Which projects require IRB Review?

The following types of projects require a full IRB review. The Expedited or Full IRB Review form should be used. These reviews are done every other month, and are conducted by the full IRB board. Annual reports and a final report are required.

1. Collecting data from anyone under the age of 18.
   
   
2. Asking vulnerable populations about their health, well-being, or family dynamics. Survey or other data collection in which we're contacting participants and asking them about their health, well-being or family dynamics.
   
   Example: A study involves a survey of pregnant women to obtain information about their use of a visiting nurse program, and the health of mother and baby. Since this study is directly contacting a vulnerable population (pregnant women) and asking them about their health, it requires full IRB review.
   
   
3. Collecting data from vulnerable populations in a setting where others will hear the questions and answers (e.g. focus group).
   
   Example: Evaluators would like to conduct a focus group with participants of a program for people with mental illness. Since this study collects data from a vulnerable population (people with mental illness) in a setting where others will overhear questions and answers, it requires full IRB review.
   
   
4. Asking participants about highly sensitive information. By highly sensitive the IRB means a topic which might place them at risk of criminal or civil liability or might be damaging to their financial standing, employability, or reputation. Personally identifiable data on a person's chemical or mental health or child protection involvement is always considered to be highly sensitive. If you are unsure whether a type of data is highly sensitive, please contact the IRB administrator.
   
   Example: Interviews will be conducted with participants in a cash assistance program designed to gain information on all sources of family income, including sources which may not be legal or ethical. The data will be used to better understand the family's finances and to improve communication between financial workers and participants. The interview instrument will not identify the respondent, and the data will only be linked to the family using a randomly assigned ID number, a crosswalk of which will be kept in a separate location. Despite the data security precautions, given the highly sensitive nature of some of the data, this entire project will require full IRB review.
   
   
5. A project which includes the performance of a diagnostic or therapeutic intervention as part of the study design.
   
   
6. A project in which participants receive different levels of services, or type of services, and where participants are placed into different service categories NOT based on their eligibility criteria, but based on random assignment or by another scientific method which strives to create comparable groups for the purposes of analysis.

The following types of projects will receive an expedited IRB review. The Expedited or Full IRB Review form should be used. These reviews are done every month, and are conducted by a subgroup of the IRB or an experienced Board member as assigned by the IRB Chair. Annual reports, updates on significant changes, and a final report are required.

1. Sharing identifiable data outside of DHS. We will be sharing existing data with an outside entity, and we will be including identifiers such as name or RecipientID. A data sharing agreement is in place. However, if the data is highly sensitive data, it requires a full IRB review. Expedited review is not required if we are sharing identifiable data with a vendor who is collecting or analyzing data solely for DHS purposes. However, IRB review may be necessary due to other aspects of the study (e.g. if a survey is conducted with a vulnerable population).
   
   Example: DHS data is being requested by an external researcher. The data relates to utilization of dental and orthodontics care. Although no identifiers such as name or SSN is being requested, very specific indicators of the participants' geographic, racial, age, and family structure characteristics are being requested, which may result in some individuals being identifiable. The researcher has a data sharing agreement with the affected division, which is sponsoring the research. The data is not considered to be highly sensitive. The study should come to the IRB, and would be accepted for expedited review.
   
   
2. Asking non-vulnerable populations about their health, well-being, or family dynamics. Survey or other data collection in which we're contacting participants and asking them about their health, well-being or family dynamics.
   
   Example: A job training program which does not target a particular vulnerable population conducts a survey of participants. They ask about participants' satisfaction with the program as well as what kind of impact the program had on them. This would not require any IRB review. However, researchers also ask open-ended questions soliciting information on how the program has impacted participants' emotional and financial well-being. The study should come to the IRB, and would be accepted for expedited review.
   
   
3. Asking vulnerable populations about the impact of the program. Survey or other data collection in which we're contacting participants and asking them about the impact of the program, or asking whether the program met their needs. If there are questions relating to health, well-being, or family dynamics on a vulnerable population, this requires a full IRB review.
   
   Example: An employment program for people with disabilities would like to conduct a continuous improvement initiative using weekly data collection. They plan to hand out short 'Tell us how we're doing' feedback forms to participants, and ask them to fill out the open-ended questions about how the program is helping them, and how the program could do better to meet their needs. The study should come to the IRB, and would be accepted for expedited review.
   
   
4. DHS analysis of existing data, findings intended to be generalized beyond the study population. We will not be contacting recipients as part of this project, and will only be analyzing existing DHS data. No individual-level data will leave DHS.
   
   Most internal evaluation projects where there is no participant contact, and no additional data security risk (as no data is leaving DHS), do not require IRB approval. However, if the study is designed to be generalizable to populations other than the one studied, federal regulations require IRB review. The one-time review meets this requirement. If the study is not originally intended to be generalizable research, but staff later decide to consider it so, a one-time IRB proposal can be submitted at that time.
   
   Example: A DHS division is evaluating an initiative to reduce unnecessary Emergency Department visits using Medicaid claims and enrollment data. The staff plan to disseminate their findings to a wide health care audience, including people who are likely to apply the findings to different populations than the one studied (e.g. populations other than Medicaid participants). A one-time review is required.
   
   
5. Sharing individual-level data which is not identifiable with an organization outside of DHS. We will not include identifiers such as name or RecipientID, and will not include indicators which in combination could identify individuals. A data sharing agreement is in place. The study is not being conducted solely for DHS purposes.
   
   Example: A professor at a local university is investigating whether continuity of Medicaid enrollment in parents is related to their children's use of preventive health care. Health care staff provide the professor with an individual-level dataset with no identifiers in which no combination of indicators together can identify anyone. A one-time IRB review is required.

Studies which do not require IRB review.

Four types of projects do not require review by the DHS IRB. If your research or evaluation project fits all criteria for any of these types, you do not need to submit a proposal to the IRB.

Internal analysis which uses existing administrative data only

1. We will analyze existing administrative data only (e.g. enrollment, program data).
   
   
2. The dataset could include identifiers such as RecipientID or Name, or all identifiers could have been deleted from the dataset.
   
   
3. We won't be contacting clients or consumers to get information about them (e.g. we won't send out paper surveys, or conduct focus groups).
   
   
4. We will not share individual-level data outside of DHS or other state agencies.
   
   
5. Our investigation is not designed to contribute to generalizable knowledge. (We do not plan to use the findings to inform decisions on populations other than the population we studied.)
   
   Example: A DHS staff is using Medicaid claims and enrollment data to identify the regions in Minnesota with the greatest prevalence of babies born with a diagnosis of neonatal opioid exposure. This will be used to encourage the development of programs in those regions to reduce exposure in future newborns, and will not be generalized to populations outside of Minnesota or outside of the Medicaid population. This study does not require IRB review.
   
   At the conclusion of the study, the analysts feel that the findings are especially interesting and they decide to publish in a national medical journal. At that time, they will need to submit an IRB proposal (one-time) because it is now generalizable research.

If your project meets all five criteria above, it is not necessary to submit a proposal or any documentation to the IRB.

Survey of service providers or professionals, including those requesting client data

1. We will solicit information from service providers or professionals in a survey, focus group, or some other data collection method.
   
   
2. We will ask them about program quality, or ask them to report something about participants.
   
   
3. We will not ask providers about their own health, well-being, or family dynamics.
   
   
4. Our investigation is not designed to contribute to generalizable knowledge. (We do not plan to use the findings to inform decisions on populations other than the population we studied.)
   
   Example: A survey of residential managers of supportive housing sites asks them to report particular residents' length of stay, the resident's age, racial group, and gender. This will be used to look at disparities in services. No questions are asked about the residential managers themselves. This study does not require IRB review.

If your project meets all four criteria above, it is not necessary to submit a proposal or any documentation to the IRB.

Sharing existing data outside of DHS solely for DHS purposes.

1. We will share existing participant data with a vendor who is collecting or analyzing data solely for DHS purposes. The data sharing can include identifiable data.
   
   
2. A data sharing agreement is in place.
   
   
3. Neither we, nor the vendor, will be collecting data from participants.
   
   
4. Our investigation is not designed to contribute to generalizable knowledge. (We do not plan to use the findings to inform decisions on populations other than the population we studied.)
   
   Example: A DHS division needs to determine the potential costs associated with a proposed payment rate change for waiver services. They send their vendor a list of recipients, their providers, services they have received, and those services' costs. The vendor analyzes the information and reports on costs that would result from the proposed change. The analysis is strictly for DHS' need to determine costs, and is not generalizable. This study does not require IRB review.

If your project meets all criteria above, it is not necessary to submit a proposal or any documentation to the IRB.

Survey or other data collection from program participants, questions are limited in scope

Two types of data collection with participants requires no IRB review. The first type is a survey which asks participants questions directly related to the program only, such as questions about the program's accessibility. The second type is a survey which has questions on how the program affected participants, but this is only outside IRB purview if this survey is conducted with non-vulnerable populations. Please review the next section on surveys if you are conducting a survey with DHS participants.

The grid below shows which types of questions require an IRB review, and which do not. The determination depends on whether or not the DHS IRB considers the participants you will be surveying to be part of a vulnerable population.

Which participants are considered vulnerable? Vulnerable populations include: people who are mentally ill, chemically dependent, disabled, cognitively impaired, or pregnant. Vulnerable populations also include participants of mandatory programs (e.g. child protective services) and people in institutions (e.g. correctional facilities, residential treatment, nursing homes, group homes, foster care).

If participants are receiving the services, program, or are participating in the study because they are members of one of these categories, then they are considered vulnerable. For example, a study of prenatal service use would only include pregnant women and thus the study involves a vulnerable population. In contrast, some participants of a cash assistance program may be pregnant, but this is not considered a vulnerable population unless their pregnancy is a criteria for their participation.

If anyone under the age of 18 is to be contacted for a survey or another form of data collection, a full IRB review is necessary, regardless of the questions asked.

Survey topic areas	Non-vulnerable Participants	Vulnerable Participants	Participants under age 18
Questions about the program itself Participants' satisfaction with a program Participants' views on program accessibility	No need for IRB review	No need for IRB review	Full IRB review
Questions about how the program affected participants (except personal topics below) Did the program meet participants' needs or goals Were the program's goals for the participant met	No need for IRB review	Expedited IRB review	Full IRB review
Questions about personal or protected topics Participants' health Participant well-being Family dynamics	Expedited IRB review	Full IRB review	Full IRB review

If your survey does not require IRB review due to the survey grid above, it will still require review if it meets any of the following criteria:

• Our investigation is designed to contribute to generalizable knowledge gr. (We plan to use the findings to inform decisions on populations other than the population we studied.) This type of project will require an expedited or full IRB review.
• Some of the questions are highly sensitive. This would requires a full IRB review.
• You will be collecting any type of data from vulnerable population in a setting where others will hear the questions and answers.
• Identifiable DHS data is being shared with an outside entity, and the entity is not a vendor conducting the work for DHS purposes. The project will require an expedited or full IRB review.

Survey example: A survey is planned for low-income people who attended a job skills training program. This training is not specifically targeting people with disabilities or any other vulnerable population. Questions might include accessibility (was the location and timing appropriate), satisfaction (was the content of the training what they needed), and program goals (did their final resumes look professional, did they get any interviews, did they get a job) and because this is a non-vulnerable population, this project will not require IRB review.

However, this project will require IRB expedited review if the survey asks about their health, well-being, or family dynamics (does the participant feel better about her/himself, have reduced family conflict).