The Food and Drug Administration (FDA) assesses compliance with the regulations governing clinical investigations, reviews the progress and conduct of these studies, and ultimately evaluates safety and effectiveness of the test article by reviewing a variety of documents. For these purposes, accurate and complete study documentation is important. Following is a description of certain clinical investigator documentation requirements contained in 21 CFR parts 312 (drugs and biologics) and 812 (devices). The recordkeeping requirements for devices apply to significant risk devices, not to devices determined to be non-significant risk devices.
Certain FDA regulations focus on those records considered necessary to ensure strict control over distribution of test articles used in clinical investigations. Clinical investigators are required to maintain records pertaining to the test articles received from the sponsor, the date the test articles were received, and their disposition. When the study is terminated, discontinued, or completed, the investigator is required to return the unused supply of drugs, biologics, or devices to the sponsor or, if authorized by the sponsor, the investigator may dispose of the test article(s) in some mutually agreeable way.
Generally, accountability records do not include details on each subject's use of a drug or biologic product. For device studies, the investigator must maintain a record of the exposure of each subject to the investigational device, including the date and the time of each use, and any other therapy. Such records may also be useful in the event of the product recall. Records on the amount of product returned by the subject can serve as a check of subject compliance with the testing regimen.
Investigators are not permitted to supply an investigational drug, biologic, or device to any unauthorized person and should maintain an inventory that reflects overall disposition of test articles to ensure that the articles are distributed only to those authorized to receive them [21 CFR 312.61, 812.110(c), and 813.107(a)]. Investigators are expected to maintain records to assure proper control of the investigational product.
If test articles are sent directly to the pharmacy department or other appropriate department in the institution, rather than to the investigator, the pharmacy or other department may distribute the test article to the investigator or the subject and maintain the accountability records. When the test article is distributed in this manner, the investigator remains responsible for ensuring that adequate records are maintained.
Two types of records are kept regarding a clinical investigation: 1) case history records and 2) the study protocol and related documentation.
Investigators are required to prepare and maintain adequate and accurate case histories, recording all observations and other data pertinent to the study about each subject treated with the investigational article or enrolled as a control. Investigators are to maintain these records even though the research sponsor may also have such records.
Case history records include both case report forms and documentation that supports data in those forms. They should contain:
- Basic subject identification information
- Information showing that each subject meets the subject selection criteria or justification for otherwise enrolling the subject
- Sufficient information to support data in the case report form as submitted to the sponsor
- Information on each subject's exposure to the test or control article, including the date (and time, if relevant) of each administration and the quantity administered
- Copies of case report forms submitted to the sponsor
Case history records must be retrievable in such a fashion that all information regarding each individual in a study is attributable to that individual.
Case history records also include information obtained form tests and examinations, such as:
- Physical examinations
- Lab results
- Progress notes
- Information and data on the subject's condition before, during and after the clinical investigation
- All diagnostic test results
- Diagnosis made
- Concomitant or concurrent therapy
- Factors that might alter the test article's effect (e.g., development of an illness while on study)
To substantiate that each subject meets the subject selection criteria, the investigator should maintain records of the subject's medical history before entry into the study (if these records exist in the investigator's own files or hospital files). In most cases, the investigator is not obligated to seek past medical history records from other physicians who may have treated the subject or who referred the subject to the investigator. If the subject has had no previous contact with the investigator, the medical history taken when the subject enters the study should show that the subject meets the study inclusion criteria.
Investigators should maintain correspondence sent to, or received from, the study sponsor and the monitor, including the protocol or investigational plan, materials to be used in obtaining informed consent, protocol modifications, and records of Institutional Review Board (IRB) approval and of other communications/actions pertaining to the study.
The FDA assesses study results through a scientific evaluation of the data contained in case report tables summarizing the data in case report forms. The case report format and content often varies from investigation to investigation. Case report forms are a critical part of the investigation records, but in most cases, they cannot serve as the complete investigation record. The case report form should contain all data required by the protocol but need not duplicate all the investigator's records on the subjects' medical histories. Likewise, everything in the case report forms need not be duplicated in the medical records. The FDA does not require that special medical records be established to meet its requirements.
When the FDA needs to verify validity and completeness of the case report form data submitted to the Agency, it may audit case history records in the possession of the investigator or investigator's institution [21 CFR 312.68 and 812.145].
All study protocols are required to contain:
- A statement of the study's objectives and purpose.
- For studies of drugs (including biologics, medical foods and food additives): each investigator's name, address, and statement of the qualifications; each sub-investigator's name; the research facility's name and address; and each reviewing IRB's name and address. (For device studies, this information is contained in the Investigational Device Exemption (IDE) investigational plan, rather than the protocol.)
- Subject selection and exclusion criteria and the estimated number of subjects to be studied; for device studies, the actual number of subjects must be given.
- A description of the study design, including any controls to be used, and a description of methods to minimize bias on the part of the subjects, investigators, and analysts.
- For drug and biologic studies, the method of determining doses to be administered, the planned maximum dosage, and the duration of subject exposure to the drug. For device studies, the method for determining the treatment parameters, including administration and duration of subject exposure to the medical device.
- A description of the observations and measurements to be made to fulfill the study's objectives.
- A description of clinical procedures, laboratory tests, and other measures to be taken to minimize risk and to monitor the effects of the test and control articles.
The investigator must maintain a current study protocol with any amendments. The investigator must submit any changes to, or deviations from, the protocol to the IRB for review and approval before they are initiated, unless the changes or deviations are necessary to eliminate apparent immediate hazards to the human subjects. For device studies, the research sponsor must also approve changes before they are instituted. However, in drug and biologic studies, the investigator is only required to notify the sponsor of proposed changes in the protocol. The FDA does not require the sponsor to explicitly approve the change. For significant risk devices, except when the deviations are necessary to protect the life or physical well-being of a subject in an emergency, the FDA must approve all changes or deviations from the investigational plan that may affect the plan's scientific soundness or subject rights, safety or welfare. Deviations that are necessary to prevent harm should be reported to the IRB and to the sponsor, who should report the deviations to the FDA.
FDA regulations require investigators to retain records for a specified period. [Note: these time periods are different from those required for IRB records.] For investigational new drug (IND) studies (and medical food and food additive studies), records are to be maintained for two years following the date of marketing application approval for the drug for the indication for which it was being investigated. If no application is filed, or if the application is not approved for the indication, the records are to be retained for two years after the investigation (i.e., the IND) is discontinued, and the FDA is notified of that fact. For device studies, records are to be maintained for two years after the later of the following dates: the date on which the investigation is terminated or completed; or, the date that the records are no longer required to support a pre-market approval application or a notice of completion of a product development protocol.
To comply with FDA record retention requirements, clinical investigators should arrange with study sponsors to be kept informed of the status of the application for their respective studies. To illustrate, for an IND study, FDA regulations require sponsors to notify each investigator if the FDA approves the new drug application (NDA) or product license application (PLA), or if the investigation is discontinued. Therefore, the FDA recommends that all investigators insist that their contract with the sponsor include a provision requiring the sponsor to notify the investigator of any action with regard to the test article (e.g., submission or approval of an NDA or PLA, withdrawal of an IND, or the placement of an IND on inactive status).
Retention of accurate and complete records is essential to establish the validity and completeness of a report on a clinical investigation that is submitted to the FDA in support of an application for a research or marketing permit. The investigator, not the sponsor, is responsible for the accuracy and completeness of his or her study records, and the investigator is responsible for any discrepancies found in these records during an inspection.
The investigator may maintain all the study records, or, when the subject is a patient in a hospital or other facility, the records may be maintained as part of the patient's hospital or clinical records. If a hospital or clinic keeps the records, the investigator must still ensure that the records are retained for the length of time set forth by the regulations.
An investigator may retain records either in their original form, on microfilm or microfiche, as photocopies, or as other accurate reproductions of the originals. However, If reproductions are used, they must be legible; and, the investigator is required to assure that such reproductions are true and accurate copies of the original. When reproduction techniques (e.g., microfilming) are used, a reader and photocopying equipment should be readily available. If written notes, erasure marks, or other changes are not apparent, a notation of this fact should be clear on the reproduction of the record, and the original record should be retained for the time required.
Raw data, entered directly into a computer system, is considered to be the original or true copy of the data, whether it is printed out as hard copy or stored on computer files. An acceptable computerized data collection system would be on that (1) allows data entry only by authorized individuals; (2) controls the ability to delete or alter previously entered data and provides an audit trail for such data changes (e.g., a modification file); (3) protects the database from tampering; and (4) ensures data preservation. If records are retained in a computer data system, suitable equipment should be readily available to produce a hard copy of the data. Data should be retrievable in such a fashion that all information regarding each individual in a study is attributable to that individual, i.e., case report form equivalents must be available.
The FDA may audit any and all records that might support microfilm, microfiche, or other stored data. These records must be maintained just as paper case report forms must be maintained.
The FDA understands the need to protect the privacy of research subjects. Study records need not identify subjects by name, but do need to provide some type of identifier to permit cross-indexing. Identifying information must be available to respond to allegations that may arise, such as the claim that a subject's consent was not obtained, or that the study records do not represent actual studies or present actual results. When an individually identifiable medical record is copied and reviewed by the Agency, the FDA safeguards the information and uses or disseminates it only under conditions that protect the individual's privacy to the fullest possible extent consistent with laws relating to public disclosure and the Agency's law enforcement responsibilities.
Research subjects' expectations concerning confidentiality will vary, depending on the study and the subject's relationship to the clinical investigator. Because FDA oversight responsibilities may compromise subject confidentiality, the Agency requires consent documents to note that the FDA may inspect the records and describe the extent to which confidentiality will be maintained by the investigator [21 CFR 50.25(1)(5)]. The subjects must be informed if anyone other than authorized hospital or office personnel will have access to records containing their identifiers. Although FDA regulations neither require nor prohibit sponsor access to study records, subjects must be made aware of the extent to which such access will be allowed.
When clinical investigators conduct a study for submission to the FDA, they agree to allow the FDA access to the study records. The investigator is responsible for making the subject's records available to the FDA for inspection and copying. The Agency will inspect and copy whether or not the subject has agreed to such review [21 CFR 312.68 and 812.145].