The relationship between indicators of sexual compulsivity and high risk sexual practices among men and women receiving services from a sexually transmitted infection clinic.

Sexual compulsivity is associated with high rates of sexual behavior and may increase risks for sexually transmitted infections (STIs), including HIV infection. Individuals who are preoccupied with sex and lack control of their sexual impulses may engage in high-risk sexual acts despite the threat of HIV infection and other potential adverse outcomes (Gold & Heffner, 1998). Sexual compulsivity is a heterogeneous psychological construct that can encompass a preoccupation with sexual desires and behaviors to the degree that a person experiences disruptions in social relationships, occupational difficulties, and problems in daily living (Barth & Kinder, 1987; Black, Kehrberg, Flumerfelt, & Schlosser, 1997; Gold & Heffner, 1998). Sexual compulsivity is not a formal psychiatric diagnosis, and it is likely that sexual compulsivity has multiple forms and multiple etiologies. Our conceptualization of sexual compulsivity is not synonymous with sexual addiction, hypersexuality, or other clinically defined categories (Carnes, 1990; Schneider, 1994). Rather, we define sexual compulsivity as a propensity to experience sexual disinhibition and under-controlled sexual impulses and behaviors as self-identified by individuals. In the current research, we investigated the association between indicators of sexual preoccupation and poor sexual impulse control (sexual compulsivity) and risks for sexually transmitted infections in a sexually transmitted infections clinic sample.

In studies of people who are already infected and living with HIV-AIDS, indicators of sexual compulsivity correlate with continued sexual risk practices, including behaviors that transmit HIV to uninfected sex partners. Kalichman, Greenberg, and Abel (1997), for example, found that HIV-positive men who had recently engaged in unprotected sexual activities with multiple sex partners scored higher on a sexual compulsivity scale than did individuals who engaged in high-risk activities with only one sex partner. Benotsch, Kalichman, and Kelly (1999) also found that HIV positive men scoring higher in sexual compulsivity engaged in more frequent unprotected sex acts with more partners, reported greater use of cocaine in conjunction with sexual activity, and rated high-risk sex acts as more pleasurable. Finally, Benotsch, Kalichman, and Pinkerton (2001) investigated indicators of sexual compulsivity as a factor contributing to high-risk sexual behavior in a sample of HIV-positive men and women. Individuals who scored higher in sexual compulsivity engaged in unprotected anal and vaginal intercourse with more HIV-negative partners or partners of unknown HIV status compared to persons with lower sexual compulsivity scores. In this study, mathematical modeling of sexual risk behavior indicated that four times as many new HIV infections could be expected among the HIV-negative sex partners of people who report more indicators of sexual compulsivity compared to partners of people who report fewer indicators of sexual compulsivity. HIV-positive individuals scoring higher in sexual compulsivity were more likely to report cocaine use and scored higher on measures of psychopathology than individuals lower in sexual compulsivity. Taken together, these studies suggest that sexual compulsivity may be a factor in sexual risk behavior for some people living with HIV-AIDS and may therefore be important in predicting risk behaviors in people at risk but not yet infected.

The current study examined indicators of sexual compulsivity in a sample of persons at high risk for STIs by conducting three sets of analyses. First, we examined the prevalence of indicators of sexual compulsivity and the dimensional composition of sexual compulsivity among STI clinic patients. Next, we conducted descriptive analyses by comparing STI clinic patients who were characterized as either relatively higher or lower in sexual compulsivity on measures of substance use and substance use outcome expectancies (Leigh & Stall, 1993). For these analyses, we operationally defined higher sexual compulsivity as the within-gender 80th-percentile scores on a sexual compulsivity scale. Finally, we conducted regression analyses to test the independent effects of sexual compulsivity as a predictor of sexual risks after controlling for known correlates of sexual risk behavior. We hypothesized that (a) indicators of sexual compulsivity would be prevalent in an STI clinic sample, (b) individuals who scored higher in sexual compulsivity would be at higher risk for HIV and other STIs, and (c) indicators of sexual compulsivity would significantly predict sexual risk behaviors over and above established correlates of sexual risks for HIV and other STIs.

METHOD

Participants and Setting

Participants were 432 men and 193 women receiving services from the largest public health clinic in Milwaukee, WI. Although a city of moderate size (population 1.2 million), Milwaukee is characterized by ethnically diverse communities and an impoverished inner-city. Sexually transmitted infections surveillance has shown that Milwaukee has ranked highest among U.S. cities in chlamydia rates and 13th for gonorrhea (CDC, 1999). The majority of STIs (59%) and HIV (52%) in Wisconsin are diagnosed in Milwaukee. The clinic site for the study is the largest publicly funded STI clinic in Wisconsin and serves approximately 6,000 patients annually; the majority of clients are African American (83%), 11% are White, 5% are Latino, and less than 1% are of other ethnic backgrounds. The clinic provides free and confidential STI diagnostic and treatment services, including confidential HIV testing.

Measures

All assessment instruments were administered using audio computer-assisted structured interviews (ACASI). Participants viewed assessment items on a 15-inch color monitor, heard items read by machine voice using head-phones, and responded to items by clicking a mouse. Research has shown that ACASI procedures yield reliable responses in sexual behavior interviews (Metzger et al., 2000). The assessment included measures of demographic characteristics, sexual compulsivity, alcohol use outcome expectancies, substance use, and sexual behaviors.

Demographic characteristics. Participants were asked their age, years of education completed, income level, self-identified gender, and ethnicity.

Sexual compulsivity. We assessed sexual compulsivity using a scale developed by Kalichman et al. (1994) that consists of 10 Likert-type items that ask respondents to endorse the extent to which they agree with a series of statements related to sexually compulsive behaviors and thoughts. Items were originally derived from a self-help guide for sexual addictions (Carries, 1990; see Table 1 for items). The items were anchored on 4-point scales from 1 = not at all like me, to 4 = very much like me. The sexual compulsivity scale has previously shown good reliability and has demonstrated criterion-related validity (Benotsch et al., 1999; Kalichman et al., 1997; Kalichman & Rompa, 1995). Responses from the present sample were internally consistent, [alpha] = .88.

Alcohol use outcome expectancies. We assessed sexualized substance use outcome expectancies with a measure reported by Kalichman et al. (1998). Participants who indicated lifetime alcohol use (99%) completed an 8-item measure reflecting sexual enhancing outcome expectancies from drinking alcohol. We originally adapted items from a widely used general alcohol outcome expectancies measure (Brown, Goldman, Inn, & Anderson, 1980). Sexual outcome expectancy items did not overlap in content with the sexual compulsivity scale. Example items include "I feel horny or sexual after I have been drinking," "I am a better lover after I have been drinking," "Sex is better after I have been drinking," and "It is easier to get turned on sexually after drinking." Items were answered on 4-point scales ranging from 1 = strongly disagree, to 4 = strongly agree. The alcohol outcome expectancy measure was internally consistent, [alpha] = .91.

Substance use. Participants indicated whether they had used alcohol in the past 3 months and how often they drank during that period. We also assessed marijuana, crack cocaine, powder cocaine, amphetamine, injection drug, and other drug use in the past 3 months. Participants also indicated the number of times they had used alcohol and other drugs in sexual situations in the past month using open response formats to record the number of occurrences.

In addition, participants completed the Alcohol Use Disorder Identification Test (AUDIT; Saunders, Aasland, Babor, DeLaFuente, & Grant, 1993), a 10-item self-report instrument designed to identify individuals who are at risk for developing alcohol problems or who are experiencing such problems. The time reference of the AUDIT items is the past year. AUDIT total scores range from 0 to 40, and scores of 8 or above have been used to identify individuals who may be at risk for or who are experiencing alcohol problems (Conigrave, Hall, & Saunders, 1995). The AUDIT has been used extensively in research and is a reliable and valid measure (Allen, Litten, Fertig, & Babor, 1997).

We administered the Drug Abuse Screening Test (DAST-10) to detect potential drug abuse disorders. The DAST-10 is a brief version of the 28-item DAST designed to identify drug-use related problems in the previous year. It has demonstrated good internal consistency, [alpha] = .94, and test-retest reliability, r = .71. The DAST-10 has also demonstrated excellent predictive validity in identifying drug use disorders (Maisto, Carey, Carey, Gordon, & Gleason, 2000).

Sexual behaviors. We assessed frequencies of vaginal and anal intercourse, both unprotected (without condoms), and protected (with condoms) for participants' female and male partners over 1-month and 3-month retrospective periods. For 1-month reports of behavior, participants answered questions about their numbers of sex partners and rates of alcohol and drug use during sexual encounters in the past month. Rates of sexual behavior over a 1-month recall period are generally reliable (Schroder, Carey, & Vanable, 2003). For the 3-month behavioral recall, we used a partner-by-partner format for structuring rates of specific sexual behavior responses. Specifically, we asked participants to recall their most recent four sex partners in the previous 3 months. For each partner, participants indicated whether the person was a regular or main partner, or a casual or one-time partner, and told us the number of times they engaged in vaginal and anal intercourse with and without condoms for each partner. We gave participants calendars, and the ACASI assessment used backdated cues to assist participants in recalling their responses. Internal consistency checks were included in the assessment interview to reduce random response patterns. We summed rates for each behavior across partners and within partner types, separately for regular or main partners versus the casual or one-time partners. Partner-specific data also allowed us to compute the proportion of sexual acts occurring with casual and one-time partners relative to total sex acts. We also assessed the number of times that participants reported being diagnosed and treated for sexually transmitted infections in the previous 3 months, including current clinic visits, which may or may not have involved an STI.

Procedures

Participants were approached as they waited for clinic services and were asked to participate in an STI risk-reduction counseling project. All participants were attending the clinic for either sexually transmitted infection diagnosis or treatment services. The mean age was 35.7 years (SD = 10.4), 40% of participants had completed less than a high-school education, 40% had a high school education without college, and 88% had annual incomes under $20,000. Closely approximating the clinic population, the majority of participants were African American (85%), 9% were White, 3% were Latino, and 3% were of other ethnicities. Finally, 17% of participants reported a history of receiving treatment for psychiatric conditions, and 67% had been incarcerated.

Data Analyses

Three sets of data analyses examined sexual compulsivity among STI clinic patients. First, we performed analyses to describe participant responses to sexual compulsivity indicators. Frequencies, percentages, and mean responses to the sexual compulsivity indicators were computed separately for men and women. We performed a principle components factor analysis to examine the dimensional composition of sexual compulsivity. We extracted factors with eigenvalues greater than one and orthogonally rotated the factor solution using a varimax procedure.

In a second set of analyses, we compared persons who were higher and lower in sexual compulsivity on measures of substance use and sexual behavior. Consistent with previous research (Benotsch et al., 1999, 2001), we defined participants as higher and lower in sexual compulsivity based on an extreme score approach to the Sexual Compulsivity Scale. We used the 80th percentile to empirically define higher and lower sexual compulsivity, assuring at least one standard deviation separation between groups. Because men and women differed on sexual compulsivity scores, t (623) = 2.56, p < .01 (see Table 1), we used gender-separate distributions to define sexual compulsivity groups; 85 men and 38 women were defined as scoring higher on the sexual compulsivity scale. We compared sexual compulsivity groups on substance use variables, numbers of sexual partners, and sexual behaviors using multiple logistic regression analyses and controlling for participant gender and age. Results are reported as adjusted odds ratios with associated significance levels.

Finally, we conducted two linear regression analyses to test the independent effects of sexual compulsivity scores on two separate markers of sexual risk: (a) numbers of sex partners in the past month and (b) rates of unprotected intercourse with casual or one-time sex partners in the past 3 months, controlling for participant age, gender, and alcohol and drug use. In both regression analyses, we entered demographic characteristics, substance use variables, and sexual compulsivity scores in separate blocks to allow for inspection of incremental changes.

RESULTS

Results indicated variability in responses to the sexual compulsivity scale among STI clinic patients. As shown in Table 1, participants frequently endorsed indicators of sexual compulsivity. For example, 26% of the sample responded that their desires to have sex had disrupted their lives, 43% indicated their sexual appetite had gotten in the way of their relationships; and 43% reported that they thought about sex more than they would like. When examined separately for men and women, we found that men were significantly more likely to endorse six sexual compulsivity indicators than were women, including those indicators that referred to the social disruptiveness and disturbances of sexual compulsivity.

Results of the principle components factor analysis using a varimax rotation indicated a two-factor solution (see Table 1). The first factor, eigenvalue = 5.08, accounted for 50.8% of the variance and included four items that represented a social disruptiveness dimension whereby sexual compulsivity causes interpersonal relationship problems and social maladjustment. The second factor, eigenvalue = 1.07, accounted for 10.7% of variance and represented a personal discomfort dimension to sexual compulsivity. Factor scores, which are weighted linear composites with means of zero and standard deviations of one and are by definition internally consistent, were computed for each of the two sexual compulsivity factors: social disruption and personal discomfort. Comparisons of men and women on the two sexual compulsivity dimensions showed that men (M = .02) did not differ from women (M = -.04) on the social disruptiveness factor. However, men (M = .07) scored significantly higher than women (M = -.16) on the personal discomfort factor, t (623) = 2.56, p < .01.

Substance Use Among Higher and Lower Sexual Compulsivity Groups

In analyses controlling for participant gender and age, higher scores on the sexual compulsivity scale were associated with greater expectancies that substances will enhance sexual experiences. Higher sexual compulsivity scores were also related to greater use of alcohol and other drugs in sexual contexts in the previous month. Comparisons between higher and lower sexual compulsivity groups on substance use variables are shown in Table 2. Persons with higher sexual compulsivity scored higher on the alcohol (AUDIT) and drug abuse (DAST) screening tests. In addition, we found that persons reporting greater indicators of sexual compulsivity were significantly more likely to have used powder cocaine, crack cocaine, and inhalants in the previous 3 months.

Sexual Behaviors Among Persons With Higher and Lower Sexual Compulsivity

Comparisons of higher and lower sexual compulsivity groups, defined by the within-gender 80th-percentile scores, showed that STI clinic patients with higher sexual compulsivity scores reported more sexual partners over the previous month (see Table 3). Using partner-specific sexual behavior data recalled over the previous 3 months, we found that sexual compulsivity groups were not different in the total number of sex acts reported or the number of unprotected sex acts reported with main sex partners. However, individuals higher in sexual compulsivity reported greater rates of unprotected sex with casual or one-time sex partners than did the lower sexual compulsivity group. We also found that the proportion of all sex acts occurring with casual or one-time sex partners was greater for persons with higher sexual compulsivity scores compared to those with lower sexual compulsivity scores. Finally, group comparisons indicated that clinic patients with higher sexual compulsivity were significantly more likely to have been diagnosed with multiple (two or more) sexually transmitted infections in the previous 3 month period than were their lower sexual compulsivity counterparts.

Tests of Independent Effects of Indicators of Sexual Compulsivity on Sexual Risks

Because the association between indicators of sexual compulsivity and sexual behaviors may have been confounded by demographic characteristics and co-occurring substance use, we conducted regression analyses to examine the relationships between sexual compulsivity and sexual risks after controlling for participant gender, age, and substance use. We entered gender, age, AUDIT scores, DAST scores, alcohol use in sexual contexts, and drug use in sexual contexts in an initial block, followed by the two sexual compulsivity factor scores: social disruptiveness of sexual compulsivity and personal discomfort with sexual compulsivity. We performed separate analyses using (a) numbers of sex partners in the previous month and (b) rates of unprotected intercourse with casual or one-time sex partners as the dependant variables. Results of the two regression analyses shown in Table 4 indicate that male gender, alcohol use in sexual contexts, DAST scores, drug use in sexual contexts, and the social disruptiveness dimension of sexual compulsivity were significantly associated with greater numbers of sex partners in the past month, F(8,619) = 18.02, p < .001, [R.sup.2] = .189. Changes in incremental variance showed that the sexual compulsivity factor scores significantly contributed to the explained variance over and above demographic and substance use variables, F(2,619) = 5.90, p < .01.

Results of the regression analysis for rates of unprotected sex with casual or one-time partners also indicated a significant equation, F(8,619) = 5.25, p < .01, [R.sup.2] = .052. When we examined changes in variance accounted for, we found that sexual compulsivity scores explained a significant amount of variance in unprotected sex over and above participant gender, age, and substance use, F(2,619) = 3.17, p < .05. In this analysis, the social disruptiveness dimension of sexual compulsivity was the only significant predictor of rates of unprotected intercourse with casual or one-time partners.

DISCUSSION

The current study is among the first to examine indicators of sexual compulsivity as predictors of sexual risks among individuals vulnerable to HIV and other STIs. Although we used a convenience sample that was primarily African American and lower income, the participants in this study were at risk for STIs by virtue of their seeking STI diagnostic and treatment services. Although the study sample was not atypical of people receiving services for sexually transmitted infections from public health clinics, caution should be taken in generalizing the results to other populations. In particular, our sample over-represented African Americans, people with a history of incarceration, and people with sexually transmitted infections. The characteristics of individuals who scored higher on sexual compulsivity in the current study should not be considered representative of people who report greater indicators of sexual compulsivity in general. In addition, our definition of sexual compulsivity is not related to clinical diagnoses of compulsive disorders, paraphilias, or other clinical syndromes. Even people who score above the 80th percentile on the Sexual Compulsivity Scale cannot be considered pathological in any clinical sense. Rather, people with extreme scores on the Sexual Compulsivity Scale have self-identified multiple markers of sexual preoccupations and undercontrolled sexual impulses.

Consistent with previous research (Benotsch et al., 2001; Kalichman & Rompa, 2001), we found that men scored significantly higher on the Sexual Compulsivity Scale than women. Examination of individual indicators of sexual compulsivity showed that nearly one in three women attending the STI clinic experienced social disruption associated with sexual compulsivity as well as feelings of personal discomfort. For both men and women, the most prevalent indicators of sexual compulsivity involved disruptions to interpersonal relationships and occupational functioning attributable to undercontrolled sexual thoughts and behaviors. However, men consistently reported greater frequencies of indicators of sexual compulsivity than did women. Indeed, the distribution of sexual compulsivity scores for men was shifted in a positive direction relative to the distribution of women's scores. Gender differences on the composites resulting from the factor analysis suggest that men experience greater personal discomfort related to sexual compulsivity than do women. Understanding the nature of these gender differences and their meaning requires further research.

Sexually transmitted infection clinic patients who demonstrated higher scores on the Sexual Compulsivity Scale, operationally defined by the within-gender 80th percentile, were at greater risk for HIV and other STIs. Individuals with higher sexual compulsivity scores reported a significantly greater number of recent sexual partners than individuals with lower sexual compulsivity scores. Rates of sexual behaviors across all partner types and behaviors practiced with main or regular sex partners were not different for STI clinic patients with higher and lower sexual compulsivity. However, persons with higher sexual compulsivity scores reported more sex partners and engaged in greater rates of unprotected intercourse with casual sex partners. We also found that a significantly greater proportion of sex acts occurred with casual partners among people with higher sexual compulsivity relative to the lower sexual compulsivity group. These findings show that increased rates of risk behaviors observed in people with higher sexual compulsivity scores confer high risks for HIV and STIs, particularly risks associated with casual and one-time sex partners. Indeed, higher sexual compulsivity participants were nearly four times more likely to have multiple STI diagnoses in the 3 months preceding data collection. The pattern of results suggests that people with higher sexual compulsivity may go outside their main sexual partnerships to meet their sexual needs. Future research is needed to explain whether seeking casual sex partners is the result of differences in domestic arrangements, access to casual sex partners, novelty seeking, need for more frequent sex than main partners can provide, higher sex drives, sexual openness, or other aspects of sexuality.

Not surprisingly, risk behaviors in people with greater sexual preoccupations and poorer sexual impulse control co-occurred with a greater likelihood of problems associated with substance abuse. Higher sexual compulsivity was associated with a greater number of problems stemming from alcohol and drug use. Persons with higher sexual compulsivity were also significantly more likely to report recent cocaine and inhalant abuse, drugs that are associated with sexual risk taking (McKirnon, Ostrow, & Hope, 1996). Cocaine use in particular is closely related to sexual risk behavior because of its highly addictive nature, involvement in sex trade, and euphoria-inducing properties (Edlin et al., 1994). Perhaps most relevant to risk for HIV and STIs, however, was the finding that higher sexual compulsivity was related to greater use of alcohol and other drugs in sexual contexts and expectancies that substance use will enhance sexual experiences. Unfortunately, our study did not assess substance use by sexual partners in sexual contexts. Although sexual compulsivity scores and substance use were related, regression analyses demonstrated that the association between indicators of sexual compulsivity--particularly the interpersonal disruptiveness of sexual compulsivity--and sexual risks for HIV and STIs cannot be accounted for by substance use.

The overall prevalence of sexual preoccupation and poor impulse control in this lower income mostly African American STI clinic sample supports a vulnerability for STIs among people of similar demographic characteristics with persistent and undercontrolled sexual thoughts and impulses. If validated through further research, these results indicate an urgent need for STI preventive interventions targeted toward people who lack control of sexual thoughts, behaviors, and impulses. Traditional models of HIV risk reduction that are most often delivered in STI clinics rely on public health education and cognitive-behavioral skills building techniques. Unfortunately, these intervention approaches will likely prove insufficient for reducing the sexual risks of STI clinic patients who experience sexual preoccupations and poor impulse control. The most promising intervention models for this population may be those that integrate elements of mental health, substance abuse treatment, and sexual risk reduction. For example, behavioral self-management approaches used in cognitive behavioral therapy for sexual preoccupations and poor impulse control can be adapted for inclusion in STI risk reduction counseling. Public health clinics should also be prepared to refer their clients who express distress about feeling out of control of their sexual desire and behavior for help that goes beyond services that an STI clinic can provide (Kalichman et al., 1997).

Table 1. Mean Scores, Number (Percent) Endorsing, and Factor Loadings
for Indicators of Sexual Compulsivity

                                                    Men

Indicator                                M           N           %

My sexual thoughts and behaviors
  are causing problems in my life.     .64 (a)      167        39 (a)
My desires to have sex have
  disrupted my daily life.             .42          121        28 (a)
My sexual appetite has gotten
  in the way of my relationships.      .80 (a)      198        46 (a)
I sometimes fail to meet my
  commitments and responsibilities
  because of my sexual behaviors.      .46          137        32
I think about sex more than I
  would like to.                       .78 (a)      207        49 (a)
I find myself thinking about sex
  while at work.                      1.05 (a)      276        64 (a)
It has been difficult for me to
  find sex partners who desire
  having sex as much as I want to.     .69 (a)      170        40 (a)
I feel that my sexual thoughts
  and feelings are stronger than I
  am.                                  .48          136        32 (a)
I sometimes get so horny I could
  lose control.                        .59          153        36
I have to struggle to control my
  sexual thoughts and behavior.        .48          134        31
Mean (SD)                                              .63 (.65)
Median score                                           .45
80th percentile score                                  .90

                                                   Women

Indicator                                M           N           %

My sexual thoughts and behaviors
  are causing problems in my life.     .46 (b)       57        30 (b)
My desires to have sex have
  disrupted my daily life.             .34           40        21 (b)
My sexual appetite has gotten
  in the way of my relationships.      .58 (b)       69        36 (b)
I sometimes fail to meet my
  commitments and responsibilities
  because of my sexual behaviors.      .41           51        26
I think about sex more than I
  would like to.                       .53 (b)       59        31 (b)
I find myself thinking about sex
  while at work.                       .72 (b)       88        46 (b)
It has been difficult for me to
  find sex partners who desire
  having sex as much as I want to.     .48 (b)       50        26 (b)
I feel that my sexual thoughts
  and feelings are stronger than I
  am.                                  .43           46        24 (b)
I sometimes get so horny I could
  lose control.                        .52           63        33
I have to struggle to control my
  sexual thoughts and behavior.        .47           48        25
Mean (SD)                                              .49 (.60)
Median score                                           .25
80th percentile score                                  .80

                                                              Factor
                                              Total           loading

Indicator                                M           N           %

My sexual thoughts and behaviors
  are causing problems in my life.     .58          224         36
My desires to have sex have
  disrupted my daily life.             .40          161         26
My sexual appetite has gotten
  in the way of my relationships.      .73          267         43
I sometimes fail to meet my
  commitments and responsibilities
  because of my sexual behaviors.      .45          188         30
I think about sex more than I
  would like to.                       .70          266         43
I find myself thinking about sex
  while at work.                       .95          364         59
It has been difficult for me to
  find sex partners who desire
  having sex as much as I want to.     .63          220         36
I feel that my sexual thoughts
  and feelings are stronger than I
  am.                                  .47          182         29
I sometimes get so horny I could
  lose control.                        .57          216         35
I have to struggle to control my
  sexual thoughts and behavior.        .48          182         29
Mean (SD)                                              .59 (.64)
Median score                                           .35
80th percentile score                                  .90

                                         Factor loading

Indicator                                I           II

My sexual thoughts and behaviors
  are causing problems in my life.     .81          .23
My desires to have sex have
  disrupted my daily life.             .77          .29
My sexual appetite has gotten
  in the way of my relationships.      .74          .13
I sometimes fail to meet my
  commitments and responsibilities
  because of my sexual behaviors.      .73          .26
I think about sex more than I
  would like to.                       .23          .77
I find myself thinking about sex
  while at work.                       .16          .75
It has been difficult for me to
  find sex partners who desire
  having sex as much as I want to.     .13          .71
I feel that my sexual thoughts
  and feelings are stronger than I
  am.                                  .52          .62
I sometimes get so horny I could
  lose control.                        .47          .61
I have to struggle to control my
  sexual thoughts and behavior.        .52          .60
Mean (SD)
Median score
80th percentile score

Note. Values with different superscripts indicate significant
differences between men and women, p < .05.

Table 2. Substance Use and Alcohol Outcome Expectancies Among
Higher and Lower Sexual Compulsivity Groups

                                  More sexually     Less sexually
Behavior                           compulsive        compulsive

                                   M        SD       M        SD

Substance use in sexual
  situations (past month)
  Alcohol use                      3.3      9.4      1.9      4.7
  Other drugs                      3.4      8.4      1.3      3.5
Substance use (past 3 months)
  Marijuana                       56       46      212       43
  Powder cocaine                  53       43      142       29
  Crack cocaine                   46       37      130       26
  Amphetamine                      4        3        7        1
  Inhalants                        9        7        7        1
AUDIT scores                      10.9     10.3      6.6      7.5
DAST scores                        3.5      3.1      2.6      2.9
  Alcohol outcome expectancies    10.0      7.0     61.      59

                                  Adjusted
                                    OR           p <

Substance use in sexual
  situations (past month)
  Alcohol use                       2.0          .01
  Other drugs                       2.7          .01
Substance use (past 3 months)
  Marijuana                         1.2           ns
  Powder cocaine                    2.0          .01
  Crack cocaine                     1.8          .01
  Amphetamine                       2.3           ns
  Inhalants                         5.7          .01
AUDIT scores                        1.1          .01
DAST scores                         1.1          .01
  Alcohol outcome expectancies      1.1          .01

Note. Odds ratios (OR) adjusted for participant age and gender.

Table 3. Sexual Behaviors for Higher and Lower Sexual Compulsivity
Groups

                                        More sexually     Less sexually
Behavior                                 compulsive        compulsive

                                        M        SD       M        SD

Number of sex partners (past month)     2.6      5.7      1.5      1.9
Sexual behaviors (past 3-months)
  Total sexual intercourse             25.9     45.7     24.3     44.8
  Unprotected intercourse              21.2     42.2     18.8     41.2
  Unprotected sex with main/regular
    partners                           16.0     29.7     18.4     41.8
  Unprotected sex with casual/one-
    time partners                       6.8     29.5      3.4     14.5
  % sex acts with casual/one-time
    partners                           39.7     37.3     30.1     34.4

                                        N        %        N        %

Multiple STI diagnoses                  12       10       14       3

                                       Adjusted
Behavior                                 OR           p <

Number of sex partners (past month)      3.4          .01
Sexual behaviors (past 3-months)
  Total sexual intercourse               0.9           ns
  Unprotected intercourse                1.1           ns
  Unprotected sex with main/regular
    partners                             0.8           ns
  Unprotected sex with casual/one-
    time partners                        1.6          .05
  % sex acts with casual/one-time
    partners                             2.1          .01

Multiple STI diagnoses                   3.9          .01

Note. Odds ratios (OR) adjusted for participant age and gender.

Table 4. Summary of Regression Analyses Testing Effects of
Sexual Compulsivity on Sexual Risks

                                    Number of sex partners

Risk                                B       SE       [beta]

Age                               -.022    .020      -.043
Gender                            -.002    .001      -.109 *
AUDIT score                       -.001    .001      -.007
Alcohol use proximal to sex        .159    .004      -.156 *
DAST score                        -.012    .034       .247 *
Drug use proximal to sex           .124    .035       .184 *
Sexual compulsivity--social
  disruptiveness                   .028    .009       .118 *
Sexual compulsivity--personal
  discomfort                       .011    .009       .049

                                     Unprotected sex with
                                       casual partners

Risk                                 B       SE      [beta]

Age                                .027     .038       .029
Gender                             .018     .002       .028
AUDIT score                        .001     .003       .038
Alcohol use proximal to sex        .089     .065       .077
DAST score                         .011     .007       .075
Drug use proximal to sex           .120     .069       .098
Sexual compulsivity--social
  disruptiveness                   .043     .017       .099 *
Sexual compulsivity--personal
  discomfort                       .007     .017       .018

* significant at p < .01.

REFERENCES

Allen, J. P., Litten, R. Z., Fertig, J. B., & Babor, T. (1997). A review of the research on the Alcohol Use Disorders Identification Test (AUDIT). Alcoholism: Clinical Experimental Research, 21, 613-619.

Barth, R. J., & Kinder, B. N. (1987). The mislabeling of sexual impulsivity. Journal of Sex and Marital Therapy, 13, 15-23.

Benotsch, E. G., Kalichman, S. C., & Kelly, J. A. (1999). Sexual compulsivity and substance use in HIV seropositive men who have sex with men: Prevalence and predictors of high-risk behaviors. Addictive Behaviors, 24, 857-868.

Benotsch, E. G., Kalichman, S. C., & Pinkerton, S. D. (2001). Sexual compulsivity in HIV-positive men and women: Prevalence, predictors, and consequences of high-risk behaviors. Sexual Addiction and Compulsivity: Journal of Treatment and Prevention 8, 83-99.

Black, D. W., Kehrberg, L. L. D., Flumerfelt, D. L., & Schlosser, S. S. (1997). Characteristics of 36 subjects reporting compulsive sexual behavior. American Journal of Psychiatry; 154, 243-249.

Brown, S., Goldman, M. S., Inn, A., & Anderson, L. (1980). Expectancies of reinforcement from alcohol: Their domain and relation to drinking patterns. Journal of Consulting and Clinical Psychology, 48, 419-426.

Carnes, P. J. (1990). Sexual addiction: Progress, criticism, challenges. American Journal of Preventive Psychiatry and Neurology, 2, 1-8.

Centers for Disease Control and Prevention, Division of STD Prevention. (2000). Sexually Transmitted Disease Surveillance, 1999. Atlanta, GA: Author.

Conigrave, K. M., Hall, W. D., & Saunders, J. B. (1995). The AUDIT questionnaire: Choosing a cut-off score. Addiction, 90, 1349-1356.

Edlin, B., Irwin, K., Faraque, S., McCoy, C., Word, C., Serrano, Y., et al. (1994). Intersecting epidemic: Crack cocaine use and HIV infection among inner-city young adults. New England Journal of Medicine, 331, 422-427.

Gold, S. N., & Heffner, C. L. (1998). Sexual addiction: Many conceptions, minimal data. Clinical Psychology Review, 18, 367-381.

Kalichman, S. C., Adair, V., Rompa, D., Multhauf, K., Johnson, J., & Kelly, J. (1994). Sexual sensation-seeking: Scale development and predicting AIDS-risk behavior among homosexually active men. Journal of Personality Assessment, 62, 385-397.

Kalichman, S. C., Greenberg, J., & Abel, G. G. (1997). Sexual compulsivity among HIV positive men who engage in high-risk sexual behavior with multiple partners: An exploratory study. AIDS Care, 9, 441-450.

Kalichman, S. C., & Rompa, D. (1995). Sexual Sensation Seeking and Sexual Compulsivity Scales: Reliability, validity, and predicting HIV risk behaviors. Journal of Personality Assessment, 65, 586-602.

Kalichman, S. C., & Rompa, D. (2001). The Sexual Compulsivity Scale: Further development and use with HIV positive persons. Journal of Personality Assessment, 76, 379-395.

Kalichman, S. C., Tannenbaum, L., & Nachimson, D. (1998). Personality and cognitive factors influencing substance use and sexual risk for HIV infection among gay and bisexual men. Psychology of Addictive Behaviors, 12, 262-271.

Leigh, B. C., & Stall, R. (1993). Substance use and risky sexual behavior for exposure to HIV: Issues in methodology, interpretation, and prevention. American Psychologist, 48, 1035-1045.

Maisto, S., Carey, M. R, Carey, K., Gordon, C., & Gleason, J. (2000). Use of the AUDIT and the DAST-10 to identify alcohol and drug use disorders among adults with a severe and persistent mental illness. Psychological Assessment, 12, 186-192.

McKimon, D., Ostrow, D., & Hope, B. (1996). Sex, drugs, and escape: A psychological model of HIV-risk sexual behaviors. AIDS Care, 8, 655-669.

Metzger, D. S., Koblin, B., Turner, C. F., Navaline, H., Valenti, E, Holte, S., et al. (2000). Randomized controlled trial of audio computer-assisted self-interviewing: Utility and acceptability in longitudinal studies. American Journal of Epidemiology, 152, 99-106.

Saunders, J. B., Aasland, O. G., Babor, T. F., DeLaFuente, J. R., & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption II. Addiction, 88, 791-804.

Schneider, J. P. (1994). Sex addiction: Controversy within mainstream addiction medicine, diagnosis based on the DSM-III-R, and physician case histories. Sexual Addiction and Compulsivity, 1, 19-44.

Schroder, K., Carey, M. P., & Vanable, P. (2003). Methodological challenges in research on sexual risk behaviors: II. Accuracy of self-reports. Annals of Behavioral Medicine, 26, 104-123.

Manuscript accepted November 4, 2003

The National Institute of Mental Health (NIMH) Grant R01-MH61672 supported this research.

Address correspondence to Seth C. Kalichman, Department of Psychology, 406 Babbidge Road, University of Connecticut, Storrs, CT 06269; e-mail: [e-mail].

Seth C. Kalichman and Demetria Cain

University of Connecticut