
INTRODUCTION
Black men “on the down-low” have gained considerable attention from both mainstream media and public health officials over the past few years as climbing rates of HIV infection (Generic Retrovir еreating HIV infection when used along with other medicines) among black women have come under increasing focus. Several nuances of the term “down-low” have been in use in the African-American community since the early 1990s. However, it was not until the release of statistics citing high HIV (Generic Zerit еreating HIV infection when used in combination with other medicines) incidence among black MSM and the fact that a significant proportion of black MSM identify as heterosexual that the term in the popular media became synonymous with heterosexually identified men who have sex with other men without the knowledge of their main female sex partners.

The results of this study suggest that while HIV prevention messages have been seen by APIs at risk of HIV (Generic Retrovir treating HIV infection when used along with other medicines), additional efforts may be needed specifically for APIs. Cultural norms prohibiting discussion of sexuality and discomfort regarding drug use and homosexuality may prevent standard HIV messages from reaching at-risk APIs. Further, foreign-born APIs may have language barriers and fear of deportation if they are infected or even at risk of HIV
Of the 187 individuals who responded to the HITS-API survey, 165 (88%) met the inclusion criteria. HITS-2000 recruited 314 participants, of whom 235 (75%) were included. Table 1 shows the demographic characteristics of both cohorts. HITS-API respondents were younger, had higher incomes and were less likely to be homeless or IDU. API respondents in HITS-2000 were comparable to those in HITS-API except that more HITS-API participants were of multiple race/ethnicity (data not shown). The most common API ancestries of participants in HITS-API were Filipinos (33%), followed by Chinese (19%), Vietnamese (17%) and Japanese (15%). Most (96%) HITS-API respondents had heard or seen HIV (Viramune 200mg treating HIV infection) prevention messages within a year of the survey; the most common venues were newspapers/magazines (72%), health centers (72%), television (68%), bus/bus stop (61%), billboards (58%) and radio (50%). Questions about prevention messages were not asked in the HITS-2000 survey.
The questionnaire for HITS-API was primarily that used nationally for HITS in 2000. Participants were recruited from three populations at highest risk for HIV (Zerit drug treating HIV infection when used in combination with other medicines) men who have sex with men (MSM) who have had sexual contact with other men in the past 12 months; higher-risk heterosexuals (HRH) who had two or more opposite sex partners in the past 12 months; and injection drug users (IDU) who had injected in the past 12 months. The eligibility requirements for both studies included being >18 years of age, a resident of Washington State for at least six months, English speaking and meeting the risk criteria (MSM, IDU and/or HRH, as above). HITS-API also required subjects to self-identify as APIs. Formative research was conducted to select venues for both projects and to identify other recruitment methods for HITS-API, including advertising, use of outreach workers and respondent-driven sampling (RDS). All participants provided informed consent prior to the interview. Trained staff conducted the interviews and respondents received monetary incentive for participation. Analyses were done using SAS v.8.2 software (Cary, NC, 1999-2001). Survey methods were reviewed and approved by CDC and Washington State Institutional Review Boards.

By December 2002, the CDC estimated 6,924 Asians and Pacific Islanders (APIs) had been diagnosed with AIDS in the United States, representing 0.8% of 886,575 total AIDS cases. APIs made up 0.4% of the estimated 312,133 HIV/AIDS cases reported in 30 U.S. areas with confidential name-based HIV reporting. APIs represent almost 4% of the total U.S. population. Although APIs represent a small proportion of all reported HIV/AIDS cases, APIs may be underreported due to several factors. APIs are often misclassified as other races in medical records and HIV surveillance. National HIV (Generic Retrovir treating HIV infection when used along with other medicines) reporting data for the most recent period available did not yet include areas with large numbers of APIs, including the entire West Coast and New York City. Also, an underlying social stigma against discussing sexuality and drug use among API cultures may prevent many from testing for HIV Finally, APIs are underrepresented among patients at facilities where HIV testing occurs, indicating that APIs may not be tested as often as other racial/ethnic groups.
Despite higher rates of morbidity and mortality from many cancers and nonmalignant chronic diseases, African Americans have been poorly represented in epidemiologic studies. One impediment to inclusion has been the tendency of studies to sample from populations where African Americans have limited representation. Moreover, African Americans have often been reluctant to participate because of mistrust resulting from the historical exploitation of blacks in medical research studies, the belief that medical research is not carried out ethically, and the view that chronic disease prevention is a low priority in the context of more immediate socioeconomic and psychosocial pressures.
A high level of participant enrollment was achieved in the CHCs. The calculation of a standard response rate, however, is not possible, due to the recruitment strategies employed by the interviewers. In addition to randomly approaching single individuals, some interviewers found that a successful means of recruitment was to speak to an entire CHC lobby at once, explaining the study and its eligibility requirements, then enrolling the resulting volunteers. Moreover, we found that there were various routes by which potential participants came to the interviewer already prescreened for interest and eligibility: 1) through word-of-mouth referrals from previous participants; 2) through referrals from the CHC receptionists or medical staff; and 3) from volunteers who sought out the interviewer in response to seeing posters, flyers, brochures or even articles and advertisements placed in local newspapers by some CHCs.