A conjoint analysis study on self-sampling for human papillomavirus (HPV) testing characteristics among black women in Indiana.

ABSTRACT

BACKGROUND: Self-sampling for HPV testing may be a method to increase overall cervical cancer screening rates among Black women, who are underscreened for cervical cancer in parts of the US. The purpose of this study was to assess preferred characteristics for delivery of HPV self-sampling kits, return of HPV self-sampling kits, and communication of HPV test results and explore sociodemographic factors (income, education, and marital status) associated with acceptability of self-sampling for HPV testing. METHODS: Survey data were gathered at an Indiana minority health fair. Participants evaluated 9 scenarios that varied along 3 dimensions HPV self-sampling kit delivery (mail, pharmacy pick-up, or clinic pick-up), HPV self-sampling kit return (mail, pharmacy drop-off, or clinic drop-off), and HPV test results (mail, phone call, or text message). The 9 scenarios were produced from a fractional factorial design and rated on a 0 to 100 scale. Ratings-based conjoint analysis (RBCA) determined how each dimension influenced ratings. A measure for acceptability of self-sampling was obtained from the ratings of all 9 scenarios. The acceptability measure was regressed on sociodemographics. RESULTS: The 98 participants ranged in age from 21 to 65 (M = 45). Across the 9 scenarios, overall acceptability to self-sample had a mean of 60.9 (SD = 31.3). RBCA indicated that HPV self-sampling kit return had the most influence on ratings, followed by HPV self-sampling kit delivery, and finally, HPV test result communication. Thirty-six percent of participants rated all self-sampling scenarios the same. Sociodemographic characteristics were not associated with acceptability of self-sampling. CONCLUSIONS: Self-sampling for HPV testing was found to be generally acceptable to Black women in this pilot survey study. This information could be used by researchers developing self-sampling interventions and the implementation of self-sampling among providers.