Appendix P-IRB Application

Appendix P

IRB Application

OFFICE USE:

FILE NO.________

DATE REC'D:__/__/__/

PUBLIC/PRIVATE VENTURES

Submission for Institutional Review Board Approval

PROJECT TITLE: Effectiveness of Efforts to Increase Supplemental Nutrition Assistance Program Participation among Medicare’s Extra Help Population Pilot Projects

PRINCIPAL INVESTIGATOR: Laura Castner

NAME & TITLE: Senior Researcher

DEPARTMENT: Mathematica Policy Research, Inc.

CO-INVESTIGATOR(s): Emily Sama-Miller

NAME & TITLE: Researcher

DEPT. OR ORGANIZATION: Mathematica Policy Research, Inc.

FUNDER(s): U.S. Department of Agriculture, Food & Nutritional Services

The information submitted herewith is complete and accurate to the best of my knowledge.

______________________________ _______________

PRINCIPAL INVESTIGATOR DATE

______________________________________________________________________________

INSTITUTIONAL APPROVAL:

UNIT DIRECTOR DATE___________

IRB PROVISIONAL (INITIALS)________________ DATE___________

IRB APPROVED (SIGNATURE)__________________ DATE___________

EXECUTIVE OFFICE__________________________ DATE___________

ABSTRACT OF THE PROPOSED STUDY

Currently, caseloads for the Supplemental Nutrition Assistance Program (SNAP) are at record high levels. Participation rates among the elderly, however, remain low. Prior research on the factors that cause lower participation among this group indicate that elderly clients may not apply for benefits because of difficulty in accessing accurate information about the program, trouble traveling to a SNAP office, or because of stigma associated with participation. Eligible elderly individuals may also struggle to complete the application process because of difficulty with submitting the full range of verification required in order to qualify for benefits. Three pilot projects (SNAP Extra Help programs) funded by the U.S. Department of Agriculture (USDA) Food and Nutrition Services (FNS) will address some of these challenges through three different approaches (targeted outreach, simplified eligibility criteria, and standardized SNAP benefits). A key motivation for the pilots was new legislative authorization, beginning in January 2010 for the Social Security Administration (SSA) to send Extra Help application data to state Medicaid agencies to asses an individual’s Medicare Savings Program (MSP) eligibility. The idea behind this new legislation was that an additional transfer of these data to the SNAP agency could help target low-income seniors. Based on this, FNS created the SNAP Extra Help pilot grants and awarded them to three states (Washington, Pennsylvania and New Mexico).

The Impact Evaluation Design

Mathematica Policy Research will conduct an impact evaluation consisting of a client survey, documenting the experiences, outcomes, and opinions of those individuals eligible to receive SNAP services. This will provide insight from a participant perspective into how services are received and delivered. The study will test the effectiveness of each program by using a double difference design. This will be accomplished by comparing outcomes of the SNAP Extra Help pilot sites to those of comparison sites (non-pilot sites with similar demographic compositions to pilot sites), and thus generate evidence that states and communities need about effective SNAP programs. Overall, the evaluation will be based on a sample of as many as 6000 elderly individuals eligible for SNAP. Comparison of experiences and outcomes for the program and comparison groups will indicate the effectiveness of the grantee’s pilot programs in changing enrollment rates and experiences of the elderly population.

Target Population and Sample Enrollment

This evaluation will consist of a telephone survey of client experiences with elderly (age 65+) individuals, who are eligible for SNAP. These individuals are targeted by grantees based on the SSA data transfer to state Medicaid agencies. Generally, once the data transfer occurs, pilot programs use this to identify individuals that are not currently receiving SNAP and attempt to enroll them into SNAP or both MSP and SNAP (depending on the state).

Mathematica will receive the program’s list of potential enrollees and send an advance letter. In addition to providing general information about the study, this letter will also inform them of a future attempt by Mathematica to contact via telephone to complete a telephone. Also included in the letter is a statement reminding eligible individuals that participation is voluntary. Mathematica will then administer a brief survey (~20 minutes) to those who give their verbal consent to participate over the telephone.

Data Collection

The data collection will help us control for variables (e.g. previous experience with SNAP) and measure program effectiveness. Program effectiveness will be judged by differences in key variables: previous SNAP experience, household food security, reasons for applying for SNAP, SNAP application process, knowledge of SNAP, reasons for nonparticipation, and experience with Extra Help pilot grantees. Individuals will only be asked questions relevant to their circumstances; for example, participants who say they did apply for SNAP will not be asked questions regarding a decision to not apply for SNAP. To aid in this, the survey will be conducted through a computer-assisted telephone interview (CATI). The same overall structure of outcomes will be used in all sites to maintain the consistency of the evaluation. Appendix A presents the survey instrument. Once completed, participants will receive $25 for their participants.

IRB Questions

1. PARTICIPANTS (ages, sex, numbers per year and total, source of treatment pool or records, inclusion or exclusion characteristics)

The expected enrollment in the SNAP Extra Help sites is as follows:

Expected Samples, by Type of Site (Includes Both Program and Control Groups)

Site Location	Grantee	Comparison Site	Total Individuals
Pennsylvania	1500	1500	3000
Washington	1000	1000	2000
New Mexico	500	500	1000
Total			6000

The above estimates are based on current enrollment numbers at both the comparison and pilot sites in each state. Grantee sites began enrollment in August and October, 2011.

It is expected that the sample will consist equally of males and females. The age distribution of sample members will vary but all will be age 65 or older. All participants will be low-income. Participants from the pilot and comparisons sites are expected to be similar as comparison sites were selected due to their similarity to pilot sites in terms of: (1) median income, (2) elderly poverty rate, (3) unemployment rate, (4) county size (i.e. populations and density), and (5) size of the elderly population.

2. NONPARTICIPANTS/CONTROLS (ages, sex, numbers per year and total, source of treatment pool or records, inclusion or exclusion characteristics)

See above for projected number of sample enrollments. Half of the sample will be from comparison sites, not participating in the SNAP Extra Help pilot program (n = 3000).

3. SPECIAL POPULATIONS (if applicable, name and justify the use of any special population--e.g., pregnant teens, institutionalized persons, etc.)

This study will not focus on special populations. Since SNAP Extra Help programs focus on elderly individuals, our sample will only consist of such. It is possible that some of these participants will, by chance, have a disability, but disabilities are not part of the definition of the target population. The questions asked in the survey will be adjusted to accommodate any disabilities (e.g. given via TTY for those who are deaf or hard of hearing).

4. RECRUITMENT:

Sites will conduct their own recruitment for participation in food assistance programs. Mathematica will receive the list of individuals (containing names and contact information) eligible for the pilot from each site. A secure FTP site established by either the state or Mathematica will be used. For instances where FTP may not be possible, Mathematica will request that each State send us encrypted data extracts via overnight mail on a CD. Once the data is in house, project staff will use normal corporate technological and procedural safeguards to protect the electronic data files that contain client information. Files containing Personally Identifying Information will be stored on a secure network drive accessible via a network log-on ID that is accessible only to team members who need access to such files. These files will be kept separate from other project documents and files. Data on network drives is protected by the security mechanisms of our network operating systems (Novell Netware). Both are compliant with the C2/E2 Red Book security specifications. The network is protected from unauthorized external access through a firewall from Cisco Systems, Inc. All servers with confidential information are in a controlled-access area. When staff using sensitive data step away from their desk, they will close files and applications and use a password protected screen saver, as well as logging off from the system at the end of the day. Any sensitive information printed in hard copy will be catalogued and shredded after it is used and will be stored in locked file storage during periods of ongoing use.

Mathematica will then send an advance letter on behalf of USDA to potential participants (see Appendix B for advance letter). This letter will give a brief description of the study and notify participants that a Mathematica interviewer will be contacting them shortly for their permission to participate in the telephone survey. This letter will also expressly state that participation is voluntary and will not affect any benefits they are receiving or will receive.

Once a Mathematica interviewer contacts potential participants, they will be given a more thorough overview of the study, reminded again that participation is voluntary and then asked if they consent to participate in the telephone survey.

5. INFORMED CONSENT. (Indicate yes or no, and whether consent will be written or oral. Provide justification for why consent will not be obtained or why consent will be oral.)

YES. We will obtain oral consent before beginning the telephone interview. We believe oral consent is sufficient in this study due to the lack of any experimental manipulation. Additionally, the questions asked are not of a highly sensitive nature. Questions only consist of experiences and knowledge of SNAP programs. Participants are reminded that may choose to not answer any questions or withdrawal their consent at any time without penalty.

6. ASSENT. (Indicate yes or no, and whether consent will be written or oral. Provide justification for why consent will not be obtained or why consent will be oral.)

NO. No minors are included in this study.

7. WHO WILL GIVE CONSENT. (Indicate whether consent will be given by the participants or by a legally authorized representative, and provide reason).

Consent will be given by participants.

8. TRANSLATIONS. (Indicate whether translations will be made and into which languages)

All explanatory materials and questionnaires will be translated into Spanish.

9. RISKS.

Some subjects might be uncomfortable answering some questions on the survey. Subjects will be informed that they do not have to answer any questions that make them uncomfortable.

Despite procedures to safeguard confidentiality, there is a slight risk that data on subjects might be divulged and cause a loss of confidentiality. However, careful steps are taken to prevent this. To protect subjects’ confidentiality, identifying information will be separated from subjects’ data in separate files. In data files, a study ID number is used to identify each subject. A link file is used to associate each study ID with the actual name and other identifying information on each subject, and access to this link file is restricted to study staff who need it for purposes of conducting the follow-up surveys. Access to hard-copy forms bearing subject identifying information is strictly limited. Physical precautions include use of secure areas, locked files and cabinets, and shredding of discarded materials. All employees at Mathematica Policy Research sign a confidentiality pledge.

10. BENEFITS.

There are no direct benefits for participation in the study itself, but subjects will have a chance to share their opinions and experiences regarding SNAP programs which will be used to help create better programs. The results from this evaluation will be published in reports to the federal government and scientific journals that will be accessible to the public.

11. TIME SPENT.

The survey was pilot tested and the average time to completion was 20 minutes. For those with little to no SNAP program experience the survey can be completed in under 20 minutes, whereas those with SNAP program experiences may take slightly longer (~30 minutes).

12. REIMBURSEMENT. (Indicate whether reimbursement will be made to participants, including amount and justification).

Participants will be given a $25 check as a thank you for their time participating in the survey.

13. PERSONALLY SENSITIVE QUESTIONS ON QUESTIONNAIRES/ INTERVIEWS.

Questions included in this survey are of a mild sensitivity. These questions are those which focus on why or why not participants choose to apply for SNAP program benefits. Since these potential benefits are based on income, some participants may view these questions as slightly evasive or stigmatization. Participants, however, are informed that they may chose not to answer any question which makes them uncomfortable, without penalty. Additionally, questions have been word in a way that minimizes stigma. Interviewers conducting the survey receive additional training in ways to reduce stigma in interviewing.

14. STUDY INSTRUMENTS TO BE USED. (Attach Copies)

The baseline survey questionnaire is included in Appendix A.

15. LOCATION WHERE STUDY WILL TAKE PLACE.

The pilot and comparison sites are located in New Mexico, Pennsylvania, and Washington. The telephone surveys interviewers will call participants out of Mathematica’s Princeton, NJ and Oakland, CA offices.