5079.0: Wednesday, November 15, 2000 - 9:00 AM

Abstract #16826

Sex and the BRFSS, Examining refusals to sensitive questions

Karin Ann Mack, PhD1, Christopher Maylahn, MPH2, and Deborah Holtzman, PhD1. (1) NCCDPHP/DACH/BSB, Centers for Disease Control & Prevention, 4770 Buford Hwy NE K47, Atlanta, GA 30341, 770-488-2500, kim9@cdc.gov, (2) Bureau of Health Risk Reduction, Room 710 - Empire State Plaza, Tower Bldg, Albany, NY 12237

Item-nonresponse may be critical when investigating risk for HIV/AIDS because questions need to ask about private issues such as sexual behaviors. This project examines nonresponse to questions related to HIV and sexual behavior using the Behavioral Risk Factor Surveillance System. The BRFSS is a state-specific, population-based, random telephone survey active in all 50 states. In 1997, 24 states also included a set of sexual behavior questions.

The percentage of refusals ranged from .1 to 3% for HIV questions. The question with the highest refusals asked whether the respondent had donated blood. Refusals ranged from .1 to 6.5% for the sex behavior questions. The questions receiving the highest percentage of refusals asked whether a condom was used at last intercourse and the number of new sex partners in the past year. In comparison, the percentage of refusals to the income question was 6.4%. Those who refused to answer at least one question HIV or sex behavior question were slightly more likely to be male (54%). Those who refused at least one question in both sections were slightly more likely to be female (54%).

These analyses reveal that questions of a sensitive nature can be successfully included in large-scale population-based telephone surveys. While the questions reveal more nonresponse than other less-threatening questions (e.g., education), the error does not seem to be appreciable. Finally, a demographic profile of the refusers may assist in developing interview protocols to help obtain valid responses to sensitive questions from those likely to refuse.

Learning Objectives: 1. Identify issues related to item nonresponse 2. Assess impact of item nonresponse on sensitive questions 3. Develop ideas for interviewer protocols to assist in reducing item nonresponse

Keywords: Methodology, Surveillance

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA