It is the investigator's responsibility to explicitly supply the necessary information for each type of waiver request. Failure to address each of the elements for each type of waiver will delay the IRB's ability to act on the request.
Waiver of one or more elements of consent is permitted provided that the research is no more than minimal risk and meets specific criteria. Waiver of an element of consent is appropriate if one or more of the 8 required elements is not relevant to the research activity. Complete waiver of consent is also permitted; this is most frequently granted for retrospective research but is also possible for some types of prospective research.
Regulatory Criteria for Waiver of Consent
Waiver or alteration of consent:
§46.116(d) An IRB may approve a consent procedure which does not include, or which alters, some or all of the elements of informed consent set forth in this section,or waive the requirements to obtain informed consent provided the IRB finds and documents that:
- the research involves no more than minimal risk to the subjects;
- the waiver or alteration will not adversely affect the rights and welfare of the subjects;
- the research could not practicably be carried out without the waiver or alteration; and
- whenever appropriate, the subjects will be provided with additional pertinent information after participation.
Rights and Welfare:
It is generally accepted that subjects aren't waiving any rights to use the data that is collected as part of routine clinical care provided that adequate provisions are in place to protect the confidentiality of the data. Use of individual's data for research is analogous to use for quality improvement and other administrative purposes. The January 31, 2008 SACHRPP Letter to the Secretary of HHS put recommended the following list of things that should be considered:
- Whether there are other federal, state, or local laws that provide rights to potential subjects to require informed consent. IRBs should seek advice from their legal counsel when appropriate to help the IRB with these determinations. This would be especially important for state specific regulations.
- Whether the subject population, in general, would object if they knew of the waiver and its intent in facilitating research.
- Whether the subject population, in general, would consider that the waiver has the potential to cause adverse consequences for their welfare or general well being.
January 31, 2008 SACHRPP Letter to the Secretary of HHS
4. Regarding the criterion under HHS regulations at 45 CFR 46.116(d)(3) for IRB approval of a waiver or alteration of informed consent requirements, IRBs should consider the following points when determining whether research could not practicably be carried out without the waiver or alteration:
- The commonly accepted definitions of the term "practicable" are (a) feasible; (b) capable of being effected, done or put into practice; and (c) that may be practiced or performed; capable of being done or accomplished with available means or resources.
It should be noted that this criterion states that the research could not practicably be carried out without the waiver or alteration. Put another way, it would not be practicable to perform the research (as it has been defined in the protocol by its specific aims and objectives) if consent was required. The emphasis being that it is impracticable to perform the research, and not just impracticable to obtain consent. The following concepts may help an IRB determine whether the research could not be practicably carried out without the waiver of consent:
- Scientific validity would be compromised if consent was required. Examples of this might include the following:
- The sample size required is so large (e.g., population-based studies, epidemiology trials) that including only those samples/records/data for which consent can be obtained would prohibit conclusions to be drawn or bias the sample such that conclusions would be skewed.
- The subjects for whom records would be reviewed are no longer followed and may be lost to follow-up. For example the proportion of individuals likely to have relocated or died may be a significant percentage of the subject population and the research results may not be meaningful and lose statistical power.
- The disclosure of the study purpose as part of the consent process would bias the research subjects so that the results will not be meaningful.
- Ethical concerns would be raised if consent were required. For example:
- There is a risk of creating additional threats to privacy by having to link otherwise de-identified data with nominal identifiers in order to contact individuals to seek consent.
- There is a risk of inflicting psychological, social or other harm by contacting individuals or families.
- The disclosure of the study purpose as part of the consent process would bias the research subjects so that the results will not be meaningful.
- There is a scientifically and ethically justifiable rationale why the research could not be conducted with a population from whom consent can be obtained.
- Practicability should not be determined solely by considerations of convenience, cost, or speed.
Not Practicable: The most important discussion should target why it is not practicable (not possible) to conduct the research without the waiver. (See the box on the left for more details.) The justification must provide a reasonable rationale for why the research would not be possible without the waiver.
Studies that are entirely retrospective (use existing data, specimens and medical records that are available as of the date of IRB submission) represent a large percentage of the studies submitted to the IRB. These studies generally request and obtain a waiver of consent. However, if there are additional prospective components to the research, e.g. some subjects will be telephoned and asked to complete a questionnaire, it then becomes entirely practicable to obtain consent for the entire study, not just the questionnaire portion. If the data will be collected prospectively (after the date of IRB approval), a thorough justification is needed since it would be (or would have been) practicable to obtain consent.
Demonstrating That Research Would Not be Practicable
The criteria in 45 CFR 46.116(d)(3)"the research could not practicably be carried out without the waiver or alteration;" does not mean that it is not practicable to get consent; it literally means that it would not be practicable to conduct the research. There are situations where the investigator has access to the prospective subject but getting consent would make it impossible to actually conduct the study or would threaten the scientific validity of the research results.
For more information on what "not practicable" means, see the January 31, 2008 SACHRP letter to HHS Secretary: Recommendations related to waiver of informed consent
Research Limited to Use of Existing Records or Specimens
The IRB's most frequent request for waivers is for research involving existing medical records or specimens. The investigator must request a Waiver of Consent and a Waiver of HIPAA Authorization. If children are subjects, Waiver of Assent. The criteria for waiver of assent and consent are the same.
The basis for both waivers (consent and HIPAA) include the requirements for the research to be no greater than minimal risk and that it not be practicable to conduct the research without the waiver.
What are existing records or biospecimens?
The terms existing, retrospective and prospective are frequently misused by investigators. This confusion leads to the potential for research non-compliance.
- Existing means that the records/biospecimens are already available as of the date of submission to the IRB.
- Retrospective means reviewing records or using specimens that are existing.
- Prospective means that the records or specimens don't exist yet as of the date of IRB submission. Reviewing data from medical charts after the information has been recorded is still prospective review from the IRB's perspective.
Common reasons for it to be impracticable to obtain consent and authorization
- Subjects are no longer seen in clinic or have moved;
- There are too many subjects to contact all;
- It is not possible to contact all of the subjects and the validity of the research depends on inclusion of all subjects.
When can the IRB grant a second waiver to extend the enrollment period?
The IRB categorizes this type of request as a successive waiver. At the time of initial approval, the IRB would have made the determination that it was impracticable to conduct the research without the waiver. If the request to the extend the enrollment period does not extend beyond the date of the original IRB submission, the IRB may grant the waiver request. However, if the request to extend the enrollment period goes beyond the original date of submission to the IRB, then that portion of the data is considered to be "prospective". In order to grant a waiver, the request for a new waiver is needed that explains why it would have been impracticable to obtain consent of the prospective subjects. The examples below may help.
The investigator requests a waiver of consent/assent and HIPAA to review records for a retrospective cohort study. The date of the original submission is January 2, 2009 and the records extend from 1985 to 2006. The investigator subsequently submits an amendment to the study and requests a second waiver of consent/assent and HIPAA to extend the review records from 2006 to 2008. The rationale for original waiver request likely covers the additional records.
The investigator requests a waiver of consent/assent and HIPAA to review records for a retrospective cohort study. The date of the original submission is January 2, 2009 and the records extend from 2005 to 2008. The investigator requests an amendment to extend the review records from 2008 to the end of 2015. The rationale for original waiver request likely covers the additional records up to and including January 2, 2009 but the records from January 3, 2009 to 2015 are considered prospective. The request to review the records of the prospective subjects would need to include a justification for the prospective subjects - all subjects on or after January 2, 2009 - and not just the existing data/specimens. See the section below on waivers for research involving prospective enrollment of subjects provides information about justifying a waiver for prospective subjects.
Research Involving Prospective Enrollment of Subjects
Research that involves an intervention or interaction with a living individual or their identifiable data that will be created after the date of IRB submission is considered to be prospective. The IRB requires a thorough rationale for a waiver of consent for prospective research. Typical justifications for a waiver of consent for prospective research include Pragmatic, Ethical and Epidemiological justifications. Often, the justification for waiver involves several reasons rather than just a single reason.
- Too Many Sites to Cover. If a study involves subjects who could be seen at any of a number of CHOP sites, it could be impossible for the study team travel to all of the sites.
- Time of Patient Visits. Patients come to the ED 24 hours a day, 7 days a week. Often the study budget won't support this level of coverage.
- Inability to Identify Subjects Ahead of Time. Often, study subjects can not be identified ahead of time (like an ED patient or an emergency admission) and therefore, the study staff can't plan for coverage to obtain consent.
- Too many Subjects* If the study involves 100's or 1000's of subjects, it might be physically impossible to obtain consent.
If the research involves subjects who are gravely ill, it might not be appropriate to approach families during the course of hospitalization. If after discharge, there is a possibility that the child has died, calling the family to request consent could precipitate an emotional burden.
Probably the most important rationale is the impact of loss of subjects on the scientific validity of the study.
- Bias. Subjects who refuse to consent usually differ systematically from those that do. If enough subjects will be lost because they could not be approached in time or because they did not wish to consent, the study results will be biased. If elimination of subjects who do not consent will have an impact on study validity, then this argument should be carefully developed and explained.
- Willingness to Participate Depends on Outcome. Parents whose child has a favorable outcome may be more willing to participate in a study than those whose child did not. This can seriously bias the study results.
- Impact on Study Power. Many diseases of interest to pediatricians are relatively rare and loss of even a few subjects could affect the power of the study and therefore the validity.
- Each of the 4 elements of §46.116(d) should be explained and justified.
- An explanation for why the research is not greater than minimal risk is generally not difficult.
- In explaining why the research will not adversely affect the rights or welfare, it is important to recognize the right of individuals to determine whether or not to participate in research.
- It is important to address why a waiver would not adversely impact the subjects.
- Breach of confidentiality is frequently the principle risk for research involving a waiver request. It is therefore, appropriate to discuss the protections in place to protect subject confidentiality.
- Providing a compelling argument for why the research could not be practicably carried out is the most difficult. Practicable means possible; it does not mean convenient. If a subject is available to consent then it is usually (but not always) possible to obtain their consent.
- For example, if the subject comes to clinic on a regular basis, it would be practicable to obtain their consent, even for a retrospective review of records. It is important to show that the research could not wait for the time it would take to wait for all subjects to come to clinic (i.e., it was possible to obtain consent but the research would not be practicable).
- Examples where it might not be practicable to obtain consent: 1) retrospective research where subjects are lost to follow up or no longer seen regularly in clinic or seen at a variety of locations at infrequent intervals or (2) a prospective observational epidemiology study involving a entire practice or inpatient unit where there is a need for 100% participation to ascertain the rate of infection.
- Example where it would be practicable: records will be reviewed and then subjects will be asked to complete a questionnaire. Consent should be obtained prior to records review and not just prior to the questionnaire.
- Generally the requirement to provide feedback to participants is not applicable, particularly for a retrospective review of records. However, for some types of research it may be appropriate to make the study results available to the individual subjects or to the broader community from which the subjects were drawn. This could be in the form of newsletters or fliers in a doctors office to inform those who were subjects.
- If the research involves individually identifiable health information, then the investigator must also request a waiver of Written Authorization. See Waiver of HIPAA for more information.