Use of conjoint analysis to

1

• Use of conjoint analysis to
• assess HIV vaccine acceptability
• in three populations
• An innovation in the assessment of consumer
  healthcare preferences (Project VIBE)

Sung-Jae Lee, Ph.D., Peter A. Newman, Ph.D.,
William E. Cunningham, MD, M.P.H., Danielle
Seiden, M.P.P., Naihua Duan, Ph.D.
2
Outline

• Background
• Assessing HIV Vaccine Acceptability
• Methodology
• Conjoint Analysis
• Results
• Three Samples
• Conclusions
• Application in HIV Vaccine Acceptability

3

• BACKGROUND

4

• What will happen when HIV vaccines become
• publicly available?

5

• If we build it, they will come!
• but will they???

6
Sub-optimal Uptake of Existing Vaccines

• Non-institutionalized U.S. adults 18 64 yrs
• 23 47 influenza vaccine
• 20 58 pneumococcal vaccine
• Lower coverage rates African Americans,
  Latinos, low SES

7

• HIV vaccine uptake is not guaranteed

8
Conjoint Analysis

• Well established research technique to
• predict consumer preferences when faced
• with a number of products varying across
• specific attributes (Green, 1999)
• Extensive use in psychology, marketing,
• and economics

9
Conjoint Analysis

• Growing application in measuring health
• care preferences (Phillips 2002, 2006),
• ranging from microbicide use (Holt, 2006)
• and anti-inflammatory drugs (Fraenkel,
• 2004) to hearing aids (Meister, 2002)
• and glaucoma treatment (Bhargava, 2006)

10
Objective

• Conjoint Analysis
• To test the feasibility of conjoint analysis
  application to measure HIV vaccine acceptability
  among three diverse communities in Los Angeles,
  California and Toronto, Canada.

11

• METHODS

12
Participants

• Multi-ethnic communities in Los Angeles,
  California
• At risk communities Individual interviews n143
• Thai Residents in Los Angeles, California
• Three focus groups n27
• Aboriginal Canadians, Toronto, Canada
• Two focus groups n13

13
Conjoint Analysis Methods

• Assigning levels of attributes
• Integrating input from focus group results, HIV
  vaccine experts, published and unpublished
  literature
• Hypothetical HIV vaccine being composed of a
  bundle of 7 dichotomous attributes

14
HIV Vaccine Attributes
15
Conjoint Analysis Methods

• Assigning conjoint scenarios
• (27 128) hypothetical scenarios
• Fractional factorial design
• 8 hypothetical HIV vaccines
• Conjoint analysis in different modalities

16
Rating Hypothetical HIV Vaccines
Rate eight hypothetical HIV vaccines from Highly
likely to Highly unlikely
17
Conjoint Scenario Cards
18
Conjoint Analysis Methods

• Estimate acceptability score for each vaccine by
  averaging individual vaccine ratings across
  participants
• One sample t-test to determine impact
  significance

19

• RESULTS

20
HIV Vaccine Acceptability
21
HIV Vaccine Acceptability
22
Impact of Vaccine Attributes on HIV Vaccine
Acceptability
23

• CONCLUSIONS

24
Conclusions

• Conjoint analysis provided insight into HIV
• vaccine acceptability among diverse potential
  consumers
• Vaccine acceptability varied widely across 8
  vaccines with different attributes
• Efficacy had the greatest impact on acceptability
• Attribute impacts varied across three samples

25
Conclusions

• We demonstrated successful applications
• of conjoint analysis
• Individual setting ( with a skilled interviewer)
• Group setting (with trained facilitators)

26
Marketing Economics
Biomedical Science
Behavioral Science
Conjoint Analysis
HIV Vaccine Acceptability
27
New Initiatives

• L.A. VOICES Post-trial HIV Vaccines
  Receptivity, Risk Disparities
  (R01-MH-069087-01A1)
• Sisters, Mothers, Daughters Aunties Protecting
  Black Women against HIV (Canadian Institutes of
  Health Research)
• HIV Vaccine Acceptability in Thailand (Social
  Sciences Humanities Research Council, Canada)
• HIV Vaccine Trial Participation Community
  Engagement (Ontario HIV Treatment Network)

28
References

• Duan N. Listening to consumers and HIV vaccine
  preparedness. Lancet. 2005 Apr365(9465)1119-21.
  See also The Lancet The trials of tenofovir
  trials. Lancet. 2005 April365(9465)1111.)
• Newman P.A., Duan, N., Lee, S-J., Rudy, E.T.,
  Seiden, D.S., Kakinami, L., Cunningham, W.E. HIV
  vaccine acceptability among communities at risk
  The impact of vaccine characteristics. Vaccine.
  2006 Mar 1524(12)2094-101.

29
Supported By

• UCLA AIDS Institute Seed Grant
• (CC99-LA-002)
• UCLA AIDS Institute and Palotta Teamworks AIDS
  Vaccine Rides
• Center for HIV Identification Prevention and
  Treatment Services (P30 MH 58107)
• Social Sciences and Humanities Research Council
  Institutional Grant (Canada)

30
Acknowledgements

• Thanks to Peter Anton, Lauren Arguelles, Phil
  Batterham, Ned Bayrd, Omar Banos, Ron Brooks,
  Coleen Cantwell, Suzi Cantwell, Mark Etzel,
  Kathie Ferbas, Neil Gajasan, Sonia Johnson, Ella
  Kelly, Faith Landsman, Kathy Mattes, Irma
  Ocegueda, Rowell Ramos, Fen Rhodes, Mary Jane
  Rotheram, Rassamee Sangthong, Dallas Swendeman,
  Julian Wang, Michael Woodford, Shin-Yi Wu, Paul
  Xue
• Special thanks to all participants, study sites
  key informants without whom this research would
  not have been possible.

31

• For more information, please contact

Sung-Jae Lee, Ph.D. Semel Institute Neuroscience
and Human Behavior, UCLA Center for Community
Health sjlee_at_mednet.ucla.edu
Peter Newman, Ph.D.University of Toronto
Faculty of Social Work 246 Bloor Street West
Toronto, Ontario M5S 1A1 CANADA p.newman_at_utoront
o.ca
Naihua Duan, Ph.D. Semel Institute Neuroscience
and Human Behavior, UCLA Health Service Research
Center naihua_at_mednet.ucla.edu