More than one-third of incident human immunodeficiency virus (HIV) cases in the United States are occurring among adolescents and adults aged <30 years,1 and other sexually transmitted diseases (STDs) are particularly common among young adults. For example, in 2008 the age-specific incidence rates of reported chlamydia, gonorrhea, and syphilis cases among 20- to 24-year olds were 5.2, 4.4, and 1.8 times higher, respectively, than the rates among all age groups combined.2 In addition to the numerous potential complications of STDs including pelvic inflammatory disease, ectopic pregnancy, cervical cancer, infertility, chronic pelvic pain, and meningitis, STDs increase a person's susceptibility to HIV infection.3,4
An important reason for the higher burden of HIV and other STDs among young adults is a higher prevalence of risky sexual behavior.5 Young adults are in an exploratory stage in their life that includes sexual experimentation.6,7 The college environment may facilitate high-risk sexual behavior in that it is an environment in which a high concentration of unmarried young people study and socialize. In contrast, there may be a higher prevalence of high-risk sexual behavior among young adults not in college if they are continuing high-risk behaviors begun in high school.8
HIV testing confers substantial benefits to people with HIV infection as well as to society at large. It allows people with previously undiagnosed HIV infection to obtain treatment, which in turn improves their survival. HIV testing benefits society by reducing the rate of transmission of HIV through at least 2 mechanisms. Diagnosis and subsequent treatment can lower people's HIV viral load that reduces the risk of HIV transmission,9 and there is evidence that people who become aware of their HIV diagnosis do decrease risky sexual behavior.10 Yet, in 2006, an estimated 47.8% of 13 to 24-year-olds infected with HIV in the United States had an undiagnosed HIV infection.11
There is limited knowledge about HIV testing among nationally representative samples of young adults, particularly among those at high risk of HIV infection. The objective of our study was to use data from the National Survey of Family Growth (NSFG) Cycle 6, which surveyed a nationally representative sample of 15- to 44-year-olds, to characterize and contrast HIV testing and risk behavior among 18- to 22-year-old students and nonstudents.
MATERIALS AND METHODS
The NSFG Cycle 6 was a survey that used a nationally representative multistage area probability sample of 12,571 participants aged 15 to 44 years living in households from March 2002 to February 2003. The survey included questions about sexual behavior. It was conducted by computer-assisted personal interviewing (CAPI) as well as audio computer-assisted self-interview (ACASI) methods.12 In this study, the analyzed CAPI survey questions included demographic questions and history of HIV testing, and the analyzed ACASI survey questions included those regarding high-risk sexual behaviors and history of STDs. The response rate of the survey was 79%.12
The sampling design has been described in detail elsewhere,12 but briefly the United States was divided up into 2402 primary sampling units (PSUs). Of these, a sample of 121 PSUs was selected, and within each selected PSU, a secondary sampling unit, a neighborhood, or set of adjoining blocks was selected. From each selected secondary sampling unit, a sample of addresses was selected. If more than one eligible person was living at an address, 1 person was selected. Household members also included people who were living away from the household, such as students in a college dormitory, fraternity, or sorority. Students not living in the household were interviewed at their college residence (e.g., dorm) or were interviewed at the household during school breaks.12
The CAPI and ACASI files were merged using the participant numbers. Analysis was limited to participants aged 18 to 22 to capture the age groups most likely to be currently in undergraduate education. Seventeen-year-olds were not included because only 7 (1.6%) of the 17-year-old participants were undergraduates. Participants were classified as students if they reported being in or having completed at least the first year of college and whether they answered “yes” to the following question regarding school attendance: “.... I'd like to talk only about regular school. By regular school I mean elementary, junior high, high school, college or graduate school. Are you now going to, or on vacation from, regular school? ” All others were classified as nonstudents. Participants who had completed ≥16 years of education and were no longer in school were excluded from the analysis.
The Centers for Disease Control and Prevention's (CDC) recommendations listed several groups as at high risk of HIV infection and in need of at least annual HIV screening; these include people who (1) have had >1 one sex partner since their most recent HIV test, (2) have had a sex partner with >1 sex partner since their most recent HIV test, (3) have sex partner(s) with HIV infection, (4) exchange sex for drugs or money, and (5) have sex partner(s) who inject drugs.13 Furthermore, the CDC recommends that people who seek treatment for STDs should be screened routinely for HIV.13 Therefore, participants were classified as having high-risk HIV behavior and thus should have had an HIV test within the year if during that year they had >1 sex partner, had been treated for an STD, had exchanged sex for drugs or money, or had a sex partner with HIV infection or who was not monogamous or who injected drugs. The NSFG ascertained information about HIV testing by first asking whether the study participant had ever donated blood because all blood donated since 1985 has been tested for HIV.14 There was also a question about any HIV testing excluding blood donations. Finally, there was a question about the month and year of the most recent HIV test excluding blood donations. If the month and year of the test was within 12 months of the month and year of the interview date, the person was classified as having a test within 12 months before the study. Because the CDC recommends HIV testing so that people can determine whether they are infected and because people who donate blood may not be aware that their blood is tested for HIV,14 and because there was no question about the date of the most recent blood donation, the analysis of HIV testing within the 12 months before the interview excluded blood donation-related testing.
Because the NSFG uses a complex survey design, analyses were conducted with sampling weights and sampling design variables provided by the National Center for Health Statistics and estimation of sampling errors using the Statistical Analysis System (SAS) survey frequency procedure (SAS 9.1, Cary, NC). All reported proportions are the weighted proportions and are therefore estimates for the 18- to 22-year-old population in the United States. Adjusted odds ratios (OR) were calculated in a design-based logistic regression model controlling for age, gender, race/ethnicity, and marital status, using the SAS survey logistic procedure. Data were not reported if the relative standard error was greater than 30% or the denominator was based on fewer than 50 cases.15 Missing values that had been imputed by the National Center for Health Statistics were used. Written informed consent was obtained from all participants. Cycle 6 of the NSFG was approved by the ethics review boards of the CDC and University of Michigan.14 The Florida International University Institutional Review Board deemed the current study exempted from Institutional Review Board approval.
On the basis of 2007 participants who met the eligibility criteria for this study, an estimated 53.0% of 18- to 22-year-olds were undergraduate students, and there was a similar distribution of males and females for both students and nonstudents (Table 1). Students were more likely to be unmarried and of non-Hispanic white race/ethnicity than nonstudents.
An estimated 37.5% of 18- to 22-year-olds had high-risk HIV behavior during the 12 months before the survey, and the percentage did not vary by student status (Table 2). The only demographic variables that were significantly associated in the regression model with high-risk HIV behavior were sex and marital status. Being female and married was protective (adjusted OR: 0.78; 95% confidence interval [CI]: 0.62-0.98 and 0.37; 95% CI: 0.20-0.70, respectively) (data not in table).
Most of the high-risk HIV behavior was attributed to either having >1 sex partner or the sex partners being nonmonogamous. An estimated 31.4% had >1 sex partner during the year, and there was no difference between students and nonstudents (Table 2). An estimated 16.0% had nonmonogamous sex partners, and this was less commonly reported by students than nonstudents (adjusted OR: 0.71; 95% CI: 0.52-0.97). An estimated 5.6% reported treatment for an STD during the 12 months before the study, and there was no difference by student status. The most commonly reported diagnosis for the entire group was genital warts followed by chlamydia. An estimated 2.4% reported exchanging sex for money or drugs, and 2.6% reported having sex with someone who injected illegal drugs. Comparisons could not be made between students and nonstudents because these behaviors were rare. The behaviors of having sex with a partner known to be infected with HIV, and injecting drugs within the previous year were so rare that estimates could not be generated.
The estimated percentage of 18- to 22-year-olds who were ever tested for HIV, either in a clinic setting or as part of a blood donation, was 53.7% (Table 3). The estimated percentage who were ever tested for HIV excluding blood donation-related testing was 34.2%, and was lower among students than nonstudents after adjusting for age, sex, race/ethnicity, and marital status (adjusted OR: 0.54, 95% CI: 0.40-0.73). There was no difference between students and nonstudents in nonblood donation-related HIV testing during the 12 months before the survey (adjusted OR: 0.76, 95% CI: 0.55-1.05). Of those who were ever tested outside of blood donation, 38.5% reported that a health care professional had spoken with them about HIV after they had the HIV test (data not in table).
Among those for whom at least an annual HIV screening was recommended because of high-risk HIV behavior, 28.3% had an HIV test within the year, and this did not vary according to student status (Table 3). Hispanics and non-Hispanic blacks in the group reporting high-risk HIV behavior were more likely to be tested than non-Hispanic whites, but the difference for non-Hispanic blacks did not reach statistical significance despite a relatively large effect (adjusted OR: 1.97; 95% CI: 1.30-2.98; adjusted OR: 1.62; 95% CI: 0.93-2.82, respectively) (data not in table). Of the 3 most common behaviors included in the definition of high-risk behavior, treatment for an STD was most strongly associated with having an HIV test (adjusted OR: 3.96; 95% CI: 2.31-6.80), followed by >1 sex partner (adjusted OR: 2.77, 95% CI: 2.08-3.68), and sex partner not monogamous (adjusted OR: 2.19; 95% CI: 1.57-3.05) (data not in table). However, only 44.8% (95% CI: 31.6%-58.1%) of those who reported STD treatment had an HIV test within the previous year. With regard to HIV testing, there were no statistically significant interactions between student status and reporting high-risk HIV behavior, including each of these most common specific behaviors.
In this report of HIV testing and risk behavior of a nationally representative sample of 18- to 22-year-old students and nonstudents, there were 2 main findings. First, the frequency of HIV testing was low for students and nonstudents, even among those at elevated risk. Second, high-risk HIV behavior was relatively common, mostly because of participants or their sex partners having multiple sex partners.
Approximately one-third of participants reported ever having been tested for HIV outside of blood donation. This result is very similar to that reported for 18- to 24-year-old students (32.1%) in the 1995 National College Health Risk Behavior Survey, a nationally representative sample of students.16 It is also similar to that (28.2%) reported in the recent (spring 2008) American College Health Association National College Health Assessment (NCHA), a survey of students from self-selected colleges and universities.17 No published study of a nationally representative sample of nonstudent young adults was found reporting lifetime testing. HIV testing may vary on the basis of enrollment in an academic program because students usually have access to student health services in addition to community services.
The percentage of participants tested within 1 year excluding blood donation in the current study was just 18.1% and was very similar to the 18.8% reported by 18- to 25-year-olds who had ever had sexual intercourse and who participated in Wave III (2001-2002) of the National Longitudinal Study of Adolescent Health.18 In that as well as in the current study there was no significant difference in the prevalence of testing between students and nonstudents.
Importantly in the current study, we found that only 28.3% of those young adults at elevated risk of HIV/AIDS were tested during the previous year. We were unable to locate another study of testing among a representative sample of young adults at elevated risk of HIV/AIDS. However, among a representative sample of 18- to 64-year-old adults who took part in the 2000-2005 National Health Interview Survey, 21% who reported a lifetime risk factor for HIV infection and 22.1% who perceived themselves to be at high risk of HIV infection had been tested for HIV during the previous year.19 Thus, low rates of testing among people at elevated risk for HIV infection seem to be a problem that is not limited to young adults.
In the NSFG, the response rate for college students living away from the chosen households (e.g., living in dorms) was 81.4%, slightly higher than that of all participants (79%) (J. M. Lepkowski, personal communication, February 10, 2010). Therefore, the results for students are likely to be representative of college students. However, it is not known whether the participating students were in a 2-year or 4-year college/university, complicating comparisons with studies such as the NCHA, which was administered mostly at 4-year colleges and universities.17 Results of the National College Health Risk Behavior Survey indicate that higher risk sexual behaviors, such as having >5 lifetime sex partners, were more common among attendees of 2-year institutions than those of 4-year institutions.16
HIV testing was self-reported, and it is possible that some participants may have been tested for HIV and forgotten or not been aware of the test. Alternatively, some participants might have thought that they were tested for HIV when in fact they had been tested for something else. To determine how these potential biases may have affected our results, we first determined what would happen if 20% of those actually tested did not report that they were tested and then whether 20% of those who had not been tested reported that they were tested. For the entire group, the prevalence of HIV testing ranged from 14.5% to 34.5%, respectively. Among those at high risk, the prevalence tested ranged from 22.7% to 42.6%. Therefore, even assuming a relatively large misclassification error, the majority of young adults at high risk were not tested for HIV infection.
At the time of Cycle 6 of the NSFG, the CDC was recommending routine HIV testing in high prevalence settings and targeted testing of high-risk groups in low prevalence settings.20 That there was no difference in testing between students and nonstudents at high-risk was somewhat surprising because in addition to community-based and clinic-based testing sites that students and nonstudents both have access to, many students have access to testing through student health services on campus.
More young adults were ever tested if blood donation-related testing was considered, but people who are only tested in a blood donation setting usually do not receive any HIV/STD prevention counseling. However, even among those who were tested outside of blood donation, less than half reported being counseled by a health care professional about HIV the last time they had an HIV test. Furthermore, in the Spring 2008 NCHA assessment, only 26.1% of students reported ever receiving any information about HIV from their college or university.17 Colleges are potentially in a good position to promote HIV testing and counseling because they serve a recognized group of people, can establish institutional policies, and serve a group of people that generally is interested in educating and improving themselves.21 Promotion of HIV testing is also needed for young adults who are nonstudents. In 2006, the CDC recommended routine HIV testing in all clinical settings and annual testing for people at increased risk.13 More than one-third of the participants in this study were in this category of increased risk. Efforts are needed to increase HIV testing, particularly among students and nonstudents at high risk of HIV infection. The current study uses data collected before the 2006 recommendations. Therefore, it will be important to assess whether HIV testing rates, especially among high-risk young adults, have improved since the 2006 HIV testing recommendations.
HIV testing rates among young adults could be low for 2 main types of reasons: (1) not being offered the test, and (2) not seeking/refusing the test. Young adults, often relatively healthy, may not access the health care system as frequently as older adults and thus be less likely to be offered an HIV test. In the National Survey of Adolescent Health, HIV testing rates were inversely associated with the length of time since the participant last accessed health care.18 Furthermore, rates were higher among those with public health insurance compared with those with private health insurance, suggesting that the type of provider (public vs. private) affects the likelihood of testing.18 Reasons for adults, including young adults, at high risk of HIV infection, to not seek the HIV test or refuse the test are low perception of risk, fear of a positive HIV test result, and confidentiality concerns.22–24 Because the approaches to address these 2 main types of barriers (not being offered the test vs. not seeking it out/refusing the test) are very different, it will be important to not only monitor testing trends, but also the reasons why people are not tested.
High-risk HIV behavior was common for students and nonstudents mostly because of having >1 sexual partner or the sexual partner being nonmonogamous. Young adults tend to have shorter sexual partnerships than older adults.5 They also tend to practice serial monogamy and believe that because they and their partner are monogamous at the time of their relationship, there is no risk of HIV or other sexually transmitted infections.25 Yet going from 1 to just 2 lifetime partners was found among 14- to 19-year-old females to increase the risk of a prevalent sexually transmitted infection from 19.7% to 38.1%.26 It is likely that percentage of participants classified with high-risk HIV behavior was an underestimate because the participants may have underreported socially undesirable behaviors, such as injection drug use, and participants may have been unaware of other sex partners that their sex partners had. Because testing rates were lower among those who did not report any high-risk HIV behaviors, if high-risk HIV behaviors were underreported, the percentage of people at high-risk of HIV who were tested would be even lower than the 28.3% we found. Additionally, people at highest risk, such as sex workers and those using injection drugs, are hardest to locate and may have been underrepresented in the survey. Therefore, this would also lead to the percentage of participants with high-risk HIV behavior being underestimated.
In conclusion, this study of a nationally representative sample of 18- to 22-year-olds indicated that high-risk HIV behavior was relatively common among young adults, regardless of student status. Despite this, even among those at high risk of HIV infection, only a minority of participants were tested within the year for HIV. Because of their young age, young adults infected with HIV have much to gain in terms of prolonged life if they become diagnosed and treated. Furthermore, undiagnosed HIV-infected young adults are likely exposing other young adults. Efforts to increase HIV testing and counseling and HIV/STD prevention education services for students and nonstudents still need to be enhanced. In addition, HIV testing and barriers to testing among young adults, especially among those at high risk of HIV infection, should be reassessed with nationally representative data to monitor changes in HIV testing patterns and guide prevention efforts in this age group.
1.Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA 2008; 300:520–529.
3.Bertozzi SM, Opuni M. An economic perspective on sexually transmitted infections including HIV in developing countries. In: Holmes KK, Sparling PF, Stamm WE, et al, eds. Sexually Transmitted Diseases, 4th ed. New York, NY: McGraw Hill Medical, 2008:13–25.
4.Fleming DT, Wasserheit JN. From epidemiological synergy to public health policy and practice: The contribution of other sexually transmitted diseases to sexual transmission of HIV infection. Sex Transm Inf 1999; 75:3–17.
6.Gayle HD, Keeling RP, Garcia-Tunon M, et al. Prevalence of the human immunodeficiency virus among university students. N Engl J Med 1990; 323:1538–1541.
7.Latman NS, Latman AI. Behavioral risk of human immunodeficiency virus/acquired immunodeficiency syndrome in the university student community. Sex Transm Dis 1995; 22:104–109.
8.Bailey JA, Fleming CB, Henson JN, et al. Sexual risk behavior 6 months post-high school: Associations with college attendance, living with a parent and prior risk behavior. J Adolesc Health 2008; 42:573–579.
9.Quinn TC, Wawer MJ, Sewankambo N, et al. Viral load and heterosexual transmission of human immunodeficiency virus type 1. N Engl J Med 2009; 342:921–929.
10.Marks G, Crepaz N, Senterfitt JW, et al. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: Implications for HIV prevention programs. J Acquir Immun Defic Syndr 2005; 39:446–453.
11.Campsmith ML, Rhodes PH, Hall HI, et al. Undiagnosed HIV prevalence among adults and adolescents in the United States at the end of 2006. J Acquir Immun Defic Syndr 2010; 53:619–624.
12.Lepkowski JM, Mosher WD, Davis KE, et al. National Survey of Family Growth, Cycle 6: Sample design, weighting, imputation, and variance estimation. National Center for Health Statistics. Vital Health Stat 2, 2006; 142:1-82. Available at: http://www.cdc.gov/nchs/data/series/sr_02/sr02_142.pdf
. Accessed April 2, 2008.
13.Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep 2006; 55(RR14):1–17.
14.Anderson JE, Chandra A, Mosher WD. HIV Testing in the United States, 2002. Hyattsville, MD: National Center for Health Statistics, 2005. Advance Data from Vital and Health Statistics. No. 363. Available at: http://www.cdc.gov/nchs/data/ad/ad363.pdf
. Accessed September 4, 2008.
15.Klein RJ, Proctor SE, Boudrealt MA, et al. Healthy People 2010 Criteria for Data Suppression. Hyattsville, MD: National Center for Health Statistics, 2002. Statistical Notes. No. 24. Available at: http://www.cdc.gov/nchs/data/statnt/statnt24.pdf
. Accessed September 4, 2008.
16.Centers for Disease Control and Prevention. Youth risk behavior surveillance: National College Health Risk Behavior Survey. MMWR CDC Surveill Summ 1997; 46(SS-6):1-54. Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/00049859.htm
. Accessed September 9, 2009.
17.The American College Health Association. American College Health Association-National College Health Assessment spring 2008 reference group data report (abridged). J Am Coll Health 2009; 57:477–488.
18.Nguyen TQ, Ford CA, Kaufman JS, et al. HIV testing among young adults in the United States: Associations with financial resources and geography. Am J Public Health 2006; 96:1031–1034.
19.Ostermann J, Kumar V, Pence BW, et al. Trends in HIV testing and differences between planned and actual testing in the United States, 2000-2005. Arch Intern Med 2007; 167:2128–2135.
21.Brener ND, Gowda VR. US college students' reports of receiving health information on college campuses. J Am Coll Health 2001; 49:223–228.
22.Awad GH, Sagrestano LM, Kittleson MJ, et al. Development of a measure of barriers to HIV testing among individuals at high risk. AIDS Educ Prev 2004; 16:115–125.
23.Kellerman SE, Lehman JS, Lansky A, et al. HIV testing within at-risk populations in the United States and the reasons for seeking or avoiding HIV testing. J Acquir Immun Defic Syndr 2002; 31:202–210.
24.Peralta L, Deeds BG, Hipszer S, et al. Barriers and facilitators to adolescent HIV testing. AIDS Patient Care STDS 2007; 21:400–408.
25.Shafii T, Burstein GR. Adolescent sexual health visit. Obstet Gynecol Clin North Am 2009; 36:99–117.
26.Forhan SE, Gottlieb SL, Sternberg MR, et al. Prevalence of sexually transmitted infections among female adolescents aged 14 to 19 in the United States. Pediatrics 2009;124:1505–1512.