In addition to asking people about their condom use, interviewers also asked respondents if they were carrying a condom. Those who said "yes" were asked to show the condom to the interviewer. These data are important because they provide an objective, verifiable measure of behavior change.

Percentage of respondents observed
to be carrying a condom over time: CDC AIDS
Community Demonstration Projects, 1991-1994

Intervention Data points	Intervention Regression
Comparison Data points	Comparison Regression

Community-level data were obtained for condom carrying in all 10 community pairs. These data showed that there was a significant increase in condom carrying in the intervention communities but not in the comparison communities. Before the intervention, 17% of respondents in the intervention communities presented a condom compared with 30% by the end of the intervention. In the comparison communities, condom carrying remained unchanged at 19%.

At the individual level, recent exposure to the intervention was significantly associated with observed condom carrying. Across the intervention period, 31% of exposed respondents were carrying a condom, compared with 20% of non-exposed respondents.