Common Problems Encountered by the IRB in Reviewing in Research Protocols
Listed below are ten problems commonly encountered in protocols submitted for full-board review during the past 18 months. In each case, a recommended solution follows the statement of the problem.
- Frequently, researchers fail to state the consequences for subjects who might decide not to participate in the study. This is crucial if you are providing a service to the study population. If research is providing any kind of service be sure to specify what happens to a participant who refuses to be part of the research. Will you still provide the service to anyone who refuses to participate? Are the proposed services available elsewhere? What steps have you taken to insure that those who decide to not participate are not made to feel uncomfortable.
- There have been cases in which it was unclear as to whether the members of the research team have the expertise to conduct the proposed research. Two basic questions have arisen: What are the credentials of the research personnel and who will be taking what actions in conducting the study? This last question is frequently of concern when there are co-investigators some of whom are students. Throughout your test be sure to clearly identify who will be doing what and specify their qualifications. For example, if a licensed psychologist will be present during observed therapy sessions clearly identify person and state that she or he is licensed. Also, if you use pronouns in your text be sure that the referent is clearly identified.
- Reviewers have been concerned with collections of sensitive data without the appropriate precautions specified. If you are dealing with confidential data, specify how confidentiality will be maintained throughout the research process. If employees will be handling data specify how they will be trained. Clearly identify everyone who will handle information and describe the process they will use and their qualifications for implementing the process in the manner proposed.
- Reviewers give consent forms particular attention and many problems have been encountered. Make sure that your consent form is consistent with your protocol. If possible have someone proof your consent form, checking for consistency with your protocol and for grammatical or spelling errors. Child assent forms should be written at the reading level of the children in the research population. And finally, you may need institutional consent from authorities responsible for the research site, e.g., hospitals, schools.
- Children are defined as under the age of 19 by HHS which technically makes 18 year-olds a vulnerable population. If you are sampling from the general population, you may want to set your bottom age range at 19.
- Reviewers have been uncertain to what is being proposed. It helps if you clearly state your design, specify procedures, and provide an operationalization of your concepts. This is particularly helpful if your independent variable is an intervention. In describing your procedures, you should use terminology that is most common between disciplines. For example, if you are planning to conduct an experiment you should specify your procedures; such as, “randomly assign subjects to the control or experimental groups which are...”.
- Proposals have referred to information which had not been included in the protocol. Reviewers are reluctant to approve measures which have not been seen. Include scripts, questionnaires, surveys, paper and pencil tests, and pictures of stimuli. If a stimulus is too bulky to submit, provide written detail so that we have a record which will demonstrate that the stimulus is innocuous.
- The validity of research can be called into question. Part of the Board's responsibility is to see to it that human subjects are not used frivolously. This not only protects the subjects but also protects serious researchers. Integrate the literature review into your Background section. Organizing your discussion around the operationalization of your independent and dependent variables aids in clearly expressing your research question to reviewers.
- Scales or instruments are presented with little comment. If you are using a scale that has been extensively used in your discipline please tell us. If a scale contains sensitive questions, but has found to be reliable and valid, then the reviewers are less likely to be concerned.
- Some researchers forget to inform their subjects that they have the right to withdraw from any study with no negative consequences, e.g., loss of credit or points or money if offered for participation. So as to be sure that your subjects understand that they have the right to withdraw for the research, state this explicitly in your consent form or in your instructions to subjects.