full-text ►
HIV Risk, Prevention, and Testing Behaviors Among Heterosexuals at Increased Risk for HIV Infection — National HIV Behavioral Surveillance System, 21 U.S. Cities, 2010
MMWR Surveillance Summaries
Vol. 63, No. SS-14
December 19, 2014
PDF
HIV Risk, Prevention, and Testing Behaviors Among Heterosexuals at Increased Risk for HIV Infection — National HIV Behavioral Surveillance System, 21 U.S. Cities, 2010
Surveillance Summaries
December 19, 2014 / 63(SS14);1-39
Catlainn Sionean, PhD1
Binh C. Le, MD1
Kathy Hageman, PhD1
Alexandra M. Oster, MD1
Cyprian Wejnert, PhD1
Kristen L. Hess, PhD1,2
Gabriela Paz-Bailey, MD, PhD1for the NHBS study group
1Division of HIV/AIDS Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention
2ORISE Research Participation Program
Corresponding author: Catlainn Sionean, PhD, Division of HIV/AIDS Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, CDC. Telephone: 404-639-8702; E-mail: csionean@cdc.gov.
Abstract
Problem/Condition: At the end of 2010, an estimated 872,990 persons in the United States were living with a diagnosis of human immunodeficiency virus (HIV) infection. Approximately one in four of the estimated HIV infections diagnosed in 2011 were attributed to heterosexual contact. Heterosexuals with a low socioeconomic status (SES) are disproportionately likely to be infected with HIV.
Reporting Period: June–December 2010.
Description of System: The National HIV Behavioral Surveillance System (NHBS) collects HIV prevalence and risk behavior data in selected metropolitan statistical areas (MSAs) from three populations at high risk for HIV infection: men who have sex with men, injecting drug users, and heterosexuals at increased risk for HIV infection. Data for NHBS are collected in rotating cycles in these three different populations. For the 2010 NHBS cycle among heterosexuals, men and women were eligible to participate if they were aged 18–60 years, lived in a participating MSA, were able to complete a behavioral survey in English or Spanish, and reported engaging in vaginal or anal sex with one or more opposite-sex partners in the 12 months before the interview. Persons who consented to participate completed an interviewer-administered, standardized questionnaire about HIV-associated behaviors and were offered anonymous HIV testing. Participants were sampled using respondent-driven sampling, a type of chain-referral sampling. Sampling focused on persons of low SES (i.e., income at the poverty level or no more than a high school education) because results of a pilot study indicated that heterosexual adults of low SES were more likely than those of high SES to be infected with HIV. To assess risk and testing experiences among persons at risk for acquiring HIV infection through heterosexual sex, analyses excluded participants who were not low SES, those who reported ever having tested positive for HIV, and those who reported recent (i.e., in the 12 months before the interview) male-male sex or injection drug use. This report summarizes unweighted data regarding HIV-associated risk, prevention, and testing behaviors from 9,278 heterosexual men and women interviewed in 2010 (the second cycle of NHBS data collection among heterosexuals).
Results: The median age of participants was 35 years; 47% were men. The majority of participants were black or African American (hereafter referred to as black) (72%) or Hispanic/Latino (21%). Most participants (men: 88%; women: 90%) reported having vaginal sex without a condom with one or more opposite-sex partners in the past 12 months; approximately one third (men: 30%; women: 29%) reported anal sex without a condom with one or more opposite-sex partners. The majority of participants (59%) reported using noninjection drugs in the 12 months before the interview; nearly one in seven (15%) had used crack cocaine. Although most participants (men: 71%; women: 77%) had ever been tested for HIV, this percentage was lower among Hispanic/Latino participants (men: 52%; women: 62%). Approximately one third (34%) of participants reported receiving free condoms in the 12 months before the interview; 11% reported participating in a behavioral HIV prevention program.
Interpretation: A substantial proportion of heterosexuals interviewed for the 2010 NHBS heterosexual cycle reported engaging in behaviors that increase the risk for HIV infection. However, HIV testing was suboptimal among the overall sample, including among groups disproportionately affected by HIV infection (i.e., blacks and Hispanics/Latinos).
Public Health Action: Increasing coverage of HIV testing and other HIV prevention services among heterosexuals at increased risk is important, especially among groups disproportionately affected by HIV infection, such as blacks and Hispanics/Latinos. The National HIV/AIDS Strategy for the United States delineates a coordinated national response to reduce infections and HIV-related health disparities among disproportionately affected groups. NHBS data can guide national and local planning efforts to maximize the impact of HIV prevention programs.
Introduction
In the United States, an estimated 872,990 persons were living with a diagnosis of human immunodeficiency virus (HIV) infection at the end of 2010 (1). Heterosexual sex is the second most common route of transmission of HIV in the United States and is estimated to account for approximately one out of every four (27%) infections diagnosed in adolescents and adults during 2011 (1). Heterosexual sex is the primary route of transmission of HIV for women; approximately four out of every five new HIV infections among women diagnosed in 2010 were attributed to heterosexual contact (1). In addition, blacks or African Americans (hereafter referred to as blacks) and Hispanics/Latinos are disproportionately affected by HIV. Estimated rates of diagnosed HIV infection in 2011 were 9 times as high for blacks and 3 times as high for Hispanics or Latinos compared with the rate for whites. Racial and ethnic differences among women were greater, with an estimated rate of diagnosed HIV infection among black women that was 20 times as high and among Hispanic/Latino women that was 4 times as high as the rate among white women (1). The National HIV/AIDS Strategy for the United States addresses these issues by calling for a coordinated national response for reducing HIV incidence and HIV-related health disparities (2).
One objective of the National HIV/AIDS Strategy, released in 2010 (2), is to decrease the annual number of new infections by 25% by 2016. As outlined in the National HIV/AIDS Strategy (2), this objective can be achieved by implementing three key steps to reduce HIV infections: intensifying HIV prevention efforts in communities where HIV is most heavily concentrated (such as urban areas, which have the highest prevalence) (3); expanding efforts to prevent HIV infection by using a combination of effective, evidence-based approaches; and educating the general public about the threat of HIV and how to prevent infection. State and local health departments as well as federal agencies are expected to monitor progress toward the strategy's goals.
The National HIV Behavioral Surveillance System (NHBS) was designed to help state and local health departments in areas with a high prevalence of AIDS monitor selected risk behaviors, HIV testing experiences, use of prevention programs, and HIV infection in three populations at high risk for HIV infection: gay, bisexual, and other men who have sex with men (collectively referred to as MSM); injecting drug users (IDUs); and heterosexuals at increased risk for HIV infection (4). NHBS is an important component of CDC's comprehensive approach to reducing the spread of HIV in the United States and is the primary source of data for monitoring the behaviors of populations at risk for HIV infection. Findings from NHBS enhance the understanding of HIV risk and testing behaviors and identify gaps in prevention efforts. NHBS data are used at the state and local levels to renew and maintain efforts to prevent HIV infection as well as other bloodborne and sexually transmitted diseases (STDs). The data from this system could be used locally and nationally to monitor efforts toward reducing HIV infections and HIV-related health disparities in these populations.
The target population for the NHBS heterosexual cycle is adults in areas of high AIDS prevalence who are at risk for acquiring HIV infection through heterosexual sex (5). In the MSM and IDU data collection cycles, all persons who report engaging in the risk behavior (male-male sex and injection drug use, respectively) in the 12 months before the interview are considered at risk for HIV infection because of the higher prevalence of HIV infection among persons who engage in these behaviors (6). However, not all persons with opposite-sex partners are at equal risk for HIV. An evaluation of the first (pilot) NHBS data collection cycle among heterosexuals (2006–2007) indicated that low socioeconomic status (SES) (household income at or below the federal poverty guidelines or no more than a high school education) was an accurate marker of HIV acquisition risk among heterosexual adults in 24 U.S. cities with high AIDS prevalence (5). Therefore, the 2010 NHBS heterosexual cycle focused on heterosexuals of low SES who lived in urban communities with high AIDS prevalence. Use of low SES to identify heterosexuals at increased risk for HIV is consistent with other published studies of HIV infection among heterosexuals documenting the association between low SES and HIV infection (7–9).
This report summarizes unweighted data from the second NHBS data collection cycle among heterosexuals at increased risk for HIV, which was conducted during June–December 2010. In the 2010 NHBS heterosexual cycle, 2.3% of participants were HIV-positive (10). This report expands on the previous report (10) of HIV infection status among participants in the 2010 heterosexual NHBS cycle by 1) reporting the distribution of HIV risk, prevention, and testing behaviors by sociodemographic characteristics and metropolitan statistical area (MSA) and 2) including data from participants who reported a lifetime history but not recent history of male-male sex or injection drug use (i.e., the behavior had ever occurred but the most recent event was >12 months before the interview). Detailed data regarding HIV-related risk behaviors from the NHBS heterosexual cycle have not been reported previously. Monitoring these data is useful for 1) assessing the extent to which heterosexuals at increased risk for HIV (i.e., those of low SES living in communities with high AIDS prevalence) engage in HIV risk and preventive behaviors and 2) identifying opportunities for HIV prevention in this population.
Methods
NHBS collects HIV prevalence and risk behavior data in selected MSAs from three populations at high risk for HIV infection: MSM, IDUs, and heterosexuals at increased risk for HIV infection. Data for NHBS are collected in rotating cycles. Each of the three NHBS populations (MSM, IDUs, and heterosexuals at increased risk for HIV infection) is surveyed once every 3 years. A period of data collection with a specific population is referred to as a cycle. The survey for each cycle is anonymous. For each survey cycle, an anonymous standardized questionnaire is used to collect information about HIV-associated behaviors, specifically sexual behaviors, substance use, HIV testing, and use of HIV prevention services. The face-to-face survey is administered by a trained interviewer using a portable computer. All participants who consent to the survey are offered an anonymous HIV test, the results of which are linked to the survey data through a unique anonymous survey identifier and provided to participants in accordance with local policy. Participants may complete the survey even if they decline the HIV test. All participating state and local jurisdictions obtained human subject protections approval before initiating data collection for the 2010 NHBS heterosexual cycle. As a component of HIV/AIDS surveillance, NHBS data are protected by an Assurance of Confidentiality under Section 308(d) of the Public Health Service Act (42 U.S.C. 242 m(d)). As a part of this assurance, funded health departments applied and trained NHBS staff members to follow data security standards consistent with those for CDC guidelines for HIV/AIDS surveillance data (11).
Participating Areas
State and local health departments that were eligible to participate in NHBS were those whose jurisdictions included an MSA or a specified MSA division with the highest AIDS prevalence in 2006 (CDC, unpublished data, 2006). The 2010 heterosexual cycle of NHBS was conducted in the following MSAs (or if a metropolitan division is indicated, the survey was conducted within that specific division of the MSA): 1) Atlanta-Sandy Springs-Marietta, Georgia; 2) Baltimore-Towson, Maryland; 3) Boston-Cambridge-Quincy, Massachusetts-New Hampshire: Boston-Quincy Division; 4) Chicago-Joliet-Naperville, Illinois-Indiana-Wisconsin: Chicago-Joliet-Naperville Division; 5) Dallas-Fort Worth-Arlington, Texas: Dallas-Plano-Irving Division; 6) Denver-Aurora-Broomfield, Colorado; 7) Detroit-Warren-Livonia, Michigan: Detroit-Livonia-Dearborn Division; 8) Houston-Sugar Land-Baytown, Texas; 9) Los Angeles-Long Beach-Santa Ana, California: Los Angeles-Long Beach-Glendale Division; 10) Miami-Ft. Lauderdale-Pompano Beach, Florida: Miami Division; 11) New Orleans-Metairie-Kenner, Louisiana; 12) New York-Northern New Jersey-Long Island, New York-New Jersey-Pennsylvania: New York-White Plains-Wayne Division; 13) New York-Northern New Jersey-Long Island, New York-New Jersey-Pennsylvania: Nassau-Suffolk Division; 14) New York-Northern New Jersey-Long Island, New York-New Jersey-Pennsylvania: Newark-Union Division; 15) Philadelphia-Camden-Wilmington, Pennsylvania, New Jersey, Delaware, Maryland: Philadelphia Division; 16) San Diego-Carlsbad-San Marcos, California; 17) San Francisco-Oakland-Fremont, California: San Francisco-San Mateo-Redwood City Division; 18) San Juan-Caguas-Guaynabo, Puerto Rico; 19) Seattle-Tacoma-Bellevue, Washington: Seattle-Bellevue-Everett Division; 20) St. Louis, Missouri-Illinois; and 21) Washington-Arlington-Alexandria, District of Columbia (DC)-Virginia-Maryland-West Virginia: Washington-Arlington-Alexandria Division. Nearly one half of all persons living with HIV in the United States and Puerto Rico as of the end of 2009 lived in one of these MSAs (12). Throughout this report, MSAs are referred to by the name of the primary principal city (Figure 1).
Eligibility
The same basic eligibility criteria are used in each NHBS cycle: age ≥18 years, a current resident of a participating MSA or specified MSA division, not a previous participant in NHBS during the current survey cycle, ability to complete the survey in either English or Spanish, and ability to provide informed consent. In addition to these basic eligibility criteria, participation in the NHBS heterosexual cycle is limited to men and women aged ≤60 years who had vaginal or anal sex with an opposite-sex partner in the 12 months before the interview. The maximum eligible age was limited to 60 years because the estimated rates of HIV diagnoses are relatively low among persons aged >60 years (1). Low SES was not an eligibility criterion but was used in the sampling strategy as described in the following section.
Sampling Method
Participants for the 2010 heterosexual cycle of NHBS were recruited using respondent-driven sampling (RDS), a type of chain-referral sampling. The RDS sampling strategy used in NHBS heterosexual cycles is similar to the strategy used in NHBS IDU cycles, the details of which have been described elsewhere (13). RDS starts with a limited number of persons (seeds) chosen by referrals from those who know the local target population well or through outreach to areas where the target population can be found. Eligible seeds who participate are asked to recruit other members of the target population. These persons, in turn, complete the survey and are asked to recruit others. This recruitment process continues until the sample size has been reached or the sampling period ends.
To reach the target population of heterosexuals at increased risk for HIV, in the 2010 NHBS heterosexual cycle, seeds were recruited from poverty areas within each MSA based on estimated 2009 poverty rates for census tracts within the participating MSAs. Poverty areas are defined by the U.S. Census Bureau as census tracts in which at least 20% of residents live below the poverty threshold (14). Seeds completed the eligibility screener; those who were eligible and consented to participate were administered the survey. Seeds who completed the survey were asked to recruit three to five persons they knew personally to participate and were given coupons with information about the project, such as phone number, hours, and locations, to provide to their recruits.
Data Collection
All persons who brought a valid coupon to a field site for the 2010 NHBS heterosexual cycle were escorted to a private area for eligibility screening. Trained interviewers obtained informed consent for the survey from eligible participants and offered an anonymous HIV test to participants who consented to the survey. Interviewers administered the survey in person using a portable computer. The survey instrument included questions regarding participants' demographic characteristics, sexual and drug-use behaviors, HIV testing history, hepatitis vaccination, STD diagnoses, and use of HIV behavioral interventions. Interviews were an average of 32 minutes. In exchange for their time and effort in taking part in the survey, participants received approximately $25 in cash or a gift certificate; participants who also agreed to the HIV test received additional compensation, typically $25. The specific amount and form of incentives for the interview and HIV test were determined locally. Participants who agreed to recruit other participants were given one to five uniquely numbered recruitment coupons. Participants received approximately $10 in cash or a gift certificate for each new participant recruited who completed the interview. The specific amount of the incentive was determined locally.
To focus recruitment on the target population, opportunity to recruit other participants was limited to those who met the NHBS definition of a heterosexual at increased risk for HIV (low SES) and who reported no recent (i.e., in the past 12 months) injection drug use. Low SES was defined as having completed no more than a high school education or having a household income that was at or below the U.S. Department of Health and Human Services poverty guidelines (15). Each participating MSA attempted to interview 450 heterosexuals at increased risk for HIV infection (i.e., heterosexual sexually active, low SES men and women aged 18–60 years who had not injected drugs in the 12 months before the interview).
Data Analysis
This report summarizes unweighted data from 9,278 heterosexual men and women interviewed in 21 MSAs during the second cycle of NHBS data collection among heterosexuals, which was conducted in 2010. Inclusion in this report is limited to participants who 1) were eligible for and consented to the interview and 2) reported low SES (i.e., either no more than high school education or an income at or below the poverty level). In addition, to assess risk and testing experiences among persons at risk for acquiring HIV infection through heterosexual sex, all analyses excluded data from participants who reported a previous positive HIV test and those who reported recent (i.e., in the 12 months before the interview) male-male sex or injection drug use or did not provide this information (Figure 2). Data from participants who did not report a previous positive HIV test during the interview but who subsequently had a positive NHBS HIV test result are included in this report because persons who do not know their HIV status are an important focus of HIV prevention efforts. Data from participants with a previous but not recent (i.e., >12 months before the interview) history of injection drug use or male-male sex were included because the reported behaviors indicated exclusively heterosexual risk in the 12 months before the interview.
Four time frames for risk and testing experiences are included in this report: 1) ever (i.e., at any point in the participant's lifetime), 2) in the 12 months before the date of the interview, 3) in the 30 days before the date of the interview, and 4) the most recent time the participant engaged in the behavior. All numbers and percentages included in this report are unweighted. Weighting methods for RDS data are still under development and are limited to estimation of outcomes with sufficient local sample sizes. In addition, unweighted analysis allows reporting of outcomes by subgroups for which the number of participants is not sufficient for weighted analysis but aggregation into a single "other" category would obscure important information (e.g., American Indians/Alaska Natives). Between-group differences of >5 percentage points are noted in the text for key outcomes deemed meaningful in the public health context, given the NHBS sample size. However, the data are descriptive; no statistical tests were conducted, and group differences should be interpreted with caution. To ensure that anonymous data reporting standards of CDC and all state and local health departments were met, numbers and percentages for numbers or numerators of fewer than six participants were suppressed.
Measures
Participant Characteristics
Data were analyzed according to the following characteristics of participants: sex, age group, race/ethnicity, marital status, education level, annual household income, health insurance status, census region, and MSA. Reponses for race/ethnicity were categorized into six mutually exclusive categories: American Indian/Alaska Native, black, Hispanic/Latino, Asian/Native Hawaiian/Other Pacific Islander, white, and multiple races. Persons of Hispanic/Latino ethnicity might be of any race. Asian was combined with Native Hawaiian/Other Pacific Islander to prevent suppression of data for each of these groups in several tables. Marital status was categorized as married or cohabiting, formerly married (separated, divorced, or widowed) and not cohabiting, and never married and not cohabiting. Education level was categorized as less than high school graduate, high school diploma or equivalent (e.g., general educational development [GED] certificate), and some college or higher level education. Participants were asked to identify which range in a list of ranges included their annual household income; in this report, income ranges were combined into four categories: $0–$4,999; $5,000–$9,999; $10,000–$19,999; and ≥$20,000. Participants were classified as living in poverty if the household income range reported during the interview was at or below the federal poverty guidelines for the total number of dependents, including the participant, for the reported income (15). Nearly all participants in the analysis sample were classified as living in poverty (men: 85%; women: 90%); approximately two thirds (67%) had at least one dependent in addition to the participant on their reported income, and approximately two thirds (63%) reported an income of <$10,000, which is below the federal poverty guidelines for persons with no additional dependents. Among participants with more than one dependent, the median number of dependents was three (interquartile range [IQR]: two to four). Therefore, income is reported rather than poverty status to demonstrate the extreme poverty among the 2010 NHBS heterosexual cycle participants. Consistent with the Stewart B. McKinney Homeless Act of 1987 (42 U.S.C. §11331 et seq.), homelessness was defined as living on the street, living in a shelter, or staying with friends or relatives. Health insurance was categorized as none; private only (i.e., health insurance obtained through a private insurance policy or employer, TRICARE, CHAMPUS, or membership in a health maintenance organization); public only (i.e., Medicare, Medicaid, or Veterans Administration coverage); or other coverage, including both public and private. Male-male sex (male participants only) was defined as oral or anal sex with another man. Injection drug use was defined as injection of drugs not prescribed for the participant. As described previously, all participants who reported male-male sex or injection drug use in the 12 months before the interview or did not provide this information were excluded from all analyses. Participants' region of residence was classified for each MSA according to the U.S. Census Bureau for each of the NHBS MSAs in the continental United States (16); an additional category (territories) was used for the San Juan, Puerto Rico, MSA.
Sexual Behaviors
Details about the number of opposite-sex partners (overall and by type) and about vaginal and anal sex with opposite-sex partners in the 12 months before the interview are presented for male and female participants. Participants were asked the number of opposite-sex partners overall and by type. Sex was defined as oral, vaginal, or anal sex. Sex partners were categorized as main or casual partners. A main partner was someone to whom the participant felt most committed (e.g., girlfriend or boyfriend, wife or husband, significant other, or life partner). A casual partner was someone to whom the participant did not feel committed, whom the participant did not know very well, or with whom the participant had sex in exchange for something such as money or drugs. Participants could report having more than one main or casual partner in the past 12 months. Within each partner type, participants were asked the number of partners with whom they had engaged in each of the following behaviors: vaginal sex, vaginal sex without a condom, anal sex, and anal sex without a condom. In this report and the accompanying tables, unless otherwise specified, sex partner refers to opposite-sex partners.
Alcohol and Drug Use
Participants were asked about their use of alcohol in the 30 days before and the 12 months before the interview and of their use, in the 12 months before the interview, of specific noninjection drugs that had not been prescribed for them. Current alcohol use was defined as drinking any alcohol, such as beer, wine, malt liquor, or hard liquor, in the 30 days before the interview. Participants reporting current alcohol use were asked on how many days they had consumed alcohol in the past 30 days and on average how many alcoholic beverages they consumed on the days they drank. Men were asked the number of times in the past 30 days they had consumed five or more alcoholic beverages at one sitting; women were asked the number of times they had consumed four or more alcoholic beverages at one sitting. Heavy drinking was defined as drinking, on average, more than two alcoholic beverages per day for men and more than one alcoholic beverage per day for women. Binge drinking was defined as having at least once, in the past 30 days, at least five alcoholic beverages at one sitting for men and at least four alcoholic beverages at one sitting for women.
Participants were asked about their noninjection drug use in the past 12 months of multiple types of drugs that had not been prescribed for them, including marijuana, crack, cocaine, heroin, methamphetamine, downers (e.g., Valium, Ativan, or Xanax), painkillers (e.g., Oxycontin, Vicodin, or Percocet), or any other noninjection drugs. Use of any noninjection drug was defined as use of one or more of the drugs listed above.
Sexually Transmitted Disease Diagnoses
Participants were asked whether they had been diagnosed by a health-care provider with any of the following in the 12 months before the interview: chlamydia, gonorrhea, syphilis, genital herpes, genital warts, or another STD. The percentages of participants reporting any STD, chlamydia, gonorrhea, or another STD are reported by sex and participant characteristics.
Use of Prevention Services and Programs
HIV Testing
History of HIV testing (lifetime or in the 12 months before the interview) is presented for all participants. The type of facility that administered the most recent HIV test and receipt of test result by facility type are reported for all participants tested in the 12 months before the interview. Reasons for not having received an HIV test in the 12 months before the interview are presented for all participants who reported their most recent test was >12 months before the interview or reported they had never been tested for HIV. Participants were asked to select from a list of reasons for not having been tested in the past 12 months (e.g., thought they were at low risk for HIV infection, fear of finding out they had HIV infection, and lack of time, money, or transportation). Participants could select more than one reason, and those who did were asked to indicate the most important reason.
Hepatitis B Vaccination
All participants were asked whether they had ever received a diagnosis of hepatitis; those reporting a previous hepatitis diagnosis were asked to indicate the type (A, B, C, or other). All participants were asked whether they had ever been vaccinated against hepatitis; those reporting hepatitis vaccination were asked the type. Hepatitis B vaccination was defined as having ever received a hepatitis B vaccine, regardless of whether it was a single or combination vaccine (i.e., against hepatitis B virus only or against both hepatitis A virus and B virus ).
Prevention Materials and Behavioral Interventions
Participants were asked whether they had received free condoms (excluding those given by a friend, relative, or sex partner) and whether they had participated in individual- or group-level HIV-related behavioral interventions in the 12 months before the interview. The definitions for both intervention types were based on those in CDC's evaluation system (17). Conversations that took place solely as a part of HIV testing (e.g., pretest or posttest counseling) were not considered HIV behavioral interventions. Participants who reported receiving free condoms or behavioral interventions were asked to report the type of organization that provided the condoms or intervention.
Results
A total of 12,544 persons were recruited to participate in the 2010 NHBS heterosexual cycle (Figure 2). Of the 12,478 participants screened for whom data were available, 11,066 (89%) were eligible for the interview. (Interview data for 66 participants were lost during electronic upload.) A total of 1,412 were not eligible for the interview: 49 lived outside the MSA; 53 did not identify as male or female; 193 were aged <18 years or >60 years; 230 had already participated; 830 did not report vaginal or anal sex with an opposite-sex partner in the 12 months before the interview; and 184 were not able to provide their consent to the survey (e.g., were too intoxicated or did not speak either English or Spanish well). Exclusion categories are not mutually exclusive. Of the 11,066 eligible persons, 11,061 agreed to participate and 10,933 completed the interview with valid data. A total of 1,655 completed interviews were excluded: 1) 730 participants did not meet the definition of a heterosexual at increased risk for HIV infection (i.e., reported income above poverty level and education greater than high school, 2) 816 participants reported engaging in male-male sex or injection drug use in the 12 months before the interview or did not provide this information, and 3) 197 participants reported a previous HIV-positive test result; exclusion categories were not mutually exclusive. A total of 9,278 participants met all inclusion criteria for analysis, including 126 participants (1.4%) who did not report a previous positive HIV test during the interview but who subsequently had a positive NHBS HIV test result.
Participant Characteristics
Approximately half (58%) of participants were aged 40–60 years. The majority of participants reported black race (72%) or Hispanic/Latino ethnicity (21%); the sample was evenly distributed by sex (Table 1). The majority of participants (64%) had never married and were not cohabiting. Most participants reported no more than a high school education (men: 86%; women: 83%) and very low incomes. For instance, most participants (87%) reported annual incomes of <$20,000, which is lower than the federal poverty guidelines for a family of four; approximately one in three (36%) reported annual incomes of <$5,000. Nearly three out of four (men: 72%; women: 73%) reported both income at or below poverty and no more than a high school education (data not shown). Approximately one third (31%) of participants were currently or had been homeless in the 12 months before the interview, and nearly half (46%) were uninsured.
With the exception of homelessness and health insurance, the demographic characteristics were similar for male and female participants. Approximately one third of male (35%) and female (27%) participants had been homeless at some time in the 12 months before the interview. Approximately one half (58%) of male and one third (36%) of female participants had no health insurance. For both male and female participants, the most frequently reported insurance was a publicly funded plan. Among those with publicly funded plans, the majority of both male (75%) and female (80%) participants reported coverage by Medicaid (data not shown).
Although the analysis sample excluded data from participants who reported they had recently (within 12 months of interview) injected drugs or had male-male sex, 14% of male participants reported that they had injected drugs or had sex with another man >12 months before the interview; 5% of female participants had injected drugs >12 months before the interview (data not shown). Each participating area contributed an average of 442 interviews (range: 131–565) to this analysis.
Sexual Behaviors
Number and Types of Partners
Among the 4,323 male participants, the median number of female sex partners in the 12 months before the interview was three (IQR: one to six); a total of 3,234 (75%) reported having more than one female sex partner in the 12 months before the interview (median: four; IQR: three to eight). A total of 3,266 (76%) reported a female main sex partner (median: one; IQR: one to two), and 3,112 (72%) reported having a female casual sex partner (median: three; IQR: two to seven). Overall, 2,055 (48%) reported having both main and casual female partners, and 801 (19%) reported having one or more female partners with whom they had exchanged money or drugs for sex in the 12 months before the interview (data not shown).
Among the 4,955 female participants, the median number of male sex partners in the 12 months before the interview was two (IQR: one to four); a total of 3,138 (63%) reported having more than one male sex partner in the 12 months before the interview (median: three; IQR: two to five). A total of 4,172 (84%) reported a male main sex partner (median: one; IQR: one to one), and 2,898 (58%) reported having a male casual sex partner (median: two; IQR: one to five). Overall, 2,115 (43%) reported having both main and casual male partners, and 910 (18%) reported having one or more male partners with whom they had exchanged money or drugs for sex in the 12 months before the interview (data not shown).
Types of Sexual Behaviors
Overall, 3,793 (88%) male participants reported having vaginal sex without a condom with a female partner in the 12 months before the interview, and 1,304 (30%) reported having anal sex without a condom with a female partner (Table 2). The percentages of men who had vaginal sex without a condom were similar among men in all categories of education and income. Anal sex without a condom was more common among male participants with lower levels of income.
A total of 4,467 (90%) female participants reported having vaginal sex without a condom with a male partner in the 12 months before the interview, and 1,420 (29%) reported having anal sex without a condom with a male partner (Table 3). The percentages of women who had vaginal sex without a condom were similar among women in all categories of education and income. Anal sex without a condom was more common among those with lower levels of income.
A total of 3,252 (75%) male participants reported having vaginal or anal sex with a female main partner, and 3,063 (71%) reported having vaginal or anal sex with a female casual partner (Table 4). Both vaginal or anal sex and vaginal or anal sex without a condom with main partners were more common among male participants who were married or cohabiting and those with higher incomes and less common among white men. Vaginal or anal sex and vaginal or anal sex without a condom with female casual partners were less common among male participants who were married or cohabiting and among those with higher incomes and were more common among white men.
A total of 4,154 (84%) of female participants reported having vaginal or anal sex with a male main partner, and 2,837 (57%) reported having vaginal or anal sex with a male casual partner (Table 5). Both vaginal or anal sex and vaginal or anal sex without a condom with a male main partner were more common among female participants who were married or cohabiting or had a higher income. Both vaginal or anal sex and vaginal or anal sex without a condom with male casual partners were less common among female participants who were married or cohabiting and those with higher incomes.
Alcohol Use
Approximately three fourths of male and two thirds of female participants had consumed an alcoholic beverage in the 30 days before the interview (i.e., were current drinkers) (Table 6). One fourth of male (25%) and female (25%) participants were classified as heavy drinkers. Heavy drinking increased with age and was most common among participants who had annual incomes of <$5,000 (men: 27%, women: 29%). Nearly one half of male participants (48%) and 40% of female participants reported binge drinking in the 30 days before the interview (Table 6). Among female participants, binge drinking was reported most frequently by those who were formerly married or had never been married and were not cohabiting. Among both male and female participants, binge drinking was least common among those aged 18–19 years (men: 36%, women: 28%) and most common among participants who participated in an alcohol or drug treatment program >12 months before the interview (men: 56%; women: 53%).
Noninjection Drug Use
Among the 9,278 participants, 5,440 (59%) had used noninjection drugs in the 12 months before the interview (Table 7). Most participants interviewed used marijuana (51%), followed by crack cocaine (15%), powdered cocaine (12%), painkillers (11%), and ecstasy (11%) (Table 7). For marijuana and powdered cocaine, a higher percentage of male participants reported use in the 12 months before the interview than did female participants. Compared with other racial/ethnic groups, lower percentages of Hispanics/Latinos reported using all drugs except powdered cocaine. Crack cocaine use was more commonly reported by participants who were aged ≥40 years (40–49 years: 28%; 50–60 years: 31%), were American Indian/Alaska Native (25%), were white (23%), or reported an annual income of <$5,000 (20%).
Sexually Transmitted Disease Diagnoses
Six percent of male and 14% of female participants reported receiving an STD diagnosis in the 12 months before the interview (Table 8). Among male participants, the percentages reporting an STD diagnosis were highest among those aged 20–29 years (8%–10%). Among female participants, percentages reporting an STD diagnosis decreased with increasing age and were highest among participants who were Asian/Native Hawaiian/Other Pacific Islander (16%), were black (16%), reported multiple racial identities (18%), reported public or other insurance (16%), or were sampled in one of the MSAs in the Midwest (22%). Chlamydia was the most commonly reported STD (men: 3%; women: 7%).
Use of Prevention Services and Programs
HIV Testing
Approximately three fourths of participants (men: 71%; women: 77%) had been tested for HIV infection during their lifetime, and approximately one third (men: 30%; women: 35%) had been tested in the 12 months before the interview (Table 9). Among both male and female participants, the percentages reporting that they had ever been tested for HIV were lowest among Asian/Native Hawaiian/Other Pacific Islander and Hispanic/Latino participants (men: 30% and 52%; women: 58% and 62%, respectively). Among male participants, the percentage ever tested was lowest among the two youngest age groups (18–19 years: 46%; 20–24 years: 63%); among female participants, the percentage ever tested was lowest among the youngest and oldest age groups (18–19 years: 61%; 50–60 years: 70%). Among participants who had visited a health-care provider in t