Supporting Statement Part B Fors Marsh 10312023

Qualitative Research on Food Safety Behaviors Among Dual Caregivers

In Depth Interview Research

OMB No. 0583–NEW

Supporting Statement

B. Statistical Methods

B.1. Respondent Universe and Sampling Methods

We will conduct two sets of interviews with adult consumers. Each series will include 15 interviews for a total of 30 interviews. Both sets will take place in FY 2024. The supporting statement describes both the first and second sets of interviews.

First and Second Set of Interviews: Food Safety Knowledge, Attitudes, Behaviors and Concept Testing

The first and second sets of 15 interviews (30 total) will take place in FY 2024 and will be conducted with African American/Black adults and first-generation Hispanic/Latino adults who are dual caregivers. All enrolled adults will be recruited via an invitation (Appendix A) and will need to have indicated via a screener (Appendix B) that they simultaneously provide support to children and/or young adults (under 18 years old) and older adults (>65 years old). Interviews will also be segmented based on ethnicity. The interviews with Hispanic/Latino populations will be further segmented based on number of years spent living in the United States (i.e., 10 or more years or fewer than 10 years) and preferred language (i.e., English or Spanish). See the table below for more information regarding enrollment in project by category.

All interviews will be held virtually and participants will represent diverse geographic locations. In each set of interviews, 10 will take place in English with English-speaking adults (i.e., African American/Black participants and Hispanic/Latino participants in US for 10 or more years) and 5 will take place in Spanish with Spanish-speaking adults (i.e., Hispanic/Latino participants in US for 10 or fewer years). Table B-1 identifies the subpopulations and languages for the interviews.

Table B.1. Enrollment and Segmentation in Interview Study

Interview Set	Subpopulation	Language	Number of Interviews
1	African American Adults	English	5
1	Hispanic/Latino (US for 10 or more years, speak English at home)	English	5
1	Hispanic/Latino (US for fewer than 10 years, speak Spanish at home)	Spanish	5
Total Interviews Set 1			15
2	African American Adults	English	5
2	Hispanic/Latino (US for 10 or more years, speak English at home)	English	5
2	Hispanic/Latino (US for fewer than 10 years, speak Spanish at home)	Spanish	5
Total Interviews Set 2			15

In addition to meeting the requirements for inclusion stated above, interview participants must meet the following eligibility criteria:

• Are adults (21+).

• Parent or guardian to minor child less than 18 years old that lives in their home.

• Provides some type of support, such as visitation, meal preparation, assistance with routine activities, phone calls, or emotional support to your parent(s) or in-law(s) at least twice a month.

• At least six participants in each interview phase will identify as male to ensure lack of homogeny in gender among participants.

• Participants will represent a diversity of geographic locations.

• Have not been employed (including immediate family members) by the federal government.

Recruitment Procedures for Phase 1 and Phase 2 of Interviews

Fors Marsh will work with a research recruitment vendor to identify potential participants for the virtual interviews. The data collection will not use statistical methods to select respondents. Instead, the recruitment vendor will use convenience sampling to recruit from their database to find potential participants who meet the eligibility criteria as defined by the screening questionnaire (see Appendix B). Recruiters at the vendor will use the questionnaire to screen and assign participants to the study. A maximum of 15 participants per group will participate in the set one interviews.

To ensure that we have a sample that meets our criteria, the research vendor will email individuals in their recruitment database who represent the inclusion criteria needed for the study. Brief screening interviews will then serve to verify demographic variables and determine interest in participation.

The databases maintained by research vendors are developed through advertisements, word-of-mouth referrals, outreach to various organizations and industries, and Internet list searches. The databases are constantly refreshed through natural drop out and additions. In addition to contact information, basic socioeconomic characteristics are collected through the database registration process and updated periodically.

B.2. Procedures for the Collection of Information

All participants who agree to take part in the interviews will be informed of the interview date, time, and Zoom link by the research vendor. Participants will also be provided with an informed consent document at this time (approved by Advarra Institutional Review Board; see Appendix C for consent form, see Appendix K for approval documentation). Once participants log on for their virtual interview, they will be asked if they’ve read the informed consent document and will be asked for verbal consent. An experienced interviewer will conduct the discussions using an interview guide (see Appendices D, E, and F for set one interview guides). The guide will serve as an outline for the interview.

The goal of the phase one interviews will be to (a) gather information about consumers’ knowledge, attitudes, and behaviors surrounding food safety, (b) learn more about preferred communication channels used by consumers when learning about food safety, and (c) gather feedback on consumers’ perspectives relating to possible themes for the campaign effort.

The goal of the phase two interviews will be to gather specific feedback about consumers’ perspectives relating to campaign materials. These conversations will involve requesting feedback on campaign statements/messages and visual elements of ready-made campaign materials that have been formulated based on the feedback provided in the phase one interviews. Each interview will be audio-recorded by Fors Marsh. A notetaker will also be present to take thorough notes on interview responses (see Appendices G, H, and I for set two interview guides). Note: All draft campaign materials developed will be submitted to OMB for review via an amendment prior to seeking participant feedback in the second set of interview.

No more than two research team members will review the interview notes and go back to original audio recordings for additional detail if necessary. Information collected from these interviews will be combined into an overall report highlighting key responses from each segment of the interview sample. The research team will identify common themes in this report and denote exceptions to those themes. Deidentified quotes will also be used in the final report to exemplify key themes and information. This information will be provided in a detailed summary to FSIS. No statistical analyses will be conducted.

Before, throughout, and after data collection, the recruitment vendor will keep the database of participants information on its own cloud (in a password protected folder). Fors Marsh will only be working with de-identified files, recordings, and notes that are de-identified. This approach will be used to make sure personally-identifiable information cannot be connected with responses. Further, all files, recordings, and notes will be kept in a password protected folder to ensure protection of information collected.

B.3. Methods to Maximize Response Rate and Deal with Nonresponse

The research vendor plans to recruit 15 individuals per ser of interviews. If a participant does not show up at the agreed-upon interview time, the research vendor will get back in touch with them to set up an alternative time to meet. If that person is not able to be reached or cannot reschedule, another person will be recruited. The expected response (show rate) among recruited participants is 83%.

To maximize response rate and address possible nonresponse, each focus group participant will receive a $75 gift card for their participation. In addition, the recruitment vendor will send confirmation emails to recruited individuals before the interview discussion.

B.4. Tests of Procedures or Methods to be Undertaken

No formal pre-testing will be undertaken on this research project. The interview guides have been reviewed and edited by multiple individuals who possess expertise in qualitative research and multiple individuals with expertise in food safety. The materials have also been created using plain language and clear communication guidance in mind.

B.5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

Carolyn Lagoe, Fors Marsh, will manage the collection and analysis of the interview data, Aaron Lavelle an FSIS employee, will review the results of the consumer interviews.

3