723
Correlates of Human Immunodeficiency Virus Infection Risk Behavior in Male Attendees of a Clinic for Sexually Transmitted Disease Richard A. Jenkins,1,2,3 Pamela R. Jenkins,4,a Ellen D. Nannis,1,a Kelly T. McKee, Jr.,5 and Lydia R. Temoshok6,a
From the 1Henry M. Jackson Foundation, Rockville, and 2Uniformed Services University of the Health Sciences, Bethesda, Maryland; 3 Armed Forces Research Institute for Medical Science, Bangkok, Thailand; 4Henry M. Jackson Foundation and 5US Army Medical Corps, Fort Bragg, North Carolina; and 6World Health Organization Programme on Mental Health, Geneva, Switzerland
Human immunodeficiency virus (HIV) infection risk behavior was evaluated in a crosssectional survey of 400 male active-duty US Army personnel who presented at a sexually transmitted disease (STD) clinic with symptoms of acute urethritis. High-risk partners were common, and nearly one-quarter of the sample had previously had STDs. Logistic regression models examined correlates of HIV exposure risk, of inconsistent condom use, and of having partners with increased risk of HIV infection. Frequent partner turnover, sex “bingeing,” negative attitudes toward condom use, and engaging in sex during military leaves were important correlates of risk. Individuals with HIV infection risk behavior generally were cognizant of their risk for HIV infection. Implications for intervention are discussed.
Sexually transmitted disease (STD) treatment settings have begun to be recognized as important venues for HIV epidemiology and prevention efforts [1–8]. Military STD clinic settings may be of particular importance, because the incidence of most STDs tends to be higher among military populations than among their civilian counterparts, particularly during deployment or wartime (E. C. Tramont, unpublished data). Military personnel with STDs are sociodemographically similar to
Received 24 March 1999; revised 29 November 1999; electronically published 20 April 2000. Presented in part: 1996 National STD Prevention Conference, 9–12 December 1996, Tampa, Florida. This research was conducted in compliance with the Standards for Human Subjects of the American Psychological Association and received review and approval from the committees for the protection of human subjects of Womack Army Medical Center and the United States Army Medical Research and Materiel Command. Articles contained in this symposium represent work conducted under Research Projects RV-56 (“Surveillance and Analysis of Sexually Transmitted Disease Patterns at Fort Bragg, NC”) and RV-81 (“Prevention of Exposure to HIV and Other Sexually Transmitted Diseases in a Seronegative Military Population: A Comparative Study of the Safety and Efficacy of Intensive STD/HIV Preventive Interventions”), supported by Cooperative Agreement DAMD 17-93-V-3004 between the US Army Medical Research and Materiel Command and the Henry M. Jackson Foundation for the Advancement of Military Medicine. The views and opinions expressed here are those of the authors and do not necessarily reflect the official policy or position of the US Army, the US Department of Defense, or the Henry M. Jackson Foundation. a Current affiliations: Care Clinic, Fayetteville, NC (P.M.J.); private practice (E.D.N.); and Division of Clinical Research, Institute of Human Virology, University of Maryland Biotechnology Institute, Baltimore, MD (L.R.T.). Reprints or correspondence: Richard A. Jenkins, Ph.D., Centers for Disease Control and Prevention, 1600 Clifton Road NE, Mailstop E-37, Atlanta, GA 30333 (
[email protected]). Clinical Infectious Diseases 2000; 30:723–9 q 2000 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2000/3004-0015$03.00
civilian STD clinic attendees but, because of their increased mobility to HIV epicenters, may have more opportunities for sexual contacts with partners whose behaviors have placed them at increased risk for HIV infection [9–11]. STDs in military populations have implications for civilian STD and HIV infection prevention because of sexual encounters with civilians during deployments, temporary duty assignments, and leaves, as well as with civilian partners encountered during normal duty assignments. Links between STDs and HIV infection risk behavior in military populations have been investigated in only a small number of studies. A large (n = 18,000), representative probability sample of US Army personnel worldwide found that self-reported STDs were associated with negative attitudes toward condoms, high levels of alcohol use, large numbers of lifetime sexual partners, and partners with genital sores. Soldiers with STDs also were more likely to have sexual partners during military deployment, to report periods of increased sexual activity (“bingeing”), and to seek out partners in locales in which the HIV prevalence is high [12]. It is interesting that those who reported STDs were more likely to perceive themselves at risk for HIV infection and were more knowledgeable about HIV infection prevention than those who did not report STDs. A 1992 US Army STD clinic study at Fort Bragg, North Carolina (G. Farr, C. Magruder, and R. Foldesty, unpublished data) found high levels of HIV knowledge, low levels of recent condom use, and widespread negative attitudes toward condom use. STDs in deployed US Navy and Marine personnel have been associated with the following factors: partners who are sex workers, single or divorced marital status, and nonwhite racial background [11]. Service-wide Navy and Marine syphilis cases from the late 1980s showed similar patterns [10]. Generalization from Navy/Marine studies to land-based services is
724
Jenkins et al.
difficult, because so much of the reported risk behavior occurs during leaves when ships are in port, which typically follow long tours of duty at sea. Building on these previous investigations of HIV/STD risk behavior in military populations, the present investigation evaluated HIV infection risk behavior patterns in US Army STD clinic attendees. The objectives of the study were to characterize core risk groups in this population and to address the development and refinement of behavioral interventions for STD clinic settings.
Methods and Measures Participants. Participants were men who attended the Epidemiology and Disease Control (EDC) Clinic at Womack Army Medical Center, in Fort Bragg, North Carolina. They were recruited from consecutive active duty personnel who presented with symptoms of acute urethritis from December 1994 through November 1995 and had no prior status as HIV-seropositive. Four hundred patients agreed to participate and gave written informed consent. Data were collected as the baseline for an evaluation of 3 experimental behavioral interventions to reduce STD/HIV infection risk behavior and of an “as is” comparison group, which received standard treatment and counseling (the interventions are described elsewhere [13]). Military personnel who had previously tested HIV-seropositive were excluded from consideration in the study, and no new seropositive cases were identified. The study also included an investigation of laboratory STD tests not previously used for routine diagnosis, as described elsewhere [14]. Data collection occurred prior to clinic attendees’ assignment to specific interventions. A total of 94 patients declined to participate in the study. The most common reasons for declining were as follows: the time required for the study (43.6%); not wishing to take part in a study (26.6%); concern that the study would be too personal (5.3%); embarrassment (4.2%); discomfort about participating (3.2%); impending leave or change in duty (3.2%); lack of motivation to change behavior (3.2%); not wanting to sign a consent form (3.2%); not wishing to have further invasive procedures (i.e., additional laboratory tests performed as part of the study [see above]; 2.1%); and other, idiographic responses (8.6%). There were no significant demographic differences between those who declined to participate in the study and study participants. Procedures. All participants received counseling based upon the Centers for Disease Control and Prevention (CDC) guidelines that were then in force [15]. The counseling was administered by the clinic’s STD health care providers, regardless of whether participants also received 1 of the experimental behavioral interventions. The counseling emphasized adherence to medication regimens, returning for test of cure, the need to inform all sexual partners to seek testing and treatment, risk reduction through abstinence or using condoms, and early response to suspicion of disease. Measures. Before they received counseling from the health care providers, participants completed a series of self-administered questionnaires. This format was chosen because it has been shown to
CID 2000;30 (April)
elicit greater disclosure of risk practices than less anonymous survey methods, such as face-to-face interviews [16]. An HIV/STD health risk appraisal, which was administered on scannable forms, included risk behavior items from the Army-Wide AIDS Survey [12, 17, 18]. The following variables were assessed for the previous year: number of sex partners; types of partners; condom use with different types of partners; biological risk factors for HIV infection (past STDs, partners with genital sores, and sex during which nonmenstrual bleeding occurred); engaging in sex in locales in which the HIV prevalence is high; and sexual encounters in situations associated with high-risk behavior (deployment, leave, and temporary change in duty). Additional items derived from the Army-Wide AIDS Survey represented hypothetical correlates of risk-relevant behavior: attitudes toward condoms, alcohol and drug use, perceived risk of contracting HIV, normative social influences, and HIV knowledge. Participants also completed a short form of the Profile of Mood States [19], which assessed current levels of anxiety, anger, and depression; a short form of the UCLA Loneliness Scale [20], of which scores have been found to be associated with social-skill deficits, particularly those related to heterosexual dating skills [21]; and a series of items evaluating readiness to change condom behavior, partner selection, and alcohol intake, adapted from Prochaska and DiClemente [22]. Participants were requested to return at 2 weeks and 2 months after this initial, baseline session and completed additional assessments of risk behavior, readiness for change, treatment compliance, and project satisfaction. Only data collected at baseline are considered here. Full descriptions of follow-up procedures and rates of follow-up, as well as longitudinal analyses of the data, are detailed elsewhere [13]. Data analysis. Analyses were conducted by use of SAS version 6.12 (SAS Institute, Cary, NC) and SPSS Windows version 7.5 (SPSS, Chicago). Frequencies and descriptive statistics were computed, and x2 and Fisher’s exact tests were performed to determine proportional relations among variables. In addition, univariate ORs were computed, and multivariate logistic regressions were performed.
Results Demographic composition. The sample was relatively young, predominantly single, and largely black or white, with high school education or some post–high school training most common. The majority lived alone or with someone other than a sexual partner (table 1). Most were from the lower enlisted ranks (71.5%), whereas 27.8% were from the senior enlisted ranks and 0.8% were officers. It is believed that most officers and many senior enlisted personnel seek private STD care in the civilian community. The post has agreements with local public STD clinics such that military personnel who seek care in the civilian public sector are referred to the Fort Bragg EDC clinic. Overview of HIV infection risk–relevant behavior. Evidence of at least some HIV exposure risk was reported by most participants. More than one-third (34.5%) reported 120 lifetime
CID 2000;30 (April)
HIV Infection Risk
Table 1. Demographic characteristics of a sample (n = 400) of US Army personnel evaluated for HIV infection risk behavior. Characteristic Race/ethnicity Black White Latino/Hispanic Other Educational attainment High school or general equivalency diploma Some college or technical training College or higher degree Marital status Single, never married Married Separated or divorced Living arrangement Live alone Live with spouse Live with steady partner (nonspouse) Live with someone (not a sex partner) NOTE. years).
% of subjects 56.8 34.8 4.5 3 52.5 44.2 3.3 60.5 18.5 21 53.5 16 8.3 22.3
Median age of patients in the sample was 23 years (range, 18–43
sex partners, 29% reported 11–20 partners, 25% reported 5–10 partners, and 11.5% reported 1 high-risk partners (i.e., partners whose behavior placed them at increased risk of HIV infection). Of those with high-risk partners, almost one-third had 11 type of high-risk partner. High-risk partners included one-night stands (60.7%), sex workers (4.5%), anonymous partners (e.g., those met in parks or on the street [8.7%]), known or suspected HIV-positive partners (4.5%), and partners who injected drugs (2.7%). About one-third of study participants (34.8%) reported >1 sexual episodes involving a potential biological facilitator of HIV exposure during the past year. The most frequent such factor was STD (24.8%); 2.8% reported partners with genital or anal warts or sores, and 14% reported partners who had nonmenstrual bleeding. Inconsistent condom use with partners whose behavior put them at increased risk for HIV infection was reported by 40.8% of the overall sample. Participants reported high rates of inconsistent condom use with high-risk partners: one-night stands (71.6%), anonymous partners (59.4%), and sex workers (62.5%). Impulse control appeared to be a frequent problem in the sample. Nearly one-third (32.3%) reported that they had >1 episodes of being “sexually out of control” or of sexual “bingeing” during the past year. Alcohol use in conjunction with sex
725
was frequent; 39% reported drinking within 2 h of their last sexual episode. Drug use was reported by 8.8%, which was relatively high for a military population; however, cannabis accounted for most drug use, and only 0.5% reported any needle-sharing. With regard to sex in risk-relevant situations, 7.5% reported engaging in sex while deployed, and 37.5% reported engaging in sex during leave. About one-fourth (27.5%) reported having had sexual partners in either domestic locales (e.g., New York City or San Francisco) or overseas locales (e.g., Thailand or sub-Saharan Africa) with significant HIV epidemics. Univariate and multivariate analyses were performed to investigate covariates of 3 major risk profiles: having a high-risk partner, having biological facilitators of HIV exposure, and inconsistent condom use with high-risk partners. Partners with increased risk of HIV infection. Several significant correlates of having had >1 sex partners with increased risk of HIV infection in the past year (anonymous partner, onenight stand, HIV-positive partner, sex worker, or injection drug–using partner) are presented in table 2. These included 2 demographic factors, a number of risk behaviors, low levels of readiness to change HIV infection risk behaviors, high levels of perceived HIV infection risk, relatively high importance given to military social norms, and engaging in sex during military leaves and deployments. When significant univariate correlates were entered into a multivariate logistic regression, only 4 variables emerged as significant: having >5 sexual partners in the past year, sexual bingeing, higher levels of perceived HIV infection risk, and engaging in sex during military leaves. Biological factors facilitating HIV exposure. Analysis of potential biological facilitators of HIV exposure during the past year (STD diagnosis, engaging in sex during nonmenstrual bleeding, and engaging in sex with partners who have sores) yielded a number of significant correlates (table 3). These included military rank; high levels of loneliness; the presence of a number of risk behaviors; lack of readiness to use condoms and to reduce sex with partners whose behavior placed them at increased risk for HIV infection; and higher perceived risk of HIV. When these significant univariate correlates were entered into a multivariate logistic regression, only 2 variables emerged as significant: being from the lower enlisted ranks and a perception of being at relatively high risk for HIV infection. Inconsistent condom use with partners at increased risk for HIV infection. Inconsistent condom use with high-risk partners was defined as condom use that occurred less often than “always” with >1 category of partner at increased risk for HIV infection (one-night stand, anonymous partner, or sex worker) in the past year. Inconsistent users of condoms were contrasted with those who reported consistent condom use with high-risk partners and those who did not report high-risk partners. Univariate analyses yielded a number of significant correlates (table 4), which included race, marital status, loneliness, having >5 partners in the past year, alcohol use in conjunction with most
726
Jenkins et al.
CID 2000;30 (April)
Table 2. Significant correlates of having partners with increased behavioral risk of HIV infection during the past year among a cohort of US Army personnel.
Variable Military rank Lower- (191/286) vs. upper-rank enlisted or officer (53/111) Marital status Never married (172/242) vs. ever married (75/158) No. of sex partners in the past 12 mo >5 (115/124) vs. !5 (132/2760) Drinking alcohol before most recent sex Yes (114/156) vs. no (133/244) Sexual bingeing in the past 12 mo Yes (119/129) vs. no (128/271) Engaging in sex in a high-risk locale in the past 12 mo Yes (81/110) vs. no (166/290) Drug use in the past 12 mo Yes (31/35) vs. no (216/365) Readiness to avoid high-risk partners Precontemplation to preparation (189/334) vs. action/maintenance (58/66) Readiness to avoid alcohol use Precontemplation to preparation (160/242) vs. action/maintenance (87/158) Perceived likelihood of HIV infection Likely (88/114) vs. not likely (155/281) Military social norms considered important Yes (222/336) vs. no (25/64) Condom use not enjoyed Agree (170/253) vs. disagree (77/147) Engaging in sex during deployment in the past 12 mo Yes (26/30) vs. no (221/370) Engaging in sex at temporary-duty location in the past 12 mo Yes (24/29) vs. no (2223/371) Engaging in sex during leave in the past 12 mo Yes (126/150) vs. no (121/250) a
Univariate
Multivariate
OR (95% CI)
OR (95% CI)
1.40 (1.13–1.69) 1.50 (1.27–1.77) a
1.94 (1.70–2.21)
4.17 (1.79–9.69)
1.34 (1.15–1.56) a
1.95 (1.68–2.27)
6.67 (3.28–14.90)
1.29 (1.10–1.49) 1.50 (1.29–1.73) 1.55 (1.36–1.77) 1.20 (1.02–1.42) a
1.40 (1.21–1.62)
2.04 (1.09–3.80)
1.69 (1.23–2.32) 1.28 (1.08–1.53) 1.45 (1.23–1.71) 1.38 (1.14–1.66) a
1.74 (1.49–2.02)
3.13 (1.68–5.82)
P ! .05 in multivariate analyses.
recent sex, sexual bingeing, errors in general HIV knowledge, higher perceived HIV infection risk, lower degrees of readiness to change HIV infection risk behavior, greater perceived importance of military social norms, negative attitudes about condoms, and engaging in sex during military leaves. When these were entered into a multivariate logistic regression, several variables remained significant. These indicated that inconsistent condom use occurred more often among those for whom the following were factors: high levels of loneliness, 15 partners in the past year, sexual “bingeing” in the past year, greater perceived risk for HIV, perception that condoms are difficult to use when aroused, and engaging in sex during military leaves.
Discussion HIV prevalence among active-duty military personnel has generally been lower than that among civilians of comparable age [23]. These lower rates of incidence may reflect, in part, the effects of the military’s preinduction screening. On the other hand, the incidence of STDs is typically higher in military populations than among their civilian counterparts (E. C. Tramont, unpublished data), and they may be deployed to countries where HIV prevalence exceeds that in the United States. Hence, military personnel manifest factors for HIV infection risk that
could lead to increased exposure risk. An economic consideration for HIV control in the military is that HIV-seropositive military personnel may remain on active duty if they are medically able [24], and medical entitlements available for their long-term care may be costly. This investigation found a relatively high potential of HIV exposure risk in this sample of US Army STD clinic attendees. The majority reported engaging in sex with partners whose behavior placed them at high risk for HIV infection, and inconsistent condom use was common with these partners. In addition, about one-third reported biological factors that have been shown to facilitate HIV exposure risk. Engaging in sex with risky partners was associated with sexual bingeing, greater numbers of partners, perceived HIV infection risk, and engaging in sex during military leaves. Biological risk for HIV exposure was associated with being in the lower enlisted ranks and having greater perceived risk for HIV infection. Inconsistent condom use was associated with higher levels of loneliness; having had more sexual partners; sexual bingeing in the past year; greater perceived HIV infection risk; perceived difficulties in using condoms when aroused; and engaging in sex during military leave. Generalizing across the models, we noted that large numbers of sexual partners, sexual bingeing, perceived HIV infection
CID 2000;30 (April)
HIV Infection Risk
727
Table 3. Significant correlates of potential biological facilitators of HIV exposure during the past year among a cohort of US Army personnel.
Variable Military rank Lower- (110/286) vs. upper-rank enlisted or officer (28/111) UCLA loneliness scale score High (37/82) vs. low–normal (102/318) No. of sex partners in the past 12 mo >5 (58/124) vs. !5 (81/276) Drinking alcohol before most recent sex Yes (64/156) vs. no (75/244) Sexual bingeing in the past 12 mo Yes (54/129) vs. no (85/271) Engaging in sex in high-risk locales in the past 12 mo Yes (47/110) vs. no (92/290) Readiness to use condoms Precontemplation to preparation (120/322) vs. action/maintenance (19/78) Readiness to avoid high-risk partners Precontemplation to preparation (109/334) vs. action/maintenance (30/66) Perceived likelihood of HIV infection Likely (56/114) vs. not likely (81/281) a
Univariate
Multivariate
OR (95% CI)
OR (95% CI) a
1.53 (1.08–2.17)
1.66 (0.99–2.77)
1.41 (1.06–1.88) 1.59 (1.22–2.09) 1.34 (1.02–1.74) 1.34 (1.02–1.75) 1.35 (1.02–1.78) 1.53 (1.01–2.32) 0.72 (0.53–0.98) a
1.70 (1.31–2.22)
2.10 (1.32–3.35)
P ! .05 in multivariate analyses.
risk, and engaging in sex during military leaves recur as significant risk correlates. A number of civilian studies have cited partner turnover as a risk factor for STDs [8, 25, 26]. These same studies also have emphasized the contributory influence of alcohol use, although our study found this to be significant in univariate but not multivariate analyses. Compared with data from a 1992 survey of STD patients at the Fort Bragg EDC clinic (G. Farr, C. Magruder, and R. Foldesty, unpublished data), condom use appeared greater, although negative attitudes toward condoms continued to be a problem. Factors that identified higher-risk participants in this study also were very similar to those factors that identified those with self-reported STDs in the Army-Wide AIDS Survey study [12], although that study evaluated STDs in the general Army population, with a much larger probability sample. In contrast to findings of a study of deployed US Navy personnel [11], contact with sex workers was relatively infrequent, which may reflect differential opportunities for having non-sex worker partners in deployed and nondeployed situations, as well as differences between land- and ship-based populations. On the other hand, our study found that engaging in sex during military leave was a significant factor and was an important context for risk behavior in the US Navy study. Hence, military leave may be important across services, although the specific types of partners may vary depending on the circumstances of a leave. In addition, the local prevalence of STDs, including HIV disease, is likely to influence actual risk. As in a study of deployed US Navy personnel, blacks reported more frequent consistent use of condoms with high-risk partners [11], although the significance of this difference was attenuated in multivariate analyses, where it appeared only as a nonsignificant trend (P ! .07 ). Race/ethnicity was not related
to frequencies of exposure risk behaviors or biological risk facilitators. Generalization of these findings to the Army population is limited because these young men were being seen for symptoms of acute urethritis, which already identified them as probably having engaged in HIV/STD risk behavior. Individual Army posts represent nonrepresentative samples of the Army population because they vary in terms of their strategic mission and the occupational specialties associated with that mission. Beyond this, there is no central STD registry in the Army, which further limits Army-wide generalization. Despite these constraints, some conclusions can be drawn about the representativeness of the sample, particularly in relation to the post demographics and to national STD data. The sample probably underrepresented officers, who are believed to seek STD treatment from private, fee-for-service providers. Relative to the local Army population, blacks were disproportionately overrepresented in the sample (56.8% of the sample, although they were only 21.7% of the male population at Fort Bragg at the time of the study [unpublished data, Department of Defense Manpower Data Center, Seaside, CA]). Even so, STD data from national samples [27] would suggest that the degree of overrepresentation of blacks in the sample was less than one would expect in a civilian STD clinic setting. Hence, it is likely that the magnitude of ethnic/racial disparity in STD prevalence is attenuated in the military [10, 11], although blacks are still more likely to be infected. This may reflect the economic leveling of the military (i.e., racial/ethnic differences here are not confounded by poverty), as well as military-specific risk factors (e.g., engaging in sex during deployment or leaves) that are independent of race or ethnicity. Some implications of our study’s findings are primarily specific to the military: the problem of military personnel engaging
728
Jenkins et al.
CID 2000;30 (April)
Table 4. Significant correlates of inconsistent condom use during the past year with partners whose behavior increased their risk of HIV infection among a cohort of US Army personnel.
Variable Race Black (80/227) vs. nonblack (83/173) Marital status Never married (112/2420) vs. ever married (51/158) UCLA loneliness scale score High (42/82) vs. low–normal (121/318) No. of sex partners in the past 12 mo >5 (84/124) vs. !5 (79/276) Drinking alcohol before most recent sex Yes (82/156) vs. no (81/244) Sexual bingeing in the past 12 mo Yes (81/129) vs. no (82/271) Test of HIV knowledge—general >1 errors (158/373) vs. 0 errors (5/27) Readiness to use condoms Precontemplation to preparation (146/323) vs. action/maintenance (17/77) Readiness to avoid high-risk partners Precontemplation to preparation (43/66) vs. action/maintenance (120/334) Perceived likelihood of HIV infection Likely (64/114) vs. not likely (96/281) Military social norms considered important Yes (144/336) vs. no (19/64) Condoms difficult to use when aroused Agree (87/163) vs. disagree (76/237) Condom use not enjoyed Agree (116/253) vs. disagree (47/147) Condoms inconvenient to carry Agree (75/1550) vs. disagree (88/245) Engaging in sex during leave Yes (89/150) vs. no (74/250) a
Univariate
Multivariate
OR (95% CI)
OR (95% CI)
0.74 (0.58–0.93) 1.43 (1.10–1.87) a
1.89 (1.03–3.48)
a
3.08 (1.67–5.69)
1.35 (1.05–1.74) 2.37 (1.88–2.98) 1.58 (1.26–2.00)
a
2.08 (1.66–2.60)
2.16 (1.22–3.84)
2.29 (1.03–5.09) 2.05 (1.32–3.17) 1.81 (1.45–2.28) a
1.64 (1.31–2.07)
1.81 (1.05–3.10)
1.44 (0.97–2.15) a
1.66 (1.32–2.10)
1.80 (1.06–3.04)
1.43 (1.09–1.88) 1.35 (1.07–1.70) a
2.01 (1.59–2.53)
2.67 (1.56–4.58)
P ! .05 in multivariate analyses.
in high-risk sex during leave needs to be addressed, and some consideration should be given to using periods before largescale leaves as propitious times for delivering preventive interventions. Several factors that were significant in univariate analyses, such as engaging in sex during deployment and engaging in sex in high-risk locales, also should be addressed and would lend themselves to interventions designed for leaves and deployments. Focus groups conducted prior to this study indicated that soldiers felt it was important to emphasize the risk and consequences of HIV infection to motivate changes in behavior. However, given that perceived risk of HIV infection was associated with greater HIV infection risk, emphasizing vulnerability to such infection may be less likely to motivate adoption of behaviors to reduce risk. The sample was drawn from an Army post whose primary missions include special-forces warfare and airborne infantry. These occupational specialty areas may attract those with greater risk-taking proclivities. On the other hand, the need inherent in these missions to manage risk and uncertainty may provide useful frames for educational messages. Because assignment to particular Army posts may reflect specific missions and overrepresentation of certain occupational specialties, STD health care providers need guidance in tailoring interventions to local norms. Study findings that are more generalizable to civilian pop-
ulations suggest the need to address impediments to behavior change in patients with STDs, such as negative attitudes toward condoms, problems with impulse control that may lead to sexual bingeing, partner turnover, and choosing risky partners. Hence, efforts to emphasize the positive aspects of condom use, the choice of safer partners, and the reduction of impulsivity should be emphasized in STD clinic counseling. Interpersonal skills may be another area for consideration, as indicated by the significance of loneliness. This variable reflects a dissatisfaction with a lack of social ties and has been associated with deficits in heterosexual dating skills [21]. Overall, the findings reflect continuing needs to address skills-development and attitude-change in STD clinic populations rather than simply offering advice and information. Acknowledgments We thank the individual participants for making this study possible. We also acknowledge the assistance of the Fort Bragg EDC Clinic for facilitating implementation of this study. The study staff, Cathy Richards, Michael Miller, and Jackie Pollock, provided recruitment and data collection. Tor-Lai Wong, Robin Garner, and Ayah Johnson conducted data analysis and provided statistical assistance. Database management was facilitated by Michele Goldschmidt. Edward Hutchins
CID 2000;30 (April)
HIV Infection Risk
and his staff at Healthier People Network developed the health-riskappraisal software. Institutional support of this project at key intervals was provided by John Lowe of the Henry M. Jackson Foundation, as well as by F. Dick Daniell, Deborah Birx, and Donald Burke, of the Walter Reed Army Institute of Research.
14. 15. 16.
References 1. Wasserheit JN. Epidemiological synergy: Interrelationships between human immunodeficiency virus and other sexually transmitted diseases. Sex Transm Dis 1992; 19:61–77. 2. Baker SA, Morrison DM, Gillmore MR, Schock MD. Sexual behaviors, substance use, and condom use in a sexually transmitted disease sample. J Sex Res 1995; 32:37–44. 3. Boyer CB, Barrett DC, Peterman TA, Bolan G. Sexually transmitted disease (STD) and HIV risk in heterosexual adults attending a public STD clinic: evaluation of a randomized controlled behavioral risk–reduction intervention trial. AIDS, 1997; 11:359–67. 4. Kamb ML, Fishbein M, Douglas JM, et al. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases: a randomized controlled trial. JAMA 1998; 280:1161–7. 5. NIMH Multisite HIV Prevention Trial Group. NIMH Multisite HIV Prevention Trial: reducing HIV sexual risk behavior. Science 1998; 280: 1889–94. 6. Wenger NS, Kuseseling FS, Shapiro MF. Misunderstanding of “safer sex” by heterosexually active adults. Public Health Rep 1995; 110:618–21. 7. Upchurch DM, Ray P, Reichert C, Celentano DD, Quinn T, Hook EW. Prevalence and patterns of condom use among patients attending a sexually transmitted disease clinic. Sex Transm Dis 1992; 19:175–80. 8. Zenilman JM, Hook EW, Shepherd M, Smith P, Rompalo AM, Celentano DD. Alcohol and other substance abuse in STD clinic patients: relationships with STDs and prevalent infection. Sex Transm Dis 1994; 21:220–5. 9. Emerson LAC. Sexually transmitted disease control in the armed forces, past and present. Mil Med 1997; 162:87–91. 10. Dembert ML, Finney LA, Berg SW. Epidemiology and reported syphilis among United States Navy and Marine Corps personnel: 1985–7. Sex Transm Dis 1990; 17:95–8. 11. Malone JD, Hyams KC, Hawkins RE, Sharp TW, Daniell FD. Risk factors for sexually transmitted diseases among deployed United States military personnel. Sex Transm Dis 1993; 20:294–8. 12. Rundell JR, Temoshok L, Blake SM, Sharp ES, Ledsky R. Factors associated with sexually transmitted diseases in the US Army [abstract Th77]. In: Program and abstracts of the 8th International Conference on AIDS (Amsterdam). Amsterdam: CONGREX Holland BV, 1992. 13. Jenkins PR, Jenkins RA, Nannis ED, McKee KT, Temoshok LR. Reducing sexually transmitted disease (STD) and HIV risk in a military STD clinic
17.
18.
19. 20. 21.
22.
23.
24.
25.
26.
27.
729
setting: evaluation of a randomized preventive intervention trial. Clin Infect Dis 2000; 30:730–5 (in this issue). McKee KT Jr, Jenkins PR, Garner R, et al. Features of urethritis in a cohort of male soldiers. Clin Infect Dis 2000; 30:736–41 (in this issue). Centers for Disease Control and Prevention. 1993 sexually transmitted diseases guidelines. Morb Mortal Wkly Rep 1993; 42:1–102. Catania JA, Gibson DR, Chitwood DD, Coates TJ. Methodological problems in AIDS behavioral research: influences on measurement error and participation bias in studies of sexual behavior. Psychol Bull 1990; 108: 339–62. Temoshok L, Blake SM, Rundell JR, Sharp ES, Ledsky R. HIV exposure risk–relevant behaviors in the US Army [abstract Th70]. In: Program and abstracts of the 8th International Conference on AIDS (Amsterdam). Amsterdam: CONGREX Holland BV, 1992. Blake SM, Sharp ES, Temoshok L, Rundell JR. Methodological considerations in developing measures of HIV risk–relevant behaviors and attitudes: an empirical illustration. Psychol Health Int J 1992; 6:265–80. McNair D, Lorr M, Droppleman LF. Profile of mood states. San Diego: Educational and Industrial Testing Service, 1971. Hays RD, DiMatteo MR. A short-form measure of loneliness. J Pers Assess 1987; 51:69–81. Gambrill E. Loneliness, social isolation, and anxiety. In: Mattaini MA, Thyer BA, eds. Finding solutions to social problems: behavioral strategies for change. Washington, DC: American Psychological Association, 1996: 101–19. Prochaska JO, DiClemente CC. Common processes of self-change in smoking, weight control, and psychological distress. In: Schiffman S, Wills T, eds. Coping and substance use: a conceptual framework. New York: Academic Press, 1986:345–63. Renzullo PO, McNeil JG, Wann ZF, Burke DS, Brundage JF, US Military Medical Consortium for Applied Retroviral Research. Human immunodeficiency virus type-1 seroconversion trends among young adults serving in the United States Army, 1985–1993. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 10:177–85. US Army Regulation 600-110. Identification, surveillance, and administration of personnel infected with human immunodeficiency virus [HIV]. Alexandria, VA, US Publication and Printing Command, 22 April 1994. Hook EW, Reichert CA, Upchurch DM, Ray P, Celentano D, Quinn TC. Comparative behavioral epidemiology of gonorrhea and chlamydia infections among patients attending a Baltimore, Maryland sexually transmitted disease clinic. Am J Epidemiol 1992; 136:662–72. Wasserheit JN, Aral SO. The dynamic topology of sexually transmitted disease epidemics: implications for prevention strategies. J Infect Dis 1996; 174(Suppl 2):S201–13. Centers for Disease Control and Prevention. 1998 guidelines for treatment of sexually transmitted diseases. Morb Mortal Wkly Rep 1998; 47(RR-1): 1–111.
