Conclusion. The significance of risk perception in predicting behavioral intention, hence actual future behaviors, is therefore evident. Conditional measures need to be used. HIV prevention can employ conditional risk perception approaches. Health behavioral theories can be strengthened by using such conditional measures on risk perceptions. The results add to this new and growing area of risk behavior research. / Introduction. Risk perception, a core element of key health behavioral theories and health interventions, is assumed to motivate people to avoid risk behaviors. Mixed findings however prevail in the literature due to methodological issues. Many of such studies are cross-sectional, using global risk perception measures that do not condition on type of risk behavior or partnership which may affect the level of risk. / Male injecting drug users (IDU) are driving the HIV epidemic in China and bridge HIV transmission to non-IDU female populations; they may be at risk of both unprotected sex and syringe sharing. HIV prevention targeting male IDU is greatly warranted and would benefit from understanding of the relationships between risk perceptions and behaviors, in the context of health behavioral theories such as the Health Action Process Approach (HAPA) model, which had not been applied to studies targeting IDU. / Objectives. This study refined the concepts and measures of HIV-related risk perception, conditioning on different types of behaviors and partners, and extended it to include others-directed risk perceptions. The relationship between such conditional risk perception measures and both prior risk behaviors and behavioral intention to avoid sex-related and drug-related risk behaviors in the future were investigated and were compared to those involving global unconditional risk perception measures. The nature of the aforementioned relationships, being motivational or reflective was investigated. A longitudinal component validated the predictive power of behavioral intention over actual future behaviors. / Results. Almost 90% of the respondents had had unprotected sex though <20% shared syringes with others in the last 6 months. Prior syringe sharing but not unprotected sex in the last 6 months was significantly associated with global unconditional risk perception measures. The picture is totally different when risk perception measures conditioned on type of sex partner and unprotected sex or syringe sharing were used instead of the global measures---higher levels of the conditional risk perception measures were significantly associated with higher levels of behavioral intention for consistent condom use and avoidance of syringe sharing, thus supporting the motivational hypothesis. Conditional others-directed risk perceptions (perceived risk of transmitting HIV to others via unprotected sex and syringe sharing) were also associated with the aforementioned behavioral intentions to avoid risk behaviors. A pilot longitudinal study showed that behavioral intentions strongly predict actual future behaviors. Other HAPA-based variables such as self-efficacy and outcome expectancies had predictive effects on behavioral intentions, independent from those of risk perceptions. / Subjects and methods. A total of 456 sexually active male IDU were recruited from Dazhou, Sichuan and Hengyang, Hunan, via snowball sampling. With informed consent, anonymous face-to-face interviews were conducted by trained and experienced staff of the local CDC in privacy settings. / Tsui, Hi Yi. / Adviser: Joseph Lau. / Source: Dissertation Abstracts International, Volume: 72-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 118-130). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese; appendix in Chinese.