Do Quality Assurance/ Quality Improvement (QA/QI) projects that deal with human subjects or data regarding personal information need to be reviewed by CUHSR?

The area of QA/QI and human subjects review has been controversial and unclear for many institutions. The confusion arises because QA/QI activities and research activities both use scientific methods, however, most QA/QI projects do not fall squarely into the definition of human subject research as established by the federal regulations thus making it technically exempt from review by an IRB. QA/QI projects are typically management activities collecting and analyzing data to assess some internal operation for monitoring and improvement purposes. These activities typically do not attempt to be systematic about subject selection related to a population. Nor do QA/QI projects try to make broad inferences (generalize) beyond the scope of the institution. Thus QA/QI typically falls outside the regulatory definition of research which states: “Research means a systematic investigation, including research development, testing, and evaluation, designed to develop or contribute to generalizable knowledge.” 45CFR46.102(l). Research tends to be an investigation with a question or hypothesis driving the activity. The answer to the question is typically interpreted broadly and inferences are made beyond the sample.

Many QA/QI projects involve interactions or interventions with human subjects. Though a project may not be technically “research,” it might involve human subjects. An institution, such as Bradley University, still has a duty in assuring human protections and follows certain ethical principles and guidelines as it applies to all investigations involving human subjects with interactions, interventions or collecting personal information. Bradley University is committed to follow the guidelines of the Belmont report. The three ethical concepts in the Belmont report are 1) respect for persons, 2) beneficence, and 3) justice. Thus, in the application of these principles, protocols are carefully considered with regard to informed consent, privacy of personal information, risk benefit assessment, and selection of subjects. These principles should be applied to all investigations involving human subjects even if that project technically is exempt or falls outside of the technical definitions of the federal regulations. QA/QI projects involving human subjects with interactions or interventions or collecting personally identifiable information (and biospecimens) should follow ethical principles. The question remains whether these projects should be reviewed by CUHSR.

What types of QA/QI projects DO NOT need CUHSR review?

1. A project that is not research (by the federal definition) and does not involve human interaction, intervention or the collection of personal information does not need a review. An example would be if an individual is looking at de-identified data for the purpose of assessing an internal process or quality factor. The only caveat would be in a setting where there are additional privacy laws and procedures such as in health care and higher education. The project leader may have to pass their project through a local privacy board or privacy official.
2. A project that is not research (by the federal definition) but involves a benign human interaction would not need a review. For example, an anonymous, voluntary survey assessing the effectiveness of a single workshop within the institution would typically not need a review. Anonymous surveys done by students as a class project within the institution would typically not need to be reviewed, as long as they are not collecting personal and sensitive information, or as long as they would not consider inferring the results of their data analysis broadly. Likewise, a teacher collecting anonymous information on a class to assess their teaching effectiveness would not need a review. By design the risk is negligible and the focus is internal with an intent to benefit those who are participating.
3. A project that is not research (by the federal definition) but involves a human interaction, but is carried out by individuals who are familiar with ethical standards would not need a review. A consultant conducting surveys for an employer might fall into this group. Though the methodology may look like high quality research, the intent is never to share the results outside the confines of that specific group. The intent is to improve a program, practice or process of a specific institution. The intent is not to contribute to generalizable knowledge. Customer satisfaction surveys might fall into this category as well. There is an expectation that the survey would be conducted using ethical standards such as it being confidential, voluntary and involves a consent process. In this case like this, CUHSR is willing to give guidance on ethical conduct if needed.

What types of QA/QI project need a review?

Technically, if it is clearly a QA/QI project, it does not need an IRB review according to the federal regulations. However, there are cases where we as an institution would highly recommended and in some cases, require that project be reviewed. When in doubt, it is always good to fill out an application and have your project reviewed.

1. If the project collects personal information that is sensitive or that a breach in confidentiality might put the subjects at risk, then a review should occur. For example, a survey asking about employee behavior could cause reputational damage should the information be shared. A survey asking students about sexual abuse on campus would be considered sensitive and thus would need a CUSHR review.
2. If the project introduces an intervention that could potentially introduce some risk might need a review. For example, a project leader may want to introduce a new behavioral modification system for agitated patients with dementia and assess its effectiveness within the institution. This type of project potentially has several ethical pitfalls such as a) does the intervention have sound scientific background, b) how will the participants give consent considering that they may be impaired in decision making, and c) how will the confidentiality of the data be maintained. Having it reviewed could help avoid any ethical violations. It is a good practice to submit an application for review for any QA/QI project that involves an intervention of some sort.
3. Some projects may need to be reviewed if they involve protected health information. Even if it is QA/QI, HIPAA standards need to be followed. It may take a review in order to determine that no HIPAA regulations are being broken.
4. If your QA/QI project is funded, the funding source may require a human subject review and approval before it can move forward. The breath of the committee charge will allow CUHSR to approve it as “Not human subject research.”
5. The Scholarship of Teaching and Learning (SOTL) research need a review. A SOTL activity might be looking an internal assessment process or a brief intervention in a classroom with the intent of improving teaching practice. However, the ordinary intent of SOTL research is to disseminate the findings in order to contribute to evidence based practice of teaching in general.

What types “research” projects DO NOT need a review?

If a project meets the definition of research and human subject interaction by the federal guidelines, then it need an IRB review. Occasionally projects are set up like research (hypothesis driven) yet there is clearly not an intent to share the results with the purpose of contributing to generalizable knowledge. The case where this happens the most is in undergraduate research classes. As part of the class, the students may be asked to conduct “research” that involves human subjects in a low risk activity, but there is not an intent that the results are disseminated beyond that class (or department). In this case, the instructor is the one responsible that the projects are carried out following ethical principles. CUHSR is more than willing to assist instructors in understanding human subject’s review process. All our review procedures and forms are available to use by the instructor and class in order to simulate an actual review. Ethical training (CITI) is available for anyone on campus. If the projects are being disseminated beyond the department, including the Bradley Student Expo, then a CUHSR review needs to occur prior to collecting data on human subjects.

What if I am not sure if I have a QA/QI project or if it is research?

Often the intent of your project is what determines whether it is research or not. So, if you are assessing an existing process with no intent to share the results broadly and with the intent to assess or improve the process, then this is clearly a QA/QI project. In this case there is no research question or hypothesis and no intent to contribute to generalizable knowledge. In doing a similar project, you might assess a process with the intent of determining the variables that affect that process, so that this could be shared broadly to contribute to the general knowledge of that process. This has the appearance of research and needs to be reviewed if it involves human subjects. When the purpose of the project is stated on the application, research intent should be clear in that brief statement. When there is a doubt, you should submit an application to CUHSR.

Can I publish or present my QA/QI project?

Yes, you can – but please note, if you are doing QA/QI it would be unlawful to refer to it as research as this implies that it meets the federal definition of research as was approved by an IRB under the federal guidelines. Intent to disseminate results is not the criteria for something to be consider research and thus require a prospective review. Some QA/QI projects are worthy of dissemination broadly, but if they are disseminated, the authors should state clearly the intent as QA/QI and not research. If your ultimate intent is to publish or have your work disseminated, we might recommend that you adapt your project to reflect research and have it reviewed. Many journals will not accept a paper unless there was proof that it was approved by an IRB.

Once I collect data on a QA/QI project that was not approved by CUHSR, can I go back and get approval of the project as research?

Once an investigation is conducted and the data is collected under QA/QI an investigator cannot change the status of their study by getting it approved as research by an IRB. The federal regulations make no provision for a retrospective review. The regulations are clear that approval of research is prospective. CUHSR recommends that you think that you might have a QA/QI project that many have results that could make a contribution to generalizable knowledge, then the investigator should seek IRB approval of the project as research before it commences. In some cases, a project might have a dual purpose to evaluate an internal process and evaluate the result in a broader context. If this is the case then the project should be reviewed prospectively as research.

Some QA/QI projects may involve personal information, such as a GPAs, or salaries, however if that information is collected in such a way that the data cannot be tracked to an identifiable person and the project does not involve interaction with individuals, then it would not necessarily need a review. This scenario may be another mater in areas where there are additional protections on personal data, such as in health care (HIPAA or FERPA). In a health care setting or an educational setting, a review should occur so there are assurances that no HIPAA or FERPA violations have occurred. When in doubt, the investigator should contact the CUHSR chair or their immediate supervisor.

The scope and scale of the QA/QI project may also dictate the need for a review. An anonymous, voluntary survey assessing the effectiveness of a single workshop within the institution would typically not need a review. Anonymous surveys done by students as a class project within the institution would typically not need to be reviewed, as long as they are not collecting personal and sensitive information, nor would consider inferring that data broadly. However, if the intent of the survey research is to make broad inferences about the effectiveness of an intervention, such as a workshop, then this should be reviewed. The investigator should ask question of intent of the project. For example, the investigator might ask “In using the survey, was my workshop (intervention) effective?” This probably does not need a review. However, if the question is “Does this type of workshop (intervention) create a change in behavior that I might infer to similar interventions?”, then this should be reviewed, and probably reviewed as research and not quality assurance. The intent often dictates the need for a review and the determination if it is research or not.

The type of interaction or intervention will also dictate the need for a review. As previously mentioned, an anonymous survey with the intent to look at an internal process and that does not collect sensitive or personal information may not need a review. However, if you survey the effectiveness of an intervention and that intervention has some risk or the population was inherently at risk, then is should be reviewed. For example, if you are a Bradley student and want to determine the effectiveness of community exercise program that brings in people who may not ordinarily exercise, then this should be reviewed because these people may be at risk of harm from your intervention. This should be reviewed even if your intent is not to broadly interpret the results. Also, if your survey has sensitive questions that might disturb or upset the participants, or if a breach of confidentiality would bring harm, then it should be reviewed.