Data management and privacy

Question 1

What is a DMP?

Answer 1

• Every study requires a Data Management Plan (DMP)
• Helps in paying attention to regulations and laws
• FAIR data principles (necessary to re-use data) (Wilkinson et al, 2016)
• -> Findable, Accessible, Interoperable and Reusable
• Almost every grant body requires a DMP nowadays
• VU also has it’s own digital DMP

Question 2

Why is a DMP important?

Answer 2

Important because enables

1. Safety of personal data (lots of laws on that)
2. To re-calculate your own data
3. Co-workers or supervisors to take over (you might get sick, switch job…)
4. To share data with others after project is finished
5. Tax payers money, we can learn more by adding data, or use for different purposes (open science: enhances science and academia)

Question 3

What is personal data?

Answer 3

• any information with which you can directly or indirectly identify a person
• E.g. name, (e-mail) address , telephone number
• But also: gender, date of birth (e.g. in relation to postal code) –> indirect

Question 4

administrative file

Answer 4

1. Personal data
   ID is created e.g. 0012
   Patient name
   E-mail address etc

2. Flow in the trial
Date screening send
Screening returned?
Date baseline assessment
Baseline assess returned?
In-or exlusion
If exlusion: reason
Date randomization
Randomisaton outcome
Date post-test questionnaire
Etc etc

• to keep track of the patients
• here you also have the personal data (not annonymous)
• only a limited amount of people have access to the file

Question 5

administrative file vs. other data

Answer 5

1. Admin file with key personal data & name
   Safely stored (password etc, not on laptop)
   Only 2 people have access to this file
2. All other data
   Anonymous (indirect –> make sure it is also not indirectly identifiable)
   Stored on ID number
   Can be accessed by more (need IC)

Question 6

software

Answer 6

Software: CASTOR or RESEARCH MANAGER

When using software to collect data:
Where is data stored (country)?
Agreement about datastorage and usage
E.g. Qualtrics not allowed

Question 7

structure of data storage

Answer 7

Data collection 
- All questionnaires
- Including data codebooks (variable lables & labels of answering options)
Data
- Admin data (patient flow, anonymous!)
- Original data set
- Cleaned data
 --> Including syntax (how to get cleaned data from original data)

Question 8

data storage after finishing

Answer 8

Sharing data after trial ended

• Data is valuable (anonymous!)
• Other people might want to answer different questions
• Or pool data: individual patient data meta-analysis
• More often: data repository’s
• Open access required nowadays

Question 9

trial registry

Answer 9

• You have to register your RCT in a trial registry
• Before inclusion of the first patient
• When submitting manuscripts: registration number required
• If you forget….. problem!

Question 10

trial registry options

Answer 10

• Worldwide 24 registries available
• Some acknowledged by International Committee of Medical Journal Editors
  For example:
  ClinicalTrials.gov
  ISRCTN.org
  trialregister.nl

Question 11

what should be reported in trial registry

Answer 11

- 20 elements should be reported:
About sponsors & How to contact & country
Title 
Health condition
Intervention
In- and exclusion criteria
Sample size
Primary & secondary outcomes
Date first enrollment & recruitment status

Question 12

trial registry : Why is this important?

Why do journals require this?

Answer 12

1. Prevents from “fiddling” with your data
   - Reviewer of your results paper can look into register
   
   • -> Are there differences?
   • -> E.g. selective reporting of outcomes!
2. Allows you to see ongoing trials (when preparing yours)
3. Allows you to find trials that are not published (e.g. meta-analyses)