Effects of a Televised Two-City Safer Sex Mass Media Campaign Targeting High Sensation-Seeking and Impulsive Decision- Making Young Adults Seth M. Noar, Ph.D. Assistant Professor Department of Communication University of Kentucky
Research Team Principal Investigator: Rick S. Zimmerman, Ph.D. Co-Investigators: Seth M. Noar, Ph.D. Pamela Cupp, Ph.D. Lewis Donohew, Ph.D. Co-PI: Philip Palmgreen, Ph.D. Graduate Research Assistants: Mia Liza A. Lustria, M.S. Hung-Yi Lu, M.A. Mary Lee Horosewski, M.A.
Acknowledgements This project was funded by the National Institute of Mental Health (NIMH), grant 1-R01-MH63705.
Research Question Can a stand-alone televised mass media campaign be effective in changing safer sexual behavior?
Limitations of Previous Campaigns Lack of formative research on audience, channels, messages Lack of audience segmentation and message targeting Failure to develop theory-based campaigns Lack of rigorous evaluation
Current Study The purpose of the current study was to develop, implement, and rigorously evaluate a 2-city safer sex mass media campaign. Target population: At-risk Heterosexually Active Young Adults (aged 18-26) Medium: Televised 30-second Public Service Announcements (PSA s) (TV only) Message: Increase Condom Use Research Design: Controlled Interrupted Time Series Design with Comparison Community (Lexington, Kentucky, and Knoxville, Tennessee)
Design of Two-City Campaign DATA COLLECTION DATA COLLECTION DATA COLLECTION LEXINGTON, KY Campaign LEXINGTON, KY KNOXVILLE, TN KNOXVILLE, TN JAN DEC NOV OCT SEPT AUG JULY JUNE MAY APRIL MAR FEB JAN DEC NOV OCT SEPT AUG JULY JUNE MAY 2002 2003 04
Campaign Design / Evaluation 100 individuals surveyed per month in each city for 21 months using (RDD) Independent samples were gathered Criteria: Young adults aged 18-26, heterosexually active in past 3 months Cohort was aged 18-24 at start, 20-26 at end of study Surveyed using laptop-based self-report survey this allowed participants to actually view the campaign PSAs
Campaign Plan Message Targeting: Sensation-seeking Targeting (SENTAR) Approach Formative Research: 3 waves of in-depth focus groups (40 focus groups total) Use of Theory: Staged Approach Campaign Evaluation: Interrupted Time Series Design with Comparison Group
Message Targeting (SENTAR) Messages that appeal to HSS (and IDM) were developed and selected through formative research (KFF spots and original spots). Messages were placed in programming preferred by these young adults. When the campaign was evaluated, subgroups were specifically examined (particularly HSS/IDM)
Airing of Campaigns Worked with TV stations early on to earn trust and build rapport We asked for and achieved a 1-1 match on airtime Initial data collected allowed us to identify TV programs viewed by the target audience PSAs were inserted (by a professional media buyer) in programs based on demographics and HSS preferences Messages aired in a sequential manner: focus on threat, personal risk, benefits/consequences, skills High saturation campaigns: 200+ gross rating points (GRPs) each week of the 13 week campaign
Results
Results For all results: Focus of analysis is on those above the median on a composite HSS/IDM variable Results reported Exposure to campaign Change in sexual risk behavior Change in safer sex mediating variables
Exposure: Saw at least 1 PSA 100.0% 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Month HSSIDM Percent
Exposure: Saw 3 different PSAs Percentage of respondents in Lexington who saw at least 3 PSAs 100.00% 90.00% 80.00% 70.00% Percent 60.00% 50.00% 40.00% LoSSDM HiSSDM 30.00% 20.00% 10.00% 0.00% Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Month
Past 3 month condom use (Knoxville) 4.00 3.80 Lexington Campaign Knoxville 3.60 3.40 3.20 3.00 2.80 Condom use 2.60 2.40 2.20 2.00 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Months No statistically significant changes (p>.05)
Past 3 month condom use (Lexington) 4.00 3.80 Lexington Campaign Lexington 3.60 3.40 3.20 3.00 2.80 Condom use 2.60 2.40 2.20 2.00 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Months Linear trends are statistically significant (p<.05)
Results: Past 3 month Condom Use 4.00 3.80 Lexington Campaign Lexington Lexington trend Knoxville Knoxville trend 3.60 3.40 3.20 3.00 2.80 2.60 2.40 2.20 2.00 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Months HSSIDM Linear Trends in both cities (Lexington, p<.05) Condom Use
Past 3 month Condom Use w/ Trend 4.00 3.80 3.60 Lexington Campaign Lexington Lexington trend Knoxville Knoxville trend Lexington trend w /o campaign 3.40 3.20 3.00 2.80 Condom Use 2.60 2.40 2.20 2.00 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep -02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep -03 Oct-03 Nov-03 Dec-03 Jan-04 Months HSSIDM Linear Trends in both cities (Lexington, p<.05)
Results: Condom Self-Efficacy 4.20 4.00 Lexington Campaign Lexington Lexington trend Knoxville Knoxville trend 3.80 3.60 3.40 3.20 3.00 2.80 2.60 2.40 2.20 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Months HSSIDM Linear Trends in both cities (Lexington, p<.05) Condom Self-Efficacy
Results: Condom Use Intentions 4.20 4.00 Lexington Campaign Lexington Lexington trend Knoxville Knoxville trend 3.80 3.60 3.40 3.20 3.00 2.80 2.60 2.40 2.20 Apr-02 May-02 Jun-02 Jul-02 Aug-02 Sep-02 Oct-02 Nov-02 Dec-02 Jan-03 Feb-03 Mar-03 Apr-03 May-03 Jun-03 Jul-03 Aug-03 Sep-03 Oct-03 Nov-03 Dec-03 Jan-04 Months HSSIDM Linear Trends in both cities (Lexington, p<.05) Condom Use Intentions
Magnitude of Effects What do our results show? Past studies r =.04. Current study r =.13. An HSS/IDM young adult engaged in a total of 10.49 fewer acts of unprotected intercourse as a result of the campaign The campaign reached an estimated 20,469 HSS/IDM young adults Thus, we estimate that 181,224 acts of unprotected intercourse were averted due to the campaign
Discussion Overall, results provide support for the efficacy of a televised media-only campaign in changing the sexual risk behavior of HSS/IDM young adults High exposure was achieved among the target audience Condom use and related variables changed in the predicted direction in response to the campaign However, there was a wear-off effect, suggesting the need for a continued presence of campaigns (or short, well-timed campaigns)
Discussion / Implications Overall, this study suggests that health mass media campaigns can be effective in changing behavior if they attend to important campaign design principles Rigorous evaluation allows for greater precision in understanding campaign effects We are currently funded to use a similar design targeting young adolescents in a delay of sex campaign