1. Data Entry and Management for Clinical Research
Matthew Simpson
Joint Clinical Trials Office

2. The Basics of Data

3. What is Data Management
The main stages of the data management process:
The raw data are collected and entered into the computer, and checked;
The data have then to be organised into an appropriate form for analysis (often in different ways, depending on the analysis)
The data have to be archived, so that they remain available throughout subsequent phases of a project, and afterwards (5, 7, 21 years).

4. Data & the Law
All data stored and used relating to an individual is governed by the Data Protection Act (DPA)
Any data relating to a clinical trial with an investigational medicinal product (IMP) has legislation based on Good Clinical Practice (GCP) governing its storage and use
Clinical research not involving an IMP must adhere to aspects of DPA and it is good practice to adhere to GCP
Investigators are ultimately responsible for the data they collect.

5. Legal Frameworks w.r.t Data
The Medicines for Human Use (Clinical Trials) Statutory Instruments:
2004 no 1031 (‘EU Directive’)
2006 no 1928 (‘GCP Directive’)
2006 no 2984
2008 no 941
Declaration of Helsinki (1996)
Data Protection Act (1998)
Human Tissue Act (2004)
Mental Capacity Act (2005)
Second Research Governance Framework (2005)

6. Anonymous Data
Truly anonymous recordings made for treating/assessing patients may be used within the clinical setting for education or research purposes without express consent as long as this policy is well publicised.
Truly Anonymous:
Apparently insignificant features may still be capable of identifying the patient to others, such as distinguishing marks, tattoos, body piercing, posture and gait.
Research shows it is usually impossible to be sure that a patient will not be identifiable from a image or voice recording

7. Data Protection
If data collected for research purposes is anonymised or pseudo-anonymised, it does not fall within the scope of the Data Protection Act and therefore will not require the usual procedures to be followed.
Special provisions for research (Research Exemption):
Data must be used exclusively for research purposes
Data must not be used to support measures or decisions relating to any identifiable living individual
Data must not be used in a way that will cause, or be likely to cause, substantial damage or distress to any data subject
The results of research or resulting statistics must not be made available in a form that identifies any data subject.

8. Sensitive Data (non-Anon.)
You must have the specific written permission of the data subject to hold sensitive data unless you already have a legal requirement to process those data.
Security must be appropriate to the degree of harm caused by the misuse of data 
Types of Sensitive Data:
Racial or ethnic origin
Political opinions
Religious, or other similar beliefs
Trade Union membership
Physical or mental health or condition
Sexual life
Convictions or alleged criminal acts
Written, informed consent is obtained before each subject's participation in the trial

9. Data Protection: Role of the Investigator
The Investigator ensures that data is to be collected (prospectively or retrospectively) with informed consent given by the data subject.
The Investigator documents in the protocol what data is to be collected and how it will be analysed (deviations from this will require an amendment to the protocol and likely resubmission to REC)
The Investigator ensures that data will not be used for anything additional to what is specified at the time of consent.
The Investigator ensures appropriate security arrangements for both electronic (back up/ password protection) and paper (locked cupboard) files.

10. What data are you collecting?
Four types of data are collected in most clinical research protocols:
Baseline data
Patients state of being prior to initiation of protocol
Efficacy data
Assessments specific to the objective of the study
Safety data
Ongoing records of patients health until E.O.S.
Compliance data
Subject deviation from protocol

11. Collecting Data Prior to beginning study: Review research protocol
Identify all the data points you need to collect
Determine complexity
The data is simple if all the records are a single type of unit, e.g. numbers (Heights, weights, ages, BP’s, single values).
The data is complex where data have been collected from a number of different units or levels. For example, oncology studies will have lesion data as images, blood and serum data from labs (separate files?).

12. Good Clinical Practice and Data
Case Report Forms / Data Collection Forms
Do you need to create a separate form to collect the clinical data – how big is your study?
Separate forms allows for more effective Auditing and Quality Control.
Designing data collection forms facilitates the collection and entry of data and reduces the number of recorded errors.
Electronic Data Storage
Will you need to record the data beyond the paper version
Can you guarantee your meeting your regulatory requirements
Do you have an audit trail of your data changes – electronic files can be locked prior to analysis – providing a guarantee to the data validity

13. Data Management

14. Data Entry Basics – pt 1 When planning a strategy for data entry:
the aim should be a fully-documented archive of validated, correct & reliable data that can be subjected to scientific scrutiny without raising any doubts in the minds of subsequent researchers.
Many research projects do not achieve this.

15. Data Entry Basics – pt 2
When planning the system, aim to make the data entry stage as simple as possible:
For example, in a replicated experiment it should never be necessary to type long names or long codes within each visit
Simplifying the keying process will speed the task, make it less tedious and hence also less error-prone.
The logical checking phase should be done by trained staff who understand the nature of the data. Usually this phase involves preliminary analyses, plotting etc (more later).

16. Annotated CRF’s ?
TRIAL MEDICATION 1- Adalimumab 2- Azathioprine 3- Ciclosporin 4- Depo-Medrone 5- Etanercept Folic Acid 6 -Gold Injections 7- Hydroxychloroquine 8- Infliximab 9-Kenalog 10- Leflunomide 11- Methotrexate 12 -None - refer to Periods off Trial Medication Form 13 -None - withdrawn patient who is not taking any of the listed trial meds 14 -Penicillamine 15 -Prednisolone 16 -Sulfasalazine
Statistician requirement?: Define numeric values for all defined data fields.
Annotate a document to provide to the Statistician. It greatly accelerates and inproves the job of analysing the data

17. Further Data Entry
Data should be collected and recorded carefully. Consider what checks can be incorporated into the data collection routine.
For example, the best and worst patients/ animals could have a one-line comment to verify, and perhaps explain, their exceptional nature. This will confirm that they were not recorded in error.
Build in further checks if your software allows.
The simplest are range checks, but other, logical checks can also be used. For example, for a particular inclusion/ exclusion criteria or that visit dates are sequential.
If possible, use software for data keying that has some facilities for data checking (validation or logic assessment).
Do not trust visually comparing the computerised data with the original records.
Though often used, it is not considered a reliable method of finding key-entry errors, print out computer and compare with originals (sign and date the comparison as proof of validity).

18. Designing a Data Entry System
Few projects generate simple data;
Try to foresee the full range of different types of data that will be collected
Build facilities in the data collection for recording all such information e.g.: comment areas, image recording, etc.
Often data will be collected from the same patient on a number of visits.
Dates of such records must be kept, with space available on the recording sheet for notes about the patient or visit issues that the investigator feels may warrant recording but is not part of the protocol
Such secondary information will be valuable at the data analysis stage to explain any curious behaviour of the data.

19. Designing a Data Entry System – pt2
Ensure that the database system clearly specifies the units of measurement used for all quantitative variables.
Changes in research staff, or in methods of data collection, may bring about changes in measurement units.
Consideration must be given at an early stage of the database design to allow for such changes to be incorporated into the data recording system
Codes maybe needed to distinguish between information collected on different visits. (Eg; continuation of an AE with change to severity)

20. Databases for clinical research
Defining standardised tables for your clinical research can accelerate creating subsequent management programs and help identify the data you need:
Adverse Events
Concomitant Medications
Past Medical History
Demographic Data
All patients in research will have information unique to them, to which all other information is related – principally their name or subject ID (blinded study)

21. Data Cleaning

22. Interim Analysis
The interim analysis is a continuation of the checking process and should include a first look at summaries of the data. Useful things to produce at this stage are:
extreme values, in particular the minimum and maximum observations;
boxplots, to compare groups of data and highlight outliers;
scatterplots, especially if you use separate colours for each treatment
tables of the data in treatment order.
- Maintaining study blindness? CTIMP SAE’s?

23. End of Study How do you guarantee data validity at EOS?
End point data review; check all critical data for every subject with source/ raw data
Sample review; take 5% or 5 patients at random, which ever is greater, and review the entire data set for each of those subject.
Write SOP’s for each process and sign and date the review upon completion.
Inform REC and R&D office of study completion

24. Archiving
The data and programs from a research project must be archived in such a way that they are safe and can be accessed by a subsequent user.
In the absence of a proper archiving scheme, a common outcome is that the researchers leave, carrying with them the only copy of their part of the data, and hoping that the analysis and write-up will be continued later.

25. Electronic Data
Ensure and document that the electronic data processing system(s) conforms to the sponsor’s established requirements for completeness, accuracy, reliability, and consistent intended performance (i.e. validation).
Maintains SOPs for using these systems.
Ensure that the systems are designed to permit data changes in such a way that the data changes are documented and that there is no deletion of entered data (i.e. maintain an audit trail, data trail, edit trail).
Maintain a security system that prevents unauthorized access to the data.
Maintain a list of the individuals who are authorized to make data changes
Maintain adequate backup of the data.

26. The Audit Trail
An audit trail is a complete record of changes to the data and decisions made about the data and the analysis, rather like a log book.
A well-maintained audit trail greatly eases the subsequent tasks of writing reports on the data and of answering data queries.
It is a legal requirement for a drug trial
It is important to record everything you do at the time that you do it
E.g.: when errors are found during checking and changes are made to the master copy of the data, a note should be made in the audit trail.
Keep notes also on the analyses that you do (including the preliminary ones done for checking purposes), writing down the names of all files created. Every change should be dated and initialled (with a pen colour different from the original – use black for recording data first time, and red for changes).

27. Validity of your Data There is nothing new here:
just re-stating a fundamental requirement of the scientific method that you should ensure that your work is repeatable by keeping good records of what you do

28. Backing Up
It is essential to develop a system for regular "back-ups" (copies) of your data files:
Omitting to do so may result in important parts of the research data being lost.
Project managers should establish a documented routine for regularly making safe copies of the data, and should insist that all members of the research team follow the routine.
Always record your processes as SOP’s (Standard Operating Procedures)

29. Further Thoughts/ Questions?
My Details:
Matthew Simpson
Joint Clinical Trials Office