April 18, 2016

By Dan Gadke & Kasee Stratton, Mississippi State University

As a researcher, there is a strong likelihood that you are familiar with the Institutional Review Board (IRB) and your University’s Office of Research Compliance (ORC). There is no doubt you are up to date on your institutional training requirements, whether they be from the Collaborative Institutional Training Initiative (CITI; www.citiprogram.org) or a program your university has put together. And, you are well versed in the importance of the IRB for the protection of potential participants. Each time you design a study, you make sure to follow the appropriate IRB submission protocols and wait for approval prior to beginning the project, which is exactly what you are supposed to do—at least in most cases.

That being said, many folks are unaware or unfamiliar with other IRB options outside this most common submission procedure; however, there are two other options that are of particular use to school psychology faculty: (1) the use of archival data sets and (2) the development of a memorandum of research understanding with your ORC. While both of these may look slightly different at your respective institutions, we provide our experience as an example. The ORC at Mississippi State University has full accreditation by the Association for the Accreditation of Human Research Protection Programs (AAHRPP; http://www.aahrpp.org), following the most rigorous of standards for ethics, quality, and protection of human subjects.

Archival Data

From a bare bones perspective, archival data is exactly what it sounds like, existing data. These data may have originally been collected with no intention what so ever of being used for research purposes. Given our field is driven on data-based decision making and treatment questions are often analogous to research questions (e.g., which intervention is most effective for this child?), it is no surprise we might have data lying around that was not originally intended for research. We have found valuable use of archival data sets in school and clinic settings.

Schools. Often we have graduate students out in the school setting for practicum and other field-based experiences conducting evidenced-based assessments and interventions. At our institution, when students are on practicum they are required to complete what we refer to as a data-based project in either academic or behavioral intervention, assessment, or consultation. Students are expected to provide services under a best practice model, which often involves measuring treatment integrity, interobserver agreement, comparing interventions to choose the most effective and efficient one (e.g., experimental analyses), and so on. In some cases, these projects, which were originally designed for training purposes related to service delivery, yield novel and unique data that are dissemination-worthy, particularly when we think about presenting (i.e., poster or paper) or publishing in the world of research-to-practice.

When data are generated in the school setting during routine practicum training requirements that you decide you would like to use for dissemination, it is time to prepare an IRB application for the use of archival data. At Mississippi State University, we have an approved “data agreement form” we worked to create with our ORC for these specific instances. In short, the data agreement form is used to obtain parent or administrator consent for data collected on-site might be used for scholarly dissemination. If it is used for these purposes, the data are di-identified (i.e., stripped of any direct identifiers) and the university ORC reviews the use of the dataset prior to dissemination. When our students are working with a child in the schools, we often send the form to their parents/guardians to sign in the event that their child’s data are ever pursued for research use. Additionally, a school administrator signs the form and agrees to release a de-identified (i.e., no direct identifiers) version of the dataset to us. After supplying the ORC with the IRB write-up, the data agreement form(s), and evidence of de-identified data, we are often then able to use those data for presentations or publications

Clinic. Like many other programs, at Mississippi State University we have our School Psychology Services Clinic (SPSC), to provide our students with additional training opportunities. At the SPSC we provide supervised intervention (i.e., academic and behavioral) and assessment services. These include individual services, groups, and summer camps. Similar to what was outlined in the school setting section above, the data collected during treatment may be dissemination worthy. In these cases, we follow the same protocol mentioned before, without an administrator. Parents are asked to sign a data agreement form during the intake appointment. That being said, our MOU (described below) provides an exception for data collected from 3 or fewer participants in our clinic.

Other. There are large existing data sets available through different organizations (e.g. NCES) or even your university that may be of interest to you. For example, we wanted to evaluate college satisfaction differences between college students who had a registered disability on campus to those who did not. Many university exit surveys conducted through the Office of University Effectiveness (or a similar entity) ask students these varied questions (i.e., do you have a register disability with the university and satisfaction related questions). Many of these surveys also have a question(s) indicating if their data can be used for research purposes in the future. In this situation, we had university administrators sign the data agreement form and we completed an IRB application. After receiving approval, we requested the de-identified data sets be sent to us. In a single day, we had approximately 14,000 data-points from the previous five years to analyze.

Key points on use of school-based archival data:

• Great option for data collected in schools and clinic but not originally intended as a research project.
• Great option for large existing data sets.
• Make sure to have a data agreement form providing permission for data to potentially be used in the future.
• Make sure to complete an IRB application as soon as you decide to use data for dissemination purposes.
• Meet with your ORC representative to discuss how to best use archival data sets from your institution.

Memorandum of Understanding

A memorandum of understand (MOU), which may be called something entirely different by your ORC, is essentially an agreement that outlines the parameters of research activities allowed in the absence of an IRB application submission. MOUs are particularly useful for individuals conducting single-subject, clinic-based research. We modeled our MOU after Johns Hopkins Medicine IRB policy and their Behavioral Psychology Department (for their full policies, see here).

Our MOU has several key features, including:

• Clear overview and break down of the services provided at our clinic (i.e., School Psychology Service Center; SPSC).
• Clear policy on research and publication of data
  • All research requires Human Research Protection Program (HRPP) approval
  • Boundaries between research and practice
• Guidelines for use and publication of clinical data
  • Defining “clinical data”
  • Defining “single case report” (i.e., clinical data collected from 3 or fewer participants)
  • Defining “case series” (i.e., clinical data from more than 3 participants)
• Reporting requirements to the HRPP

Essentially, our MOU states that as long as we follow all procedures as outlined in the MOU, when using data from 3 or fewer participants, we do not need to submit for HRPP review; however, bi-annually, we need to report all projects disseminated this way. If a project has more than three participants, than we have two options: (a) if it was a pre-planned research project, submit IRB first as you typically would; (b) if it is archival clinical data, submit an archival data IRB request. In all instances, regardless of the number of participants, data need to be consented for use using the approved data agreement form mentioned under archival data.

While the MOU is particularly handy for disseminating single-subject design data involving 3 or fewer participants, it also makes all other IRB (i.e., traditional and archival) applications easier because the ORC had a clear understanding of your procedures.

MOU Use Examples

Single Case Report – 3 or fewer participants not requiring HRPP review. You have 3 children come to your clinic, all needing reading interventions. You have parents/guardians complete the data agreement form prior to beginning treatment.  You conduct a brief experimental analysis comparing the effectiveness and efficiency of a non-electronic intervention to an iPad app. You find the non-electronic intervention is more useful than the iPad app. You deliver the non-electronic intervention across several weeks, alternating its use with a control to ensure long-term effectiveness. In the end, you realize you have interesting data that are unlike anything currently in the literature. As long as all procedures were followed as outlined in the MOU, you have consent to use the data, and you report this project in the bi-annual report to the ORC, you can disseminate the data without HRPP review.

Case Series – more than three participants. You do the same intervention outlined in the single case report above, but it turns out you have 5 clients the intervention worked for. You decide after finishing the intervention that the data are worthy of dissemination, you must submit for IRB review of archival data.

Thus, with some advanced planning, researchers can put into place procedures that allow for efficient segueing class, fieldwork, and clinic activities into research publications and presentations (see a list of the scholarship that has resulted from these efforts). This is particularly valuable where practices are implemented systematically as part of training and lend themselves to analysis of effectiveness. This also presents an ongoing opportunity to integrate teaching and research activities and model a scientist-practitioner orientation to professional development and teaching for our students. We hope that you find this information useful when working with your ORC and IRB.  If you are interested in seeing an example of our MOU, please email us and we would be happy to share.