EOC, www.ctueoc.ch
Bern, Clinical Trials Unit, CTU, www.ctu.unibe.ch
Geneva, Clinical Research Center, CRC, crc.hug.ch
Lausanne, Clinical Research Center, CRC, www.chuv.ch
St. Gallen, Clinical Trials Unit, CTU, www.kssg.ch
Zürich, Clinical Trials Center, CTC, www.usz.ch
References
ICH Topic E9 – see in particular
- 5.2.1 Full analysis set
- 5.2.2 Per protocol set
Swiss Law
ClinO – see in particular article
- Art. 2b Definition of intervention