As much as we might like to think that the purpose of clinical research is to help the patients who are subjects in the clinical trial, the real purpose of clinical research is to provide data to prove that a drug, biologic, medical device, or treatment is safe and efficacious, and to therefore benefit patients in the future. If a current subject receives benefit from the trial, that is a bonus.
In order to fulfill the purpose of performing clinical trials, data and records must be accurate, complete, and verifiable. Some of the source data and records generated during clinical research include:
- IRB approvals
- IRB correspondence
- IRB approval of protocol revisions
- Documentation of outcome of expedited IRB review
- Evidence of meeting eligibility criteria
- Signed informed consent forms
- Laboratory results
- Patient reported outcomes (PROs) and diaries
- Reports of the details of treatment, such as radiation
- Surgical reports
- Documentation of adverse event
- Treatment notes
- Documentation of toxicities and grades
- Medical Progress notes
- Dosing assignments
- Arm and cohort assignments
- Documentation of outcomes
- Results of monitoring visits
Regardless of whether source data and records are paper-based or electronic, they must be legible, accurate, secure, and indicate by whom, when, and why they were created or modified. For paper records, these requirements are part of Good Documentation Practices.
The following is an example of a Good Documentation Practices Policy that is suitable for use in written data and records generated in conjunction with clinical research
|
|
Good Documentation Practices Policy
This policy is an example of one that could be implemented to assist research staff in complying with Good Clinical Practice documentation requirements. It may be customized to fit the environment in which it will be used.
1.Purpose
This policy describes good documentation practices that apply to paper data and records generated for clinical research. This policy defines documentation practices for all study records and data, laboratory values, study subject notes and records, and regulatory documents.
2. Affected Departments
All departments and staff involved in documentation for clinical research.
3. Policy
3.1 Entering Data
3.1.1 Handwritten entries on documents must be made using permanent (indelible) blue or black ink. Erasable ink, non-waterproof ink, and pencil are not permitted
3.1.2 Correction fluid and correction tape are not allowed on documents.
3.1.3 The date and time hand written on a document will be the current date and time at the location where the handwritten entry is made
3.1.4 The preferred format for the handwritten date on documents in the U.S. is the format: 2 digit day, 3 character month, 4 digit year. The format 2 digit month, 2 digit day, and 2 digit year is also acceptable in circumstances where the date will not be confused
3.1.5 The handwritten time on documents in the U.S. will include a.m. or p.m.
3.1.6 Document entries must be made immediately after completion of a step or a process
3.1.7 Document entries must be made immediately after completion of a step or a process
3.1.8 Signature entries shall be consistent with the signature recorded in the study signature log.
3.1.9 Initial entries shall be consistent with the initials recorded in the study signature log.
3.1.10 No one shall enter a signature or initials for someone else.
3.1.11 All entries must be made directly onto the official record or document. The use of scratch paper, post-it notes, or unofficial notes to record data is not permitted.
3.1.12 All documents requiring review and approval signatures shall contain the original handwritten signature and date signed. Signatures made by rubber stamps, pre-printed labels, photocopy, or fax are not permitted.
3.2 Editing Data
3.2.1 When a correction is needed, draw a single line through the entire incorrect entry, enter the correct information and initial and date the correction. When the reason for the correction is not obvious, place an asterisk or number next to the incorrect entry and explain the correction at the bottom of the page, identified by the asterisk or corresponding number.
Note: Write-overs are not permitted.
3.2.2 All missing entries must be explained. If an entry is not completed at the time the function is performed, place an asterisk or number at the point of the missing entry and at the bottom of the page explaining the missing entry. Initial and date the entry.
3.2.3 To ensure that inappropriate entries are not made at a later date, a line shall be drawn on all blank spaces if the reason for that blank space is not apparent. If the blank space is left because an entry is not applicable, “NA” may be entered.
3.2.4 Backdating (entering a date on a day after the entry was made or the task was performed) is not permitted.
3.2.5 Postdating (entering a date in the future) is not permitted.
3.2.6 When a document requires an entry upon completion of an activity and the activity was performed but not documented, an explanation (by the performer of the activity) of why there was an omission must be included, signed, and dated.
4. Responsibilities
4.1 All staff members involved in clinical research are responsible for following these good documentation practices.
4.2 Quality Assurance is responsible for audit to ensure compliance with these good documentation practices. |
|
Electronic Data and Records
Electronic data and records are subject to many of the same controls as paper data and records and to additional requirements intended to ensure the integrity of electronic information. Each study protocol should state where computerized systems (including electronic medical records and spreadsheets) will be used to create, modify, maintain, archive, retrieve, or transmit data.
Acceptance of data from clinical trials by sponsors and regulatory agencies for decision-making purposes is dependent upon the ability to verify the quality and integrity of such data. To be acceptable the data should meet certain fundamental elements of quality whether collected or recorded electronically or on paper.
Data should be attributable, original, accurate, contemporaneous, legible, and secure. For example, attributable data can be traced to individuals responsible for observing and recording the data. In an automated system, attributability could be achieved by a computer system designed to identify individuals responsible for any input. Just as for revising a hand-written record, revisions to electronic records (that may be used in submissions to the Food and Drug Administration for approval of a test article) should indicate who changed the data, when, and a reason for the change.
The Code of Federal Regulations (21 CFR Part 11, for an explanation on the current status of 21 CFR Part 11 enforcement) refers to this revision history as the “audit trail”. And computerized systems that create, capture, and store electronic data for clinical research must be validated. (See Guidelines for the Use of Computerized Systems in Clinical Research. )
Security of electronic data and records is an issue, since electronic data is easily manipulated if adequate controls are not in place. Having a validated system and a change control process help to ensure that the data remains complete, accurate, and original. Validation establishes a baseline for the accurate functioning of software applications and networks that hold clinical research data and records, and change control ensures that no inadvertent changes that could affect data and records are made to the applications and networks. Sponsors of clinical trials and the FDA alike emphasize the importance of validation and change control in creating and maintaining data and records that can be used for submissions for drug or device approval.
Excel spreadsheets are very popular and useful in tracking events associated with trials. However, Excel does not indicate when changes are made so procedural controls must be put into place to ensure that changes to data are tracked. When dosing decisions and study-related decisions concerning adverse events and toxicities are stored on spreadsheets, those spreadsheets are source documents and are subject to security, validation, and maintenance requirements. Functional testing should be performed and documented so that the user is assured that the spreadsheet functionality is without errors. Standard Operating Procedures for the use of spreadsheets should include system set-up/installation, security access, data format conventions, quality control procedures, revising data and documenting changes, version control and change control, and back-up and restore procedures.
|
|
Best Practice – Written Procedures
One major part of the foundation of any Quality System lies in having written policies, procedures, guidelines, or standards that describe the controlled processes used to perform functions. This is true whether the quality system is ISO, GCP, TQM, or any other type.
Best Practice – Training
Of course, in order for written procedures to be effective, staff must be trained on these standard practices and must adhere to them. Adherence to the procedures is ensured by auditing and following up on audit findings.
Any of several systems can be used to document policies, procedures, standards, and guidelines, but they all should describe the intent of the organization, how to perform the functions in a standard way, and how to document that performance. One commonly used system that many PSC clients have adopted is a four-level approach consisting of
- Policy
- Process
- Procedure
- Form
A Policy is a statement of intent. It describes the organization’s philosophy and the intended consequence of that philosophy. An example would be a Policy on the Protection of Human Subjects, which would include how the organization intends to protect the safety of human subjects who participate in clinical trials.
A Process is a sequence of inputs that are turned into outputs. An example would be the process by which a subject is enrolled in a trial. Examples of inputs to this process would be the initial identification of a patient as a potential subject, explanation of the trial and its risks to the potential subject using the consent form, and determination of eligibility. Some outputs of the enrollment process would be an appropriately signed Informed Consent Form, an appropriately completed and signed eligibility checklist, and an accurate entry of the subject into the trials database.
A Procedure describes steps taken to accomplish a task or function. It describes how a function is performed, by whom, with what responsibilities and training, and the acceptable outcome. An example of a procedure would be obtaining Informed Consent. This procedure would describe who may obtain consent, what training they must have, the steps to be taken in obtaining the consent, and how to document the activity.
A Form is a tool that is associated with documenting a procedure. The benefit of forms is that they make the recording of data simpler and more standardized. A completed form in clinical research is a source record. Forms may be filled out by research staff or by subjects (as in Patient Reported Outcomes – PROs).
PSC recommends that organizations that conduct clinical research develop a hierarchy of Policies, Processes, Procedures, and Forms to provide infrastructure that is needed for high quality, compliant research.
|
|
|
