0920-0800 SS Part B 6 27 2014

Focus Group Testing to Effectively Plan and Tailor Cancer Prevention and Control Communication Campaigns

Generic Information Collection

OMB No. 0920-0800

Request for Extension

Supporting Statement Part B

June 27, 2014

Technical Monitor:

Cynthia A. Gelb

Division of Cancer Prevention and Control

3719 North Peachtree Road | Bldg. 107 | MailStop F76

Chamblee, GA 30341

Phone: 770-488-4708

Fax: 770-488-3040

E-mail: cmg7@cdc.gov

Supported by:

Centers for Disease Control and Prevention

National Center for Chronic Disease Prevention and Health Promotion

Division of Cancer Prevention and Control

TABLE OF CONTENTS

B. Data Collection & Statistical Methods……………………………………………………..

B1. Respondent Universe……………………………………………………………

B2. Procedures for Information Collection………………………………………….

B3. Methods to Maximize Response Rates………………………………………….

B4. Tests of Procedures or Methods to be Undertaken……………………………...

B5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or

Analyzing Data…………………………………………………………………..

List of Appendices

A1. Legislative Authority: Public Health Service Act, 42 U.S.C. 241

A2. Legislative Authority: Public Health Service Act, 42 U.S.C. 247b-17

B1. 60-Day Federal Register Notice

C. Example Questions for the Focus Group Screening Form

D. Example Consent Form

E. Example Questions for the Focus Group Discussion Guide

B. DATA COLLECTION & STATISTICAL METHODS

Data collection will consist of a focus group methodology. Focus groups are widely used in stages 1 and 2 of the Health Communication Process (National Cancer Institute, 2002). In a focus group, a small group of people (typically 8-12 individuals) engage in a discussion of selected topics of interest typically directed by a moderator who guides the discussion in order to obtain the group’s opinions (Edmunds, 1999; Krueger & Casey, 2000). Focus groups capture the collective insight of a group while preserving individual preferences. In this setting, participants can describe their experiences and preferences without the limitations of preset response categories. Furthermore, focus groups produce rich data complete with nuances that often may be obscured in quantitative data collection techniques.

Statistical methods will not be employed to analyze focus group data, as it is not appropriate to report the percentage of focus group participants who expressed a particular view (Carey, 1995; Morgan, 1995; National Cancer Institute, 2002; Webb & Kevern, 2001). Typically, not every participant in a group comments on every issue discussed (Carey, 1995), and the course of discussion will vary across groups, with some topics emerging in one group and not in another (Carey, 1995; Morgan, 1995). Qualifiers such as “many,” “several,” and “few” will be used to describe the number of participants who expressed a particular view.

B1. Respondent Universe

Study participants will include members of the general public and health care professionals who are non-incarcerated, non-institutionalized adults. Additional inclusion and exclusion criteria will vary depending on the target campaign. Questions, generally drawn from the customizable recruitment screening form (Appendix C) will allow us to identify respondents with the relevant characteristics.

On average, each information collection will be conducted with 16 focus groups. The design of each communication campaign focus group will allow DCPC to collect information and to explore potential contrasts in 2-4 targeted respondent groups or audience segments. In some cases, the contrast of interest will relate to race, ethnicity, and/or language; in other cases, the contrast may relate to the knowledge, attitudes and beliefs of health care providers versus the knowledge, attitudes or beliefs of the general population or a specific segment of the general population. To identify potential variation according to regional differences, each information collection will be conducted at multiple sites in the U.S.

Typically, three to four rounds of information collections will be needed to successfully develop new materials and/or evaluate existing communication campaigns as they progress through the iterative stages of the Health Communication Process. Each information collection will be based on a distinct focus group discussion guide (generally drawn from the reference set of example items, Appendix E) and, where applicable, a distinct version of the communication campaign materials under development. It is anticipated that each information collection request, using focus groups testing, would require an average of 16 focus groups, and that two such requests will be made in each calendar year:

4 U.S. sites X 2 audience segments X 4 rounds in support of the Health Communication Process = 32 FG

DCPC plans to conduct focus groups for an average of two communication campaigns per year (32 X 2= 64 groups). In addition to the planned groups for existing DCPC campaigns, we anticipate that other communication efforts in DCPC may also require focus group research. To account for this, we are adding 20% to our focus group average (20% of 64=16) to accommodate this need. The estimated annualized burden to respondents is thus based on an average of 80 focus groups (64+16=80).

The capacity to tailor information collections to specific time-limited circumstances and maintain the ability to move rapidly among phases of the Health Communication Process are major advantages of the generic clearance format. In some cases, preliminary cancer control messages may already exist, and DCPC would begin focus group testing at a point corresponding to a later stage of the Health Communication Process. The proposed generic clearance will provide DCPC with the flexibility to conduct tailored information collection involving multiple campaigns on an as-needed basis.

B2. Procedures for Information Collection

In order to elicit focus group responses to effectively plan and/or tailor existing DCPC communication campaigns, the following steps will occur:

1. Participants, either members of the general public or health care providers, will be identified and recruited from a variety of geographic regions (e.g. Northeast, South, Midwest, and West Coast) and in both large and small cities in order to collect data from a diverse group of individuals. As many as 800 respondents (up to 10 per group) will take part in focus groups each year. Participants will be recruited using proprietary databases of professional organizations (e.g. American Medical Association Masterfile®), commercial focus group companies, and other sources. Eligibility criteria will be established for all focus group participants, and potential participants will be screened using a telephone or self-administered screening form (Appendix C). Prior to conducting the individual focus groups, consent forms will be signed by all participants (Appendix D).

2. Focus group discussion, not to exceed two hours, will occur under the direction of a professionally trained moderator. A focus group discussion guide will be developed from the reference set of OMB pre-approved questions (Appendix E) and will be utilized throughout the duration of the session. The verbal discussion that ensues will be partly directed by the moderator and partly by the comments of other participants.

As all DCPC communication campaigns utilize the Health Communication Process, similar categories of questions will be used in focus groups regardless of the specific campaign being evaluated. In stage 1, individual knowledge, attitudes, behaviors, message preferences, and media preferences will be explored. In stage 2, participants’ reactions to prototype materials will be assessed. Focus group questions will essentially be the same regardless of the campaign of interest, albeit tailored to the relevant cancer type (e.g., colorectal) and behaviors (e.g., talking with a doctor about getting screened for colorectal cancer). For this reason, it is possible and appropriate to develop a reference set of example questions.

3. All stage 1 focus groups will be video- and/or audio-recorded, and a verbatim transcript will be compiled for each group. Investigators will draft a codebook that captures themes related to the discussion topics. The codebook will include definitions, examples, inclusion criteria, and exclusion criteria for each code. To refine the codebook, several members of the study team will independently code the same focus group transcript, compare their application of codes, and reconcile coding discrepancies. During this test, the codebook continually will be revised and expanded. Using the final codebook, at least two analysts will code the remaining focus group transcripts, and intercoder reliability will be assessed.

Stage 2 focus groups will be video-and/or audio-taped for note-taking purposes. All focus group tapes will be kept in a locked cabinet and will be erased at the conclusion of 24 months following the focus group. Materials tested in stage 2 focus groups will be revised immediately, based on participants’ feedback, and the newly revised materials will be tested in the next focus group. To produce the immediate analysis required to support this process, three or more investigators will monitor each focus group and take extensive notes. The study team will hold a debriefing meeting after each group to discuss results and establish consensus among the observers.

B3. Methods to Maximize Response Rates

Participants will be recruited from sources which offer an abundant supply of the target audience. In the past, physician participants have been recruited from a sample of the American Medical Association (AMA) Masterfile®, which includes all licensed U.S. physicians, both AMA members and nonmembers. Lay participants are generally recruited from the database of the commercial research facilities where the groups are held.

To minimize the possibility of having too few appropriate focus group participants (thereby forcing group cancellation), as many as 25% more participants are invited to each group than are needed. In the event that too many participants report, excess participants will receive the honorarium and will be dismissed.

B4.Tests of Procedures or Methods to be Undertaken

All DCPC communication campaigns are guided by the Health Communication Process (National Cancer Institute, 2002) which involves four stages: (stage 1) planning and strategy development; (stage 2) developing and pretesting concepts, messages, and materials; (stage 3) implementing the program; and (stage 4) assessing effectiveness and making refinements. The Health Communication Process is not linear, but rather is a circular model in which stages are revisited in a continuous loop of planning, development, implementation, and refinement. DCPC campaign staff carefully record all aspects of campaign development, operation, and evaluation. Innovations and improvements are incorporated into subsequent campaign cycles and periodically published in the peer-review literature (Cooper, et al., 2005; Cooper et al., 2011). In fact, the use of focus group methodology to inform the development and refinement of communication campaigns has been well documented throughout the literature (Bull, et al., 2002; Edmunds, 1999; Jorgensen, et al., 2001; Krueger, 1994; Krueger & Casey, 2000; Wong, et al, 2004; Cooper et al., 2011). Thus, the formative and materials-testing methods currently used by DCPC campaigns have been refined in twelve years of campaign operations.

B5. Individuals Consulted on Statistical Aspects and Individuals Collecting and/or Analyzing Data

The proposed protocol and reference set of example questions were developed and reviewed extensively by DCPC staff, an external consultant, and Ogilvy Public Relations staff identified below. DCPC and Ogilvy Public Relations staff will participate in the analysis of the data, campaign planning and/or material refinement, as well as development of scientific manuscripts.

Cynthia A. Gelb, BSJ

Division of Cancer Prevention and Control

Centers for Disease Control and Prevention

3719 North Peachtree Road, Bldg. 107, MailStop F76

Chamblee, GA 30341

Sherri Stewart, PhD

Division of Cancer Prevention and Control

Centers for Disease Control and Prevention

3719 North Peachtree Road, Bldg. 107

Chamblee, GA 30341

Phone: (770) 488-4616

awk5@cdc.gov

Nikki Hawkins, PhD

Division of Cancer Prevention and Control

Centers for Disease Control and Prevention

3719 North Peachtree Road, Bldg. 107

Chamblee, GA 30341

Phone: (770) 488-4229

cyt4@cdc.gov

Crystale Purvis Cooper, PhD

Soltera Center for Health Communication Research

9566 N. Placita Roca de Bronce

Oro Valley, AZ 85704

Phone: (520)797-1392

Crystale_cooper@comcast.net

Jennifer Chu, MPH

Ogilvy Public Relations

1111 19^th Street, NW, Tenth Floor

Washington, DC 20036

Phone: (202) 729-4157

Jennifer.Chu@ogilvy.com