IRWIN, DEBRA E. MSPH, PhD*; THOMAS, JAMES C. MPH, PhD‡; SPITTERS, CHRISTOPHER E. MD∥; LEONE, PETER A. MD§; STRATTON, JANICE D. MD†; MARTIN, DAVID H. MD¶; ZENILMAN, JONATHAN M. MD**; SCHWEBKE, JANE R. MD††; HOOK, EDWARD W. III MD††; (FOR THE STUDY GROUP)
From the *Health Advice Company and †Durham County Health Department, Durham, ‡University of North Carolina, Chapel Hill, §Wake County Health Department, Raleigh, North Carolina; ∥Snohomish Health District, Everett, Washington; ¶Louisiana State University School of Medicine, New Orleans; **Johns Hopkins University School of Medicine, Baltimore, Maryland; and ††University of Alabama, Birmingham
The authors would like to thank Willard Cates, MD, MPH, Sevgi Aral, PhD, King Holmes, MD, PhD, Louise Zimmer, MPH, and the Burroughs Wellcome Clinical Data Management Group for their valuable input.
Supported by Burroughs Wellcome Co.
The additional members of the Study Group were as follows: Burroughs Wellcome Co.: L. Gray Davis, PhD, Hugh H. Tilson, MD, Elizabeth B. Andrews, PhD, Mary Ann Mowat, MA, Monika Stender, PhD, Martha Monett, MPH, Jeff Johnston, MSPH, Michael Miles, Vanessa Williams, MS, Bob Davis, Linda Phillips, MSN, Nancy Girzaitis, Dee Connors, Kellie Souza, Gerald Garrett, Debbie Caporuscio, Donna Clover; University of North Carolina: Edith Parker, PhD; San Diego County Department of Health Services: Gil Munoz, MD, Robert Gunn, MD; Wake and Durham County Health Departments: Sara Stratton, MPH, Debbie Hollis, MPH, Kristi Foster, MPH, Martha Monett, MPH; Louisiana State University School of Medicine: Barbara Armentor, RNP; Johns Hopkins School of Medicine: Barbara Pare, MS, Sherell Jackson, Jay Nobel; Chicago Department of Health: Masai Ehehosi, Rosalie Cyrier; University of Alabama: Brenda Hunter.
Reprint requests: Debra E. Irwin, MSPH, PhD, Health Advice Company, 2515 E. Highway 54, 2200 Century Plaza, Suite 101, Durham, NC 27713.
Received for publication September 1, 1998, revised February 3, 1999, and accepted February 5, 1999.
BEHAVIORAL FACTORS are integral to the dynamics of sexually transmitted disease (STD) transmission. The risk of becoming infected can be reduced by decreasing the number of sex partners and sexual activity and by using condoms correctly and consistently. People who are infected can reduce transmission to others by seeking early treatment for curable infections, as well as by reducing sexual activity and using condoms. Little is known, however, about sexual behaviors in individuals with symptomatic STDs; most studies have looked at condom use before the acquisition of STD symptoms.
Perceived symptoms may affect treatment decisions and earlier entry into care.1-4 Patients with genital lesions may be more motivated to self-treat2 and seek medical care.1,2 For symptoms other than genital lesions, self-treatment may delay entry into care.2
One study reported that men may be more likely to limit sexual activity while symptomatic.1 In this study in Kenya, men with urethral discharge and genital ulceration reported fewer sex acts than men with other symptoms. However, this relationship did not hold true for women. Women were not only more likely to engage in sexual activity, but were also more likely to have more acts of intercourse than symptomatic men.
The purpose of this study was to identify predictors of sexual activity in patients with overt STD symptoms. A further issue under study was whether symptomatic patients who engaged in sex used condoms.
Settings and Enrollment
Patients seen at 10 public health STD clinics in seven cities (Baltimore, MD; Birmingham, AL; Chicago, IL; Durham, NC; New Orleans, LA; Raleigh, NC; and San Diego, CA) from August 1993 through June 1994 were asked to enroll in the study. Individuals 18 years and older who were either seeking treatment for STD symptoms or who had a known STD sexual contact were eligible to participate. Patients were excluded from the study if: (1) they were attending the clinic only for HIV counseling and testing, for routine gynecologic care, symptoms unrelated to STDs, or follow-up for an earlier diagnosis; (2) they had a severe communication problem; or (3) they had already participated in the survey. More detailed information on methods and data collection can be found in Irwin et al. (another primary goal of this study was to assess patterns of self-treatment among genital ulcer disease (GUD) patients and its effects on the duration of STD symptoms before medical care).2
Eligible patients electing to participate responded to an interviewer-administered questionnaire before clinician examination and diagnosis. The questionnaire collected information confidentially about demographic characteristics, perceived symptoms, symptom duration, and sexual activity and condom use while symptomatic.
Sexual activity since symptom onset was assessed in relation to the number of partners reported. Data on individuals who reported two or more sex partners was aggregated into a single category. Patients reporting one or more sexual partners while symptomatic were aggregated into categories of condom use: always, most of the time, some of the time, never. For the purposes of this study and throughout this report, sexual activity and condom use were assessed only while the patient reported STD symptoms and no assessment was made for sexual activity or condom use outside of the current STD symptomatic time period.
The duration of a symptom was calculated as the time from reported symptom onset to the interview date (date of initial clinic appointment). The longest symptom duration was used for patients who reported more than one symptom. Genital ulcer disease (GUD) symptoms were defined as a patient report of any genital sore, blister, cut, scrape, ulcer, or lesion at the time of the clinic visit. The presence of a GUD diagnosis was based on a diagnosis of syphilis, genital herpes, chancroid, or GUD-type unspecified.
The relationship of sexual activity to other factors and the relationship of always using condoms were assessed with multivariable logistic regression. Variables included in the models were study site, race, gender, age, symptom duration, education, employment status, symptom number, self-treatment status, GUD symptoms, and GUD diagnosis.
A total of 3,090 patients met inclusion criteria and were asked to enroll in the study. Of these, 3,025 (97.9%) agreed to participate and were interviewed. Almost two thirds of the participants were male, although the range of male participants varied by site from 48% to 78%. Mean age was approximately 28 years, with a range of 18 to 73 years (median age range for all sites was between 24 years and 27 years). Approximately two thirds had graduated from high school, and nearly 57% were currently employed. Participants were predominantly African American, with the exception of the San Diego site, where the greatest proportion of subjects was Hispanic (more detailed information on demographic results of the study population can be found in Irwin et al., Table 1).2
One or more symptoms were reported by 2,508 (82.9%) of study participants; approximately half (52.6%) of the symptomatic patients reported only one symptom. Genital discharge was the symptom most frequently reported by both men and women (Table 1). The median duration from symptom onset to clinic visit was between 5 and 7 days depending on presenting symptoms and self-treatment status before clinic visit (data not shown).2
Approximately 60% of all respondents reported no sexual activity since symptom onset. A lower percentage of men (34.4%) than women (49.8%) reported sexual activity while symptomatic. However, a slightly larger percentage of men who reported sex had two or more partners than did women (Table 2).
Multivariate logistic regression found that factors associated with sexual activity while symptomatic include: duration of symptoms for 7 or fewer days, African American race, and male gender. African Americans were more than twice as likely to engage in sex as other groups, primarily Hispanics (OR = 2.1; 95% CI, 1.4-3.1). African Americans did not differ significantly from whites in this regard (OR = 1.1; 95% CI, 0.86-1.5). Symptomatic men were 1.5 times as likely to engage in sex as symptomatic women (95% CI, 1.3-1.8). The variable most strongly associated with sexual activity was symptom duration of 7 or fewer days (OR = 5.9; 95% CI, 5.0-7.2). Study site, education, employment, symptom number, self-treatment, GUD symptoms, and GUD diagnosis were not related to sexual activity while these patients were experiencing STD symptoms.
Half of all symptomatic patients who were sexually active stated they never used condoms. The percentage never using condoms was higher among symptomatic women (56.1%) than among men (44.8%). A slightly higher percentage of men than women reported always using condoms while symptomatic. More men than women also reported using condoms most of the time and some of the time (Table 3).
Multivariate logistic regression found that symptomatic patients who "always" used condoms were more likely to be 30 years or older (OR = 1.5; 95% CI, 1.1-2.2) and to have at least a high school education (OR = 1.5; 95% CI, 1.1-2.1). Study site, gender, race, education, employment, symptom number, duration of symptoms, self-treatment, GUD symptoms, and GUD diagnosis were not related to always using condoms while experiencing STD symptoms.
Among this population, almost two thirds of clients (60.3%) had not been sexually active while symptomatic. African Americans were more than twice as likely to have been sexually active as Hispanics; however, African Americans were only slightly more likely than whites to report being sexually active. Male gender was also associated with sexual activity while symptomatic. African Americans and men are also groups most likely to delay seeking care.2 As such, education addressed to the needs of these groups might be an important point of intervention in reducing the transmission of sexually transmitted diseases.
Among men reporting sexual activity while symptomatic, almost one third (30.7%) had two or more partners. Among women who had sex while symptomatic, only 16.5% reported sex with more than one partner. For many women, sexual activity while symptomatic may occur with a steady partner, such as a spouse or boyfriend. These women may be unable to refuse to have sex or insist on condom use with their long-term partner; for women, being married has been associated with engaging in sex while symptomatic.1 Further, the majority of women believed they had been infected by their husband or boyfriend.1
The strongest predictor of sexual activity while symptomatic was duration of symptoms of a week or less. That is, individuals who were engaging in sexual activity were also receiving medical attention earlier. Sexual activity while symptomatic was not related to disease symptom type (GUD symptoms versus non-GUD symptoms).
Among STD patients reporting sexual activity, less than one fifth (17.2%) stated that they always used condoms while symptomatic. Other authors have also reported extremely low levels of condom use among patients attending STD clinics.1,5 Negotiating condom use successfully requires a complex set of skills. Higher rates of condom use among women are associated with conversations with partners about condoms, strong risk reduction intentions, positive attributions about condoms, and access to condoms, but not to HIV risk behavior knowledge or perceived risk for HIV infection.6 Women in a steady relationship with one risky male partner may find it difficult to assert condom use over his objections. Men are more likely to argue that condoms decrease pleasure and agree that it is safe to have sex without a condom if you know your partner.7 Because condom use is partner-specific and condoms are generally used to prevent disease rather than for contraceptive purposes, use of condoms declines as relationships progress. Both men and women are more likely to report lower levels of condom use with a primary or regular sex partner.5-7
In this study, "always" using condoms was associated with age of 30 or older. It may be that although younger individuals are more likely to have ever used a condom,7,8 older individuals may tend to be more cautious and use condoms to reduce risk to their partners when they have overt symptoms.
Condom use was also associated with higher levels of education. Among this population, individuals with a high school education or more were significantly more likely to use condoms while symptomatic. This finding is consistent with other studies in which individuals with less than an eighth grade education were least likely to use condoms and most likely to say they never used condoms.7
Many individuals report using condoms "some" or "most" of the time, but not with every sex act. Condom use has also been shown to vary according to the sexual activity performed; rates of use vary for anal, oral, and vaginal sex.5,7 A limitation of this study is that type of sexual activity was not reported. In addition, this study did not assess the effect of symptom severity on sexual activity or condom use.
Further limitations of this study are that both sexual activity and condom use were self-reported. Respondents may incorrectly recall use or give socially desirable responses rather than accurately characterizing their sexual behavior.5 Recall error may be minimal, however, because respondents were only asked about short-term condom use. In addition, respondents were not asked about the type of partner with whom they were engaging in sex or the length of their relationship. Both of these factors have been shown to affect the likelihood of using condoms. In addition, this study did not assess the effect of symptom severity on sexual activity or condom use.
Individuals infected with an STD can reduce the likelihood of infecting their partners by seeking early treatment for a curable infection, abstaining from sex while infected or under treatment, and using a condom if they do have sex. Among an STD clinic population, almost two thirds refrained from sexual activity while symptomatic. A small subset of those who did engage in sex reported "always" using condoms. However, a significant number of people engaged in sex with two or more partners without using condoms. Others used condoms inconsistently. These people may infect someone new or reinfect the person from whom they received their infection if their partner has been cured in the meantime.
Assuming those who have sex but do not use condoms consistently understand sexual transmission, their behavior suggests either lack of concern for their partner's health or forces outweighing their healthful intentions that compel them to have unprotected sex. A woman, in particular, may not feel safe informing her partner that she is infected or insisting that her partner use a condom. This situation underscores the need for means of reducing disease transmission that are more under the control of women.
Research is needed into the reasons why people who know themselves to be infected nonetheless engage in unprotected sex. Barriers to seeking early treatment, following through with treatment, abstaining from sex while infected or under treatment, and using condoms when having sex need to be identified and reduced. The barriers and means of reducing them are likely to vary by factors such as gender, type of relationship, race, age, and education level. They may also vary by factors such as community size and anonymity in seeking care9,10.
The Health Belief Model provides a theoretical basis for research further exploring care seeking and risky behaviors while knowingly infected. The Model includes considerations of cues to action (such as symptom recognition), the perceived severity of symptoms and their sequelae, and the perceived benefits of, and barriers to, treatment11. Although the model also includes one's perceived susceptibility to a condition (e.g., becoming infected), it was not designed to include the susceptibility of one's sexual partner (i.e., the likelihood of the infected person transmitting infection to their sexual partner). Although perhaps not in the context of the Health Belief Model, this would be an important factor to study. Success in altering behaviors among infected people engaging in unprotected sex is of paramount importance in reducing disease transmission and infection rates in a community.
1. Moses S, Ngugi EN, Bradley JE, et al. Health care-seeking behavior related to the transmission of sexually transmitted diseases in Kenya. Am J Pub Health 1994; 84:1947-1951.
2. Irwin DA, Thomas JC, Spitters CE, et al. Self-treatment patterns among clients attending sexually transmitted disease clinics and the effect of self-treatment on STD symptom duration. Sex Transm Dis 1997; 24:372-377.
3. Hook EW, Richey CM, Leone P, et al. Delayed presentation to clinics for sexually transmitted diseases by symptomatic patients. Sex Transm Dis 1997; 24:443-448.
4. Fortenberry JD. Health care seeking behaviors related to sexually transmitted diseases among adolescents. Am J Public Health 1997; 87:417-420.
5. Zenilman JM, Weisman CS, Rompalo AM, et al. Condom use to prevent incident STDs: the validity of self-reported condom use. Sex Transm Dis 1995; 22:15-21.
6. Heckman TG, Sikkema KJ, Kelly JA, et al. Predictors of condom use and human immunodeficiency virus test seeking among women living in inner-city public housing developments. Sex Transm Dis 1996; 23:357-365.
7. O'Donnell L, San Doval A, Vornfett R, O'Donnell CR. STD prevention and the challenge of gender and cultural diversity: knowledge, attitudes, and risk behaviors among black and Hispanic inner-city STD clinic patients. Sex Transm Dis 1994; 21:137-148.
8. Upchurch DM, Ray P, Reichart 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; 19:175-180.
9. Schuster J, Thomas JC, Eng E. Bridging the culture gap in sexually transmitted disease clinics. N C Med J 1995; 56:256-269.
10. Thomas JC, Lansky A, Weiner DH, Earp JA, Schoenbach VJ. Behaviors facilitating sexual transmission of HIV and STDs in a rural community. AIDS Behav (in press).
11. Strecher VJ, Rosenstock IM. The health belief model. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. 2nd ed. San Francisco: Jossey-Bass Publishers, 1997.