Likelihood of Condom Use When Sexually Transmitted Diseases Are Suspected

Abstract

Objective. To determine the event-level associations between perceived risk of sexually transmitted disease (STD) acquisition/transmission and condom use during penile–vaginal intercourse (PVI) among STD clinic attendees. Method. A convenience sample (N = 622) completed daily electronic assessments. Two questions were proxies of perceived risk: suspicion that the partner might currently have an STD and that “you” might currently have an STD. Participants reported whether condoms were used with PVI events in the past 24 hours. Generalized estimating equations determined the association between each of the perceived risk variables and event-level condom use. Results. For the model pertaining to suspicion of sex partner infection, there were 16,674 events of PVI, with condom use during 10,552 of these events. The effect of current suspicion was significant after adjusting for gender and whether participants identified as African American/Black (estimated odds ratio = 2.17, 95% confidence interval = 1.57-3.00, P = .0001). The model pertaining to suspicion of self-infection included 16,679 events of penile–vaginal sex, with condom use during 10,557 of these events. Again, the effect of current suspicion was significant after adjusting for gender and African American/Black race (estimated odds ratio = 2.05, 95% confidence interval = 1.43-2.40, P = .0001). Tests for interactions with gender and with race were nonsignificant (all Ps > .25). Conclusion. Using an event-level research design, strong associations were found between perceptions of STD risk and condom use in a clinical population. Health care providers and other professionals may indirectly promote condom use by helping clinic patients realistically evaluate their risk of having sex with infected partners or of being a source of STD transmission to others.

Keywords

condoms,health protective behavior perceived susceptibility sex behavior

Evidence strongly supports that condoms have protective value against sexually transmitted diseases (STDs; Crosby & Bounse, 2011; Koss, Dunne, & Warner, 2009; Warner et al., 2005). Researchers have demonstrated that a complex web of factors at the individual, relational, and broader societal level affect condom use behavior (DiClemente, Salazar, & Crosby, 2007). Among these, individual-level factors have been the focus of many interventions directed at achieving behavioral change (DiClemente et al., 2007). First brought to the forefront by the health belief model (Janz & Becker, 1984), the construct of perceived susceptibility has been a central construct in many of these interventions. Perceived susceptibility refers to one’s subjective perception of the risk of contracting a disease or illness (Janz & Becker, 1984).

Despite strong research demonstrating the utility of the concept of perceived susceptibility prior to the mid-1970s, recent research has yielded inconclusive findings (Baume, 2000). With respect to perceptions of STD acquisition and transmission, published findings suggest limited utility of this predictor variable for models of condom use behavior (Eng & Butler, 1997; Sheeran, Abraham, & Orbell, 1999). This may be attributable to inadequate research methodology (Eng & Butler, 1997) rather than an inherent lack of association between perceived risk of STDs and condom use behavior (Kowalewski, Henson, & Longshore, 1997). For example, most studies of HIV risk perception are based on cross-sectional survey data, in which perceived susceptibility is measured using retrospective recall periods that are vulnerable to memory bias (Kowalewski et al., 1997). In contrast, event-level data transcend cause-and-effect limitations by using each specific sexual event as the unit of analysis, that is, behaviors are measured in real time, linking the predictor variable with the outcome variable in temporal sequence (Crosby, DiClemente, Yarber, Snow, & Troutman, 2008; Hensel, Stupiansky, Herbenick, & Dodge, & Reece, 2011; Shrier, Walls, Lops, & Feldman, 2011; Warner et al., 2005). Although one disadvantage of daily diary assessment is possible reactivity “the potential for behavior or experience to be affected by the act of assessing it” (Shiffman, Stone, & Hufford, 2008, p. 20), there is little evidence of significant reactive effects (for review, see Shiffman et al., 2008).

Furthermore, STD risk perception can be conceptualized at two levels. In the parlance of the health belief model, only an individual’s risk of acquisition of infection is included (Rosenstock, Strecher, & Becker, 1994). Because of the dyadic nature of sexual encounters, a second level of risk is critical: transmission of disease to a partner. When risk of transmission is considered only as a unidirectional process, only one half of the disease prevention equation is addressed. Nonetheless, perceived risk of STD transmission from self to partner has not been studied empirically. Previous research has also mainly assessed perceived susceptibility without specifying the context for responding (Baume, 2000) and only rarely focused on perceived risk of a specific partner. Finally, much of the research on perceived susceptibility has relied on predominantly White (Winfield & Waley, 2002) and/or college student samples (Noar & Zimmerman, 2005; Sheeran et al., 1999). Few investigations have examined the utility of the construct among ethnically diverse or among higher risk clinic-based samples in which individuals are more likely to report a history of STD infection (Manhart et al., 2004).

Accordingly, the purpose of this study was to determine the event-level association between perceived risk of both STD acquisition and transmission and actual condom use during penile–vaginal intercourse (PVI) among female and male STD clinic attendees. As gender has been documented to influence perceived risk (e.g., Robertson, Stein, & Baird-Thomas, 2006), this study also examined the moderating effect of gender in the association between perceived risk of STDs and condom use. Because the STD disease epidemic has heavily affected the Black African American population (Centers for Disease Control and Prevention [CDC], 2012), we also examined the possible moderating role of Black versus non-Black race.

Method

Study Sample

Data were collected from December 2007 through April 2011 as part of a larger National Institutes of Health–funded study of condom effectiveness. A convenience sample was enrolled. All participants were tested for Chlamydia, gonorrhea, and trichomoniasis by urine-based nucleic acid amplification tests; those testing positive were treated with single-dose therapy. Patients were recruited from five clinics caring for individuals at high risk for STD in three U.S. cities: a publicly funded STD clinic in the Southern United States, a publicly funded STD clinic in the Midwestern United States; an STD clinic of a large teaching hospital in the Northeastern United States, and two adolescent medicine clinics affiliated with a children’s hospital, both in the Northeastern United States. The STD clinics enrolled individuals aged 18 years and older, the adolescent clinics enrolled individuals as young as 15. Eligibility criteria included reporting PVI in the preceding 3 months, speaking English, willing to provide contact information, and providing written informed consent. Institutional review boards at the participating universities approved the study protocol with a waiver of parental consent for adolescents younger than 18 years.

Recruitment procedures varied slightly across the five clinics. At the two adolescent clinics, the study was listed on a research recruitment flag attached to the appointment paperwork of age-eligible patients. The research assistant used the flag to identify eligible patients. This chart-flagging system at adolescent clinics precluded us from calculating a participation rate for those sites. Across the two adolescent clinics, 135 participants were enrolled.

Across the three remaining clinics, 1,424 patients agreed to be screened for eligibility. Of these, 1,297 were eligible and invited to participate; 794 enrolled, yielding a participation rate of 61.2%. The total number of participants enrolled was therefore 929.

Participants were trained in the use of an electronic diary report using a password-protected personal digital assistant (PDA) programmed with the Configurable Electronic Real-Time Assessment System (Personal Improvement Computer Systems, Inc. Reston, VA). Each day, participants were prompted to answer the question of whether they had sex in the past 24 hours. Because the research question pertained to urethral/vaginal acquisition of STDs, our assessments concerned penile–vaginal sex. Based on formative research (Mehta, Sunner, Head, Crosby, & Shrier, 2011), we defined sex as “putting the penis in the vagina” and the end of sex as the male orgasm. If participants responded affirmatively, they were asked questions about each instance of sexual intercourse. They were also asked to enter a report about their sexual behavior directly into the PDA after each time that they had sex to maximize capture of data on every sex event. Most participants were able to complete the daily questions in less than 3 to 4 minutes. These PDA diaries were kept by participants for up to 180 days postenrollment; however, because of attrition completed diaries for the current study were only available for 622 participants.

Measures

Two key questions from the daily diary assessments were used as a proxy for perceived risk: (a) “Before you had sex did you have any reason at all to suspect that you might currently have an STD that might be transmitted to your partner during sex?” (suspicion-self) and (b) “Before you had sex did you have any reason at all to suspect that your partner might currently have an STD that might be transmitted to you during sex?” (suspicion-partner). Daily assessments also captured condom use at last sex: “Did you or your partner use a condom for the penis-in-vagina part of the sex event?”

Data Analysis

After conducting basic cross-tabulations, multivariate logistic regression models were used to test associations between suspicions of current infections and condom use for PVI. The models included gender and race (Black versus non-Black) as well as interaction terms (entered concurrently) between these covariates and the perceived risk variables. Because of the correlations inherent to multiple observations on the same participant (i.e., the use of sex events as the observation units), generalized estimating equations models, using an exchangeable correlation matrix, were used for all analyses. Analyses were performed with SPSS (Version 19.0).

Results

Descriptive Findings

The mean age of the sample was 29.9 years (SD = 10.45). Of the 622 participants, 59.3% identified as African American/Black and 63.1% were female. About one of every four (25.7%) participants was 20 years of age or younger at enrollment. Just more than one half (54.9%) of the participants reported being unemployed. About 4 of every 10 (43.8%) had a high school education or less.

Table 1 provides the descriptive bivariate information on event-level condom use pertaining to the two variables—suspicion-self and suspicion-partner, for the overall sample and broken down by gender. As shown, condoms were used for at least 80% of the 1,202 events when the sex partner was suspected of infection, whereas this value was far lower (58.9% to 66.7%) during the 15,478 events when partner infection was not suspected. Also, condoms were used for at least 80% of the 1,177 events occurring when participants suspected self-infection whereas this value was, again, far lower (59.0% to 66.5%) during the 15,508 events occurring when self-infection was not suspected.

Table 1.

Descriptive Statistics for Event-Level Condom Use According to Suspicion of Sexually Transmitted Disease (STD) in Partner and in Self, Total and by Gender.

Descriptive Statistic	N (%) Using Condoms When Suspicion of STD	N (%) Using Condoms When No Suspicion of STD
Suspect sex partner is infected	994/1,202 (82.7)^a	9,558/15,478 (61.8)
Suspect sex partner is infected (males only)	397/493 (80.5)	3,776/5,664 (66.7)
Suspect sex partner is infected (females only)	597/709 (84.2)	5,782/9,814 (58.9)
Suspect “I am” infected	977/1,177 (83.0)	9,580/15,508 (61.8)
Suspect “I am” infected (males only)	394/472 (83.5)	3,781/5,687 (66.5)
Suspect “I am” infected (females only)	583/705 (82.7)	5,799/9,821 (59.0)

Refers to the number of times condoms were used (the numerator) and the number of sexual events (the denominator).

Multivariate Findings

Table 2 presents the results of the two generalized estimating equation models used to estimate the adjusted odds ratios for the effect of suspicion-self and suspicion-partner on condom use. Suspicion-self and suspicion-partner were significantly associated with event-level condom use in the two models.

Table 2.

Adjusted Event-Level Associations Between Suspicions of STDs and Condom Use.

Association	EOR	95% CI	P
Model 1: 621 participants (16,674 events)
Suspect sex partner is infected	2.17	1.57-3.00	.0001
Gender	0.83	0.65-1.08	.16
Race	1.84	1.43-2.40	.0001
Model 2: 622 participants (16,679 events)
Suspect “I am” infected	2.05	1.43-2.92	.0001
Gender	0.83	0.64-1.07	.15
Race	1.80	1.40-2.30	.0001

Note. STD = sexually transmitted disease; EOR = estimated odds ratio;CI = confidence interval.

The model pertaining to suspicion-partner is shown in the top half of Table 2. This model included 16,674 events of PVI, with condoms being used during 10,552 (63.3%) of these events. During 1,202 of the events (7.2%), participants suspected his or her sex partner had an STD. As shown, suspicion-partner was significantly associated with condom use, with condom use being more than twice as likely during penile–vaginal sex events when this suspicion was present. Gender was not significantly related to condom use; however, race did achieve significance in this model. The model suggested that African American/Black men and women were about 80% more likely to use condoms. Tests for interactions between suspicion-partner and gender (P = .68) and suspicion-partner and race (P = .46) were not significant.

The model pertaining to suspicion-self is shown in the bottom half of Table 2. This model included 16,679 events of PVI, with condoms being used during 10,557 (63.3%) of these events. During 1,177 of the events (7.0%), participants suspected their sex partner had an STD. As shown, suspicion-self and race was associated with condom use. The model suggested that when participants suspected self-infection they were more than twice as likely to use condoms. The model also suggests that, beyond the effect of suspicion-self, condom use was about 80% more likely for participants identifying as African American/Black. Tests for interactions between suspicion-self and gender (P = .28) and suspicion-self and race (P = .83) were not significant.

Discussion

The findings from this novel investigation of women and men attending STD clinics suggest that perceptions of STD infection that individuals held just prior to having sex were associated with condom use for that specific sexual encounter. Specifically, the event-level analysis of clinic attendees demonstrated that condom use was nearly twice as likely to occur when the event of PVI involved suspicion that either the individual or his or her partner might currently have an STD that could be transmitted during sex. This association was evident for men and women, meaning that both genders should be the focus of intervention efforts regarding perceived risk of STD acquisition and transmission. Furthermore, this relationship was present regardless of whether participants identified as African American/Black. Also, irrespective of their perception of STD risks, African American/Black men and women were more likely than non–African American/non-Black men and women to use condoms.

Though published findings suggest limited value of the construct of perceived risk relative to condom use behaviors (Eng & Butler, 1997; Sheeran et al., 1999), this study showed robust utility of both perceived risk of self- and partner-STD constructs. Event-level assessments represent a methodological improvement over retrospective, more global assessments. This study used ecological momentary assessments of both predictor and outcome variables and assessed perceived risk at two levels (self-to-partner and partner-to-self). This measurement approach may account for inconsistencies between some of the previous research that failed to find an association (Baume & Middlestadt, 1996, as cited in Baume, 2000) between perceived risk of infection and condom use and the current investigation.

Though it is reassuring that participants who perceived there was a risk for transmission of STD were more likely to use condoms, it is potentially concerning that approximately 93% of the events involved no suspicion of risk for self-to-partner or partner-to-self transmission. This is particularly troubling given the sample was composed of newly tested clinic patients (although it should be acknowledged that we have no evidence on the actual STD risk of the participants). An abundance of research supports the idea that individuals are notoriously poor at accurately assessing their risk for STD infection (East, Jackson, O’Brien & Peters, 2007; Kershaw, Ethier, Niccolai, Lewis, & Ickovics, 2003; Kowalewski et al., 1997). Moreover, individuals may be unlikely to suspect that they might acquire an STD from a partner with whom they choose to have sex. Perceptions of a relationship as characterized by trust, intimacy and connection may invoke a false sense of security against infection (Bolton, McKay, & Schneider, 2010). Furthermore, despite higher rates of STD infection among Blacks (CDC, 2012), the rate of suspicion was not elevated among this subgroup of participants. Nonetheless, they were more likely to use condoms.

Study findings have implications for clinicians and other clinical staff working with STD clinic attendees. Health providers should initiate discussion with clients about their attitudes toward condoms and in particular, their views of whether condoms imply lack of intimacy or trust of a partner. STD clinical staff should seek to educate all attendees about the behaviors that put them at risk for STD acquisition or transmission. Furthermore, clinical staff should also help individuals to raise their perceptions of personal susceptibility to STD infection to levels commensurate with their actual risk practices (Elifson, Klein, & Sterk, 2010). In addition, clinicians could discuss potential benefits of increased personal responsibility for protecting partners from contracting STDs. Thus, the clinical encounter provides an important teachable moment to educate about risk and promote condom use, regardless of diagnosis. For those testing positive, further counseling regarding heightened risk of STD transmission during the posttreatment period is vital.

Clinicians and clinical staff should note that although perceived risk is not a sufficient basis for behavior change, it is nonetheless a necessary precondition of behavior change (Napper, Fisher, & Reynolds, 2011). Therefore, clinicians should engage in the provision of behavioral skills training related to condom negotiation, correct use of condoms, and the consistent use of condoms as well as the provision of condoms that are preferred by their clients.

Limitations and Future Research

This study of STD clinic attendees from five clinics across three cities is constrained by the validity of self-report, as is true for most sexuality research, and its generalizability to other populations may be limited. The participation rate (62.3%) was fairly low and unfortunately, we could not compare possible differences in background characteristics between those who enrolled in the study and those who did not. Another limitation was the high attrition rate, which reduced the data available for analysis; this is a recognized issue in studies using diary assessment (Shiffman et al., 2008; Schroder, Carey, & Vanable, 2003). Future research should investigate strategies to enhance compliance with diary completion. There is also the possibility of reactivity when participants engage in daily monitoring (Shiffman et al., 2008). However, Schroder et al. (2003), reviewing the advantages and disadvantages of daily diary assessment, highlighted the reduced memory bias inherent in this technique and recommended that researchers make greater use of diary assessment. The event-level analysis, although valuable, may also not equally represent all study participants in that individuals varied in the number of events they have contributed to the data set; however, the generalized estimating equation models accommodated this by using estimated odds ratios.

We did not collect any information about why individuals suspected that they or their partner might currently have any STD; future qualitative studies should be carried out to elucidate the possible origins of suspicion. Also, we did not obtain information about participants’ sexually transmitted infection history; future research might also investigate whether recent treatment of an STD affects subsequent perceptions of STD acquisition/transmission risk. Finally, future studies should examine the associations between perceptions of STD risk and condom use in other groups, for example, gay/bisexual populations and other ethnic groups.

Conclusions

Using an event-level research design, this study found that condom use was associated with participants’ suspicion that they or their partner might currently have an STD that could be transmitted during the sexual event. This association occurred regardless of gender and race. Health care providers, as a part of teaching information, motivation, and behavioral skills, should help individuals develop more realistic assessments of risk for transmission of STD.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for this project was provided by a grant to the first author from the National Institutes of Allergies and Infectious Diseases, Grant #5 R01 AI068119.

References

Baume

C. A.

(2000). The relationship of perceived risk to condom use: Why results are inconsistent. Social Marketing Quarterly, 6, 33-42.

Bolton

McKay

Schneider

(2010). Relational influences on condom use discontinuation: A qualitative study of young adult women in dating relationships. Canadian Journal of Human Sexuality, 19, 91-104.

Centers for Disease Control and Prevention. (2011, November). HIV among African Americans (Fact sheet). Retrieved from http://www.cdc.gov/nchhstp/newsroom/docs/fastfacts-aa-final508comp.pdf

Crosby

R. A.

Bounse

(2012). Condom effectiveness: Where are we now? Sexual Health, 9, 10-17.

Crosby

R. A.

DiClemente

R. J.

Yarber

W. L.

Snow

Troutman

(2008). An event-specific analysis of condom breakage among African American men at risk of HIV acquisition. Sexually Transmitted Diseases, 35, 174-178.

DiClemente

R. J.

Salazar

L. F.

Crosby

R. A.

(2007). A review of STD/HIV preventive interventions for adolescents: Sustaining effects using an ecological approach. Journal of Pediatric Psychology, 32, 888-906.

East

Jackson

O’Brien

Peters

(2007). Use of the male condom by heterosexual adolescents and young people. Journal of Advanced Nursing, 59, 103-110.

Elifson

K. W.

Klein

Sterk

C. E.

(2010). Factors associated with condom use among young adult ecstasy users. International Journal of Public Health, 55, 571-579.

Eng

T. R.

Butler

W. T.

(1997). The hidden epidemic: Confronting sexually transmitted diseases. Washington, DC: National Academy Press.

10.

Hensel

D. J.

Stupiansky

N. W.

Herbenick

Dodge

Reece

(2011). When condom use is not condom use: An event-level analysis of condom use behaviors during vaginal intercourse. Journal of Sexual Medicine, 8, 28-34.

11.

Janz

N. K.

Becker

M. H.

(1984). The health belief model: A decade later. Health Education Quarterly, 11, 1-47.

12.

Kershaw

T. S.

Ethier

K. A.

Niccolai

L. M.

Lewis

J. B.

Ickovics

J. R.

(2003). Misperceived risk among female adolescents: social and psychological factors associated with sexual risk accuracy. Health Psychology, 22, 523-532.

13.

Koss

C. A.

Dunne

E. F.

Warner

(2009). A systematic review of epidemiologic studies assessing condom use and risk of syphilis. Sexually Transmitted Diseases, 7, 401-405.

14.

Kowalewski

M. R.

Henson

K. D.

Longshore

(1997). Rethinking perceived risk and health behavior: A critical review of HIV prevention research. Health Education and Behavior, 24, 313-325.

15.

Manhart

L. E.

Aral

S. O.

Holmes

K. K.

Critchlow

C. W.

Hughes

J. P.

Whittington

W. L. H.

Foxman

(2004). Influence of study population on the identification of risk factors for sexually transmitted diseases using a case-control design: The example of gonnorrhea. American Journal of Epidemiology, 160, 393-402.

16.

Mehta

C. M.

Sunner

L. E.

Head

Crosby

R. A.

Shrier

L. A.

(2011). Sex isn’t something you do with someone you don’t care about: Young women’s definitions of sex. Journal of Pediatric and Adolescent Gynecology, 24, 266-271.

17.

Napper

L. E.

Fisher

D. G.

Reynolds

G. L.

(2011). Development of the Perceived Risk of HIV Scale. AIDS and Behavior, 16, 1075-1083.

18.

Noar

S. M.

Zimmerman

R. A.

(2005). Health behavior theory and cumulative knowledge regarding health behaviors: Are we moving in the right direction? Health Education Research, 20, 275-290.

19.

Robertson

A. A.

Stein

J. A.

Baird-Thomas

(2006). Gender differences in the prediction of condom use among incarcerated juvenile offenders: Testing the Information-Motivation-Behavior Skills (IMB) model. Journal of Adolescent Health, 38, 18-25.

20.

Rosenstock

I. M.

Strecher

V. J.

Becker

M. H.

(1994). The Health Belief Model and HIV risk behavior change. In DiClemente

R. J.

Peterson

J. L.

(Eds.), Preventing AIDS: Theories and methods of behavioral interventions (pp. 5-24). New York, NY: Plenum Press.

21.

Schroder

K. E.

Carey

M. P.

Vanable

P. A.

(2003). Methodological challenges in research on sexual risk behavior: II. Accuracy of self-reports. Annals of Behavioral Medicine, 26, 104-123.

22.

Sheeran

Abraham

Orbell

(1999). Psychosocial correlates of heterosexual condom use: A meta-analysis. Psychological Bulletin, 125, 90-132.

23.

Shiffman

Stone

A. A.

Hufford

M. R.

(2008). Ecological momentary assessment. Annual Review of Clinical Psychology, 4, 1-32.

24.

Shrier

L. A.

Walls

Lops

Feldman

H. A.

(2011). Correlates of incorrect condom use among depressed young women: An event-level analysis. Journal of Pediatric and Adolescent Gynecology, 24, 10-14.

25.

Warner

Macaluso

Austin

H. D.

Kleinbaum

D. K.

Artz

Fleeno

M. E.

. . . Hook

E. W.

(2005). Application of the case-crossover design to reduce unmeasured confounding in studies of condom effectiveness. American Journal of Epidemiology, 161, 765-773.

26.

Winfield

E. B.

Whaley

A. L.

(2002). A comprehensive test of the health belief model in the prediction of condom use among African American college students. Journal of Black Psychology, 28, 330-346.