Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Aug 1.
Published in final edited form as: J Acquir Immune Defic Syndr. 2010 Aug 1;54(4):360–367. doi: 10.1097/QAI.0b013e3181d01b14

HIV Infection and Women’s Sexual Functioning

Tracey E Wilson 1, Girardin Jean-Louis 2, Rebecca Schwartz 1, Elizabeth T Golub 3, Mardge H Cohen 4, Pauline Maki 5, Ruth Greenblatt 6, L Stewart Massad 7, Esther Robison 8, Lakshmi Goparaju 9, Stacy Lindau 10
PMCID: PMC2900377  NIHMSID: NIHMS181146  PMID: 20179602

Abstract

Objective

To compare sexual problems among HIV-positive and HIV-negative women, and describe clinical and psychosocial factors associated with these problems.

Design

Data were collected during a study visit of the Women’s Interagency HIV Study (WIHS). The WIHS studies the natural and treated history of HIV among women in the United States.

Methods

Between 10/01/2006 and 3/30/2007, 1,805 women (1,279 HIV-positive and 526 HIV-negative) completed a study visit that included administration of the Female Sexual Function Index (FSFI). In addition, the visit included completion of standardized, interviewer-administered surveys, physical and gynecological examinations, and blood sample collection.

Results

Women with HIV reported greater sexual problems than did those without HIV. Women also reported lower sexual function if they were classified as menopausal, had symptoms indicative of depression, or if they reported not being in a relationship. CD4+ cell count was associated with FSFI scores, such that those with CD4 ≤199 cells/µL reported lower functioning as compared to those whose cell count was 200 or higher.

Conclusions

Given research documenting relationships between self-reported sexual problems and both clinical diagnoses of sexual dysfunction and women’s quality of life, greater attention to this issue as a potential component of women’s overall HIV care is warranted.

Keywords: HIV, Women, Sexual Behavior, Sexual Problems

Introduction

Women with HIV are living longer and healthier lives with less comorbidity due to the effectiveness of antiretroviral treatments. Most women with HIV are sexually active following their diagnosis 1, 2, and have the opportunity to remain sexually active for more years due in part to these medical advances. Research on the topic of women’s sexual behavior in the context of HIV infection has primarily taken place with the objective of assessing HIV transmission risk dynamics among women and their partners, and has focused on condom use practices, determinants of sexual risk behavior, and unintended pregnancy 3–6. Issues related to other aspects of sexual health, including satisfaction with sexual relationships among women living with HIV infection, have received relatively little attention.

Among women, sexual function problems are characterized by subjective reports of difficulties or limitations in sexual interest, desire, arousal, orgasm and pain during sexual activity. A clinical diagnosis of female sexual dysfunction requires the presence of both a problem with sexual functioning and a subjective assessment of distress due to the problem. There is reason to believe that sexual problems occur more frequently among HIV-infected than uninfected women. Chronic health conditions and diseases such as diabetes, cancer, vascular disease, arthritis, and hypertension are associated with impaired sexual functioning, as are certain drugs that are used to treat these conditions 7, 8. Factors associated with sexual problems or dysfunction in the general population, such as mental health issues and substance abuse, are common among both men and women with HIV 9, 10. Further, concerns about HIV transmission, HIV treatment, including antiretroviral therapy, and treatment side effects may have deleterious effects on female sexual functioning 11–13.

Most studies of sexual function among women with HIV describe a burden of sexual problems in this population14, 15, although published information on the pathways influencing the relationships between HIV and women’s sexual function has varied. For instance, while lower levels of sexual function among HIV+ women have been associated with greater severity of HIV-related symptoms 16 and with decreased adherence to antiretroviral therapy 17, other studies have failed to detect associations between sexual function and biomarkers of disease course, including viral load, CD4+ cell count 18, and reported use of highly potent antiretroviral regimens 11. Reports on the relationship between HIV-related body habitus changes and sexual function have also been equivocal, with one study finding no relationship 11, and another reporting lower function among those with altered body habitus 17. These differences may be due in part to variations in the methods of measurement of sexual problems across these studies.

Information about the occurrence of problems with sexual function among HIV-infected women is needed to appropriately identify, address, and ideally prevent these problems from arising, and to promote the sexual well-being of women with HIV in general. In the current investigation, we describe self-reports of sexual problems among a large, nationally representative sample of women with and without HIV in the United States. We also examine psychosocial, clinical and behavioral correlates of sexual problems among HIV-infected and uninfected women and also assess how HIV-specific parameters influence these issues among women with HIV infection.

Methods

Participants

Data on sexual functioning was collected as part of the Women’s Interagency HIV Study (WIHS). The WIHS is an ongoing prospective study of HIV infection among women in the United States. It is conducted in Bronx/Manhattan, NY; Brooklyn, NY; Washington, DC; San Francisco/Bay Area, CA; Los Angeles/Southern California/Hawaii and Chicago, IL.

Enrollment in WIHS occurred in 1994–1995 with an initial cohort of 2,054 HIV-positive women and 569 HIV-negative women matched by age, race/ethnicity, recruitment site (including HIV testing sites, community outreach sites, drug rehabilitation centers, and hospital based programs), and risk factors including history of injection drug use and number of sexual partners. In 2001–2002, an additional 737 HIV-positive and 406 HIV-negative women were enrolled; these women were similar to the original cohort in terms of proportion of HIV-positive and HIV-negative women falling into each of the matching categories. WIHS criteria for enrollment in 1994–1995 and 2001–2002 included HIV status confirmed by Western blot, an age of 13 years or older, and the ability to answer questions in English or Spanish. WIHS interviews are administered in English and in Spanish. Participants all provided written informed consent for study activities and completed a study visit every six months. These visits include completion of standardized, interviewer-administered surveys, physical and gynecological examinations, and collection of blood samples. For the current analysis, viral load and CD4+ cell count were derived from specimens; all other variables were assessed via self-report. Additional details on the WIHS recruitment methods and baseline characteristics have been previously published 19, 20.

Of the 3,766 women enrolled in WIHS, 2,899 were alive at the study visit during which the cross-sectional sexual function measures were implemented (the visit occurring between October 1, 2006 and March 30, 2007), and 2,095 (72%) women completed that study visit. Of the 2,095 women who completed the visit, 188 completed an ‘abbreviated’ study visit of select WIHS modules, which excluded the sexual function assessment. Of note, participants eligible for completing an abbreviated visit may face significant health or other contextual barriers to completing a full study visit that may be linked to impaired sexual function. An additional 102 women completed the study visit but did not complete all items on the sexual function assessment. Thus, eligible participants for the analysis include 1,805 participants (1,279 HIV-positive and 526 HIV-negative women).

Although the WIHS does not capture reasons for an abbreviated visit (e.g., incarceration, hospitalization, increased burden of other commitments), we did examine differences between several variables among those completing the abbreviated (N = 188) and the full study visit (N = 1805). The likelihood of completing an abbreviated versus full visit did not differ as a function of HIV serostatus, age, baseline history of crack or heroin use, or, among HIV-positive women, detectable viral load (all p > .05). However, employed women were more likely to complete an abbreviated visit (12%) as compared to unemployed women (6%; p < .05). Questions on sexual activity were not asked as part of the abbreviated visit. We also compared those who completed the full study visit but who did (N = 1805) versus did not (N = 102) complete all of the FSFI items (N = 102). These groups did not differ at statistically significant levels as a function of HIV serostatus, age, or whether the respondent reported having been sexually active since the last study visit. Women were more likely to have missing data on the FSFI if they were unemployed (7%) versus employed (3%; p < .01), and if they reported at baseline a history of cocaine or heroin use (7%) versus no history (4%; p < .01). Among women with HIV infection, missing data was more frequent among women with a detectable viral load (7%) versus those with an undetectable viral load (4%; p = .03).

Procedures

The Female Sexual Function Index (FSFI) is a 19-item self-report survey which includes subjective assessments along several domains of sexuality, including sexual arousal and desire, lubrication, orgasm, and pain during intercourse. For these different domains, questions assess the subjective frequency and severity of problems, as well as assessments of overall satisfaction or dissatisfaction in different domains, including orgasm, overall sexual life, and within sexual relationships 21. The scale has high internal consistency and strong associations with clinically diagnosed sexual function disorders, including sexual arousal disorder, female orgasmic disorder, hypoactive sexual desire disorder, female sexual desire disorder, and dyspareunia/vaginismus 21–23. For domains in which no sexual activity is reported, FSFI scores fall between 0 and 1 (for those with the lowest levels of sexual activity and interest over the past four weeks) to 36, with higher scores indicating better sexual functioning. Sexual activity is broadly defined in the FSFI as including caressing, foreplay, masturbation and vaginal or other sexual intercourse. Although most self-report modules in WIHS are interviewer administered, the FSFI was self-administered, with the study interviewer present to read items if necessary. This approach was chosen following review and approval of the FSFI by the WIHS Community Advisory Board and in response to a pilot of the survey and additional feedback from study personnel.

Symptoms of depression were assessed with the 20-item Center for Epidemiologic Studies Depression (CES-D) scale 24. A CES-D score greater than 23 was used to identify participants with moderate to severe symptoms of depression 25. The CES-D has high reliability in the population (Cronbach’s α = 0.93). Use of anti-depressants was also recorded for the study visit. Both sexual behaviors and alcohol and drug use are ascertained for frequency since the last study visit (i.e., over an approximately six-month window). To assess recent drug use, participants were asked to report any use of marijuana, cocaine, crack, heroin, amphetamines or methamphetamine, hallucinogens, club drugs, or non-prescribed methadone or narcotics since the last study visit. Alcohol use since the last visit was defined as a report of consumption of any wine, beer, hard liquor, or other alcohol. Participants also completed questions on number and sex of sexual partners since the last study visit. A sex partner was defined as either a man or woman with whom the participant had engaged in vaginal, oral, or anal sex. Condom use consistency with male partners was defined as either no unprotected anal or vaginal intercourse (UAVI) or one or more episodes of UAVI since the last study visit.

Several factors associated with sexual function in the general population were also assessed. Menopausal status was defined by self-reported menstrual pattern, with postmenopausal women reporting no menses for the past 12 months, and others defined as non-menopausal (including premenopausal, early perimenopausal, late perimenopausal). These definitions were developed by the World Health Organization and the STRAW working group 26, 27 and have been used in other studies of menopause in women with HIV 27–30. We were not able to classify several groups of women based on this definition, including women who were currently pregnant or lactating or who reported ever having had a hysterectomy or oophorectomy. We codified these women as “other/currently unknown” for this variable, and do not comment on their sexual functioning in this analysis, given the heterogeneity in regard to factors that may drive sexual problems. We also included self-reported use of hormone therapy (on therapy, not on therapy). Based on the history and physical examination, we also included information on Body Mass Index (BMI <25 indicating not overweight, BMI ≥25 indicating overweight/obese), reported diabetes status, and whether or not participants were taking medication for seizures, or cardiovascular disease. Women with HIV completed additional questions regarding antiretroviral therapy and adherence, and provided blood samples for measurement of HIV viral load. We defined highly active antiretroviral therapy (HAART) regimens based on DHHS/Kaiser Panel guidelines in place at the time of data collection 31.

Antiretroviral therapy was categorized among HIV+ participants as (1) not on antiretroviral therapy, (2) on antiretroviral therapy but not HAART, or (3) on a HAART regimen. Among women on a HAART or non-HAART antiretroviral regimen, we assessed medication adherence by asking how often they had taken their medication as prescribed during the 6 months before the interview. These answers were categorized into <95% versus ≥95% adherence. In a meta-analysis examining the relationship of self-reported adherence to viral response among a pooled sample of over 15,000 patients, the odds of detectable viral load were higher when self-reported adherence was set at a threshold of 95% or higher 32. CD4+ cell counts and HIV RNA measurements were measured at the study visit corresponding with the interview. Plasma HIV-1 RNA quantification was performed using the isothermal nucleic acid sequence-based amplification method in laboratories certified by the National Institutes of Health Virology Quality Assurance program, with a lower limit of detection set at 80 copies / mL. Indications of body fat loss symptoms were recorded by study personnel during the physical exam. All WIHS study procedures were reviewed by study investigators and by the WIHS Community Advisory Board, and were approved by the Institutional Review Boards at each site.

Statistical Analysis

We first conducted a principal components analysis employing varimax rotation to assess the factor structure of the FSFI (N = 1805). This analysis was performed for the overall sample as well as for HIV-infected and uninfected women separately. Univariate analysis of variance was used to examine relationships between FSFI scores and posited sociodemographic (i.e., age, racial/ethnic identification), psychosocial and behavioral (i.e., relationship status, drug and alcohol use, sexual behavior), and clinical characteristics (i.e., diabetes, BMI, depression symptoms, mental health medications, medications for seizures, blood pressure, and/or heart disease, hormone replacement therapy, menopause status). We next conducted factorial analysis of variance for all women in the sample (N = 1805), selecting variables for inclusion in the model that exhibited statistically significant univariate associations with the sexual functioning measure (p < 0.05). We assessed whether there were statistically significant interactions with HIV infection status, and each of the factors included in the model, controlling for age. We then utilized univariate analysis of variance to examine the relationships between sexual behavior (number of partnerships, consistency of condom use) and FSFI scores. Finally, univariate analysis of variance was used to assess the relationships between antiretroviral regimen features, adherence to regimens, changes in body fat, CD4+ cell count, and viral load among HIV-positive women on HAART (N = 846). Factorial analysis of variance examined the relationships between FSFI and CD4, viral load, and body fat changes, controlling for age. All analyses were conducted with SPSS version 17.0.

Results

FSFI scale characteristics

Results of the principal components analysis for the scale indicated similar factor structures for both HIV-negative and HIV-positive women, and therefore results of this analysis are reported for the combined sample. The original FSFI scale posited six factors (desire, arousal, lubrication, orgasm, dissatisfaction, and pain) that contribute to overall sexual function, and several analyses demonstrate correspondence between these factors and clinical diagnoses of sexual dysfunction such as pain, satisfaction, and sexual arousal 21–23. However, our principal components analysis did not produce these six distinct factors; instead, most of the FSFI items in our sample loaded onto a single factor. That first factor, comprising 17 of the 19 items, explained 84% of the variance in sexual functioning, and had an eigenvalue of 15.8. The second factor, comprising the 2 remaining items, accounted for 6.4% of the variance and had an eigenvalue of 1.2. We repeated this analysis separately for both HIV-positive and HIV-negative women, and found a similar patterns of results. For this reason, and because previous validation studies were conducted on samples of higher income, HIV-negative women with relatively low proportions of participants of racial/ethnic minority status, we did not calculate factor scores for each of the original FSFI domains, nor did we use the cutpoint that has been described in a previous paper 23. Instead, we computed total scores for the 19 items. These total scores ranged from 1.2 to 36.0 (Median = 14, Mean = 15, SD = 13, skewness = 0.17, Cronbach’s α = 0.99).

Characteristics of the study population

Women with HIV had lower mean FSFI scores (Mean = 13.8, SD = 12.7) than did women without HIV infection (Mean = 18.0, SD = 13.2) in this sample (p < .001). Table 1 presents posited social, demographic, and clinical predictors of FSFI scores in the study population, along with associated FSFI scores, as a function of HIV status. Among both HIV positive and negative women, higher FSFI scores, indicating fewer sexual problems, were seen in women who were younger and who were married or living with a sexual partner. Similarly, HIV positive and negative women scored lower on the FSFI, indicating greater levels of sexual problems, if they were classified as having reached menopause, diabetic, if they reported symptoms indicative of depression, and if they reported taking medication to treat mental health problems or for seizures, hypertension, or heart disease (all p < .05). For women without HIV infection, fewer sexual problems were reported among women whose BMI fell into the normal to underweight range and who were not on hormone therapy (all p < .05).

Table 1.

Univariate associations with FSFI scores among participants in the Women’s Interagency HIV Study (N = 1805), stratified by HIV status*

Behavioral, Sociodemographic Factors
HIV+ HIV−
Mean, SD N p-
value		 Mean, SD N p-
value
Age 1279 526
20–29 19.6, 12.7 71 23.1, 11.3 97
30–39 17.1, 12.5 352 22.0, 12.0 159
40–49 14.2, 12.8 545 16.0, 13.4 181
50+ 7.8, 10.3 311 9.7, 11.6 89
< .001 < .001
Race and ethnicity 1279 526
White 13.7, 12.8 168 16.4, 13.3 47
Black 13.7, 12.7 718 18.1, 13.1 317
Hispanic 13.7, 12.6 355 17.4, 13.4 139
Other 14.9, 13.1 38 23.5, 12.2 23
0.95 0.17
Married or living with a
Partner
 1278 526
Yes 18.2, 12.4 394 20.7, 12.8 169
No 11.8, 12.4 884 16.8, 13.2 357
<.001 <.001
Alcohol use since last visit 1276 523
No 13.8, 12.7 741 17.4, 13.3 297
Yes 13.6, 12.7 535 18.9, 13.0 226
0.84 0.21
Substance use since last visit 1275 523
No 13.8, 12.8 989 17.6, 13.1 403
Yes 13.4, 12.5 286 19.6, 13.3 120
0.57 0.14
Clinical Factors
HIV+ HIV−
Mean, SD N p-
value 		Mean, SD N p-
value
Diabetes 1220 506
No Diabetes 14.1, 12.7 977 18.9, 12.9 407
Diabetes 11.7, 12.3 243 14.4, 13.4 99
.009 0.002
BMI 1244 507
Underweight/Normal 13.3, 12.6 419 20.2, 12.7 124
Overweight/Obese 14.1, 12.8 825 17.4, 13.3 383
0.27 0.04
On medication for seizures, blood
pressure, heart disease		 1279 526
No 14.7, 12.8 932 19.5, 13.0 425
Yes 11.1, 12.1 347 12.0, 12.1 101
<.001 <.001
On medication for depression/mental
health		 1279 526
No 14.5, 12.8 970 18.8, 13.1 446
Yes 11.5, 12.3 309 14.0, 13.1 80
<.001 0.003
On hormone therapy 1278 526
No 13.9, 12.7 1224 18.3, 13.1 509
Yes 11.1, 12.0 54 9.7, 12.0 17
0.12 0.008
Menopause status 1153 514
Other/not classified 12.2, 12.5 331 14.5, 13.3 108
Not menopausal 16.5, 12.9 611 20.5, 12.6 351
Menopausal 7.8, 9.9 211 10.9, 12.0 55
<.001 <.001
Symptoms of depression 1276 524
Lower / CESD le 23 14.9, 12.9 1020 18.9, 13.2 424
Higher / CESD gt 23 9.2, 10.8 256 14.1, 12.5 100
<.001 0.001
Open in a new tab
*

Continued on next page.

Factors associated with sexual function

Results of a factorial analysis of variance describing relationships between FSFI and posited influential variables are reported in Table 2. For this model, we included variables which had statistically significant univariate relationships with FSFI among either HIV positive or HIV negative women, and also included HIV serostatus as a predictor. The final model excluded diabetes status, BMI category, hormone therapy, medication for seizures, hypertension, or heart disease, and medication for mental health conditions; these variables were not statistically significant after adjusting for other variables in the model, and their exclusion did not impact parameter estimates in the final model, nor the overall variance accounted for in the model. The resulting model produced an adjusted R2 = .18, with statistically significant main effects for age (F (3, 1652) = 23.8, p < .001), HIV status (F (1, 1652) = 15.3, p < .001), relationship status (F (1, 1652) = 55.3, p < .001), menopausal stage (F (2, 1652) = 12.0, p < .001), and depressive symptoms (F (1, 1652) = 36.8, p < .001). There were no statistically significant interactions detected between HIV status and any of the other variables in the model (all p >.05).

Table 2.

Predictors of Female Sexual Function Inventory (FSFI) scores among HIV-negative and HIV-positive women (N = 1661).

B Std. Error t p-value Adj 95% CI
Intercept 1.76 0.98 1.80 0.07 −0.16 3.68
HIV Status
HIV-negative 2.53 0.65 3.91 < .001 1.26 3.80
HIV-positive 1.00
Age
20–29 9.15 1.25 7.30 < .001 6.69 11.61
30–39 7.10 0.96 7.40 < .001 5.21 8.98
40–49 4.33 0.85 5.12 < .001 2.67 5.98
50+ 1.00
Married or living with a partner
Yes 4.79 0.64 7.43 < .001 3.53 6.05
No 1.00
Menopause status
Other/not classified 1.88 0.96 1.97 .05 0.01 3.76
Not menopausal 4.42 0.97 4.57 < .001 2.52 6.32
Postmenopausal 1.00
Symptoms of depression
Lower / CESD le 23 4.43 0.73 6.07 < .001 3.00 5.86
Higher / CESD gt 23 1.00
Open in a new tab

Sexual behavior and sexual function

Thirty-two percent of the study population reported having had no vaginal, oral, or anal sex with a male or female sexual partner since the last study visit (23% HIV−, 35% HIV+), 60% reported having had one partner (61% HIV−, 59% HIV+), and 8% reported having had two or more partners (15% HIV−, 6% HIV+). Among the HIV-positive women who reported one or more partners (N = 829), 774 reported at least one male partner, 40 reported at least one female partner, and 15 reported both male and female partners. Among the HIV-negative women who reported one or more partners (N = 404), 361 reported at least one male partner, 36 reported at least one female partner, and 7 reported both male and female partners. As described earlier, the FSFI defines sexual behavior broadly (e.g., including caressing, masturbation), and attributes the lowest levels of sexual function to those reporting no recent sexual activity 33. Therefore, those reporting no vaginal, oral or anal sex are likely to have lower FSFI scores, but may not have lower scores if the individual engages in behaviors that confer low HIV transmission risk and does not report problems with engaging in these activities. A general linear model including age, HIV serostatus, and number of sexual partners (0, 1, or 2 +) and a product term for the latter two variables revealed a significant interaction between HIV serostatus and number of male and female sexual partners on FSFI scores, F (2, 1796) = 6.83, p < .001. Among HIV-negative women, FSFI increased as number of partners increased from 0 (Mean = 2.7, SD = 2.7) to 1 (Mean = 22.7, SD = 11.9), but did not differ as partners increased from 1 to 2 or more (Mean = 22.7, SD = 9.6). Among HIV+ women, FSFI increased as number of partners increased from 0 (Mean = 2.9, SD = 4.6) to 1 (Mean = 19.2, SD = 11.9) to 2 or more (Mean = 23.8, SD = 10.1).

In terms of unprotected anal or vaginal sex (UAVI) with a male partner, 22% of HIV-positive women reported this behavior at least once since the last study visit, compared with 55.5% of HIV-negative women. Among the 1,233 women who reported one or more sexual partner since the last study visit, 579 (47%) reported at least one episode of unprotected anal or vaginal sex (72% for HIV-negative, 35% for HIV-positive women, p < .001). Among women with one or more partners, those who reported no unprotected sex had lower sexual function scores (Mean = 19.0, SD = 12.2) than did women who reported 1 or more unprotected episodes (Mean = 22.4, SD = 11.0, p < .001). No statistically significant interactions were detected between HIV status and UAVI on FSFI scores, after controlling for age (p = .14).

HIV-specific parameters and sexual function

Of women with HIV infection, 846 (66%) were on HAART, 41 (3%) were on an HIV antiretroviral regimen that was not classified as HAART, and 392 (31%) were not on antiretroviral therapy at the time of the study visit. FSFI scores did not differ by ARV status at the time of the visit (F (2, 1276) = 0.81, p = .44). Among women on antiretroviral therapy, 77% reported adherence levels of at least 95%; adherence was not related to FSFI scores (p = .20). Among women on HAART, 62% were on a regimen that included at least one PI, and 36% women were on a regimen that included an NNRTI. All women on HAART reported that their regimen included an NRTI. We examined whether women with a PI-containing regimen differed from those with regimens that did not include at least one PI, and found no difference in FSFI scores (p = .44). There were also no differences in FSFI for those regimens containing at least one NNRTI (p = .97).

Forty-five percent of all women with HIV-infection had a detectable viral load; this was not associated with FSFI scores (p = .29). Sixteen percent of all HIV-positive participants exhibited evidence of body habitus changes. In univariate analysis, those with body habitus changes had lower FSFI scores (Mean = 11.4, SD = 12.2) than did those with no changes (Mean = 14.2, SD = 12.8, p = .003). Fourteen percent of women were classified as having CD4 ≤199 cells/µL, 41% between 200–499, and 45% ≥500 cells/µL. CD4+ cell count was associated with FSFI scores (p = .005); post hoc comparisons revealed that those with CD4 ≤199 cells/µL had lower scores (Mean = 11.0, SD = 11.8) as compared to those whose count fell between 200–499 cells/µL (Mean = 13.9, SD = 12.7, p = .019) or greater than or equal to 500 cells/µL (Mean = 14.5, SD = 13.0, p = .004); the two higher groups did not differ from one another (p = .77). Factorial analysis of variance assessed the relationship between FSFI and body fat loss, CD4+ cell count, and age. This model revealed that CD4+ cell count remained statistically significant, F (2, 1220) = 4.1, p = .02, as did age, F (3, 1220), = 35.0, p < .001, but that body fat loss was not related to FSFI after controlling for those variables, F (1, 1220) = 0.3, p = .60).

Discussion

Our analysis reveals that the burden of sexual problems is significantly higher among women with HIV-infection compared to uninfected women. Subjective reports of sexual problems are common among women in the United States. Data from the 1992 National Health and Social Life Survey, the most recent nationally-representative study on sexuality among younger adults in the United States, suggest that among sexually active women ages 18–59, over 40% report symptoms indicative of sexual problems over a 12-month period 34. As would be expected from past research on sexual functioning in women, sexual function scores in the WIHS cohort were associated with age, menopause, symptoms of depression, and relationship status. However, the inclusion of these factors did not mitigate the influence of HIV infection on impaired sexual function scores. In addition, HIV did not influence the relationships between these factors and sexual function scores.

Although we confirmed univariate associations between several previously published correlates of sexual function among HIV-infected women, including BMI, therapeutic regimens for mental health, therapeutic regimens for seizures, blood pressure or heart disease, hormone therapy, and diabetes, these relationships were not sustained in multivariate analysis. In addition, although previous research has described the impact of chronic drug and alcohol use on impaired sexual function 35, we did not find a relationship between sexual problems and alcohol or drug use. We note, however, that our measures of alcohol and other substance use only estimate whether one or more instances of use were reported since the last study visit; these measures do not account for the role of drug and alcohol abuse and dependence on sexual problems nor the role of drug or alcohol use during sexual activity.

In addition, while CD4 cell count was associated with sexual function in this study, we did not replicate previous findings on factors associated with sexual problems in persons with HIV. We did not, for instance, detect statistically significant associations between sexual problems and being on HIV antiretroviral therapy, being on certain classes of antiretroviral therapy (e.g., a PI containing regimen), or HIV antiretroviral therapy adherence. Given that little investigation in this area has been conducted in populations consisting primarily of women, further research is needed to help clarify these inconsistent findings.

The role of sexual function and HIV/STI transmission risk behaviors has not been clearly established in the literature. In this analysis, we found a positive correlation between reporting fewer sexual problems over the past four weeks and reporting one or more episodes of unprotected anal and/or vaginal sex with male sexual partners since the last study visit. We cannot make causal assumptions given the cross-sectional nature of the analysis and the different time parameters for assessment of sexual behavior and sexual function. However, we believe that this finding warrants further examination. For instance, if additional research demonstrates a causal link between increased sexual function and transmission risk behavior, then HIV/STI prevention approaches may need to be incorporated as part of assessment and treatment of sexual problems. On the other hand, this relationship could simply reflect a measurement artifact, whereby those in sexual relationships tend to score higher on the FSFI than those with no sexual relationships. Those who were not in sexual relationships were scored as having no unprotected anal or vaginal sex, which may account for some of this finding.

Our data did not allow us to examine the relationships between FSFI and sexual risk behavior in different types of relationships, although we did report higher function among those living with a sexual partner or married. The greater level of perceived intimacy that may for instance be present in more established (versus casual) relationships may be linked to increased risk for unprotected sex, given that condom use is less frequent in relationships defined as main or primary, versus those defined by participants as being more casual in nature 36. Although papers describing patterns of partnerships and transmission risk behaviors with HIV concordant and discordant partners in the WIHS have been published previously 37, 38, the self-report instrument at the time of the FSFI administration did not include measures of condom use or partner serostatus on a partner specific level. Therefore, we were unable to disentangle these relationships.

We include a note of caution in the interpretation and analysis of scales such as the FSFI in the assessment of sexual function among women living with HIV/AIDS. The FSFI is designed to be utilized for both sexually and non-sexually active women. However, many of the items are scored such that those who report no sexual activity, defined as any sexual activity or sexual intercourse in the past 4 weeks, are assigned a numeric value of “zero,” indicating the lowest level of social, psychological and physical dimensions of sexual functioning. Although sexual activity is defined broadly in the inventory to include activities that occur with or without physical contact with a sexual partner (e.g., masturbation, foreplay), we believe that this measure may over-categorize some women as low functioning. For instance, data from the HIV Cost Services Utilization Study (HCSUS) reveal that approximately 50% of HIV-positive respondents reporting no anal, vaginal or oral sex in the prior 6 months said that their abstinence was deliberate 39. In an analysis of men and women with HIV who were homeless or had unstable housing, nearly 20% reported intentional abstinence in the past 90 days. In this analysis, common reasons cited for abstinence included unavailability of a sex partner and concern about transmitting HIV 40. Thus, it could be assumed that at least for some respondents, lack of sexual activity may reflect a temporary unavailability of sexual partners at any given time, or may be based on intentional abstinence for both HIV+ women and women who are at risk for infection. It is unknown whether sexual dysfunction may be less pronounced when intended abstinence explains a lack of sexual activity. Our study did not assess whether a lack of sexual behavior reflected a lack of available partners, a decision to abstain from sex, or some other combination of factors. However, this would be important to examine as a component of furthering this area of study.

We believe that our study contributed to the literature on sexual function in HIV by using a commonly used and widely accepted tool for the assessment of self-reported sexual problems, and by comparing women with HIV to a seronegative cohort. However, we noted some differences in the psychometric properties of the scale in this sample as compared to previous reports. For instance, we note that the FSFI factor structure previously reported by Rosen and colleagues 21 was not reproduced within our sample using the same methods of analysis. In addition, the very high internal consistency suggests that there may be some redundancy that could be addressed in future studies with HIV-infected samples. Thus, some refinements of measurement of the construct of sexual function among HIV-positive women may be useful to help further delineate the extent and nature of these issues.

Although we did not have data available on overall subjective assessments of HIV-related quality of life for the study visit in which we administered the FSFI, we did include some markers of overall health in our analysis. For instance, women with HIV who had CD4+ cell counts of less than 200, which is AIDS defining, had significantly lower FSFI scores than did those with higher cell counts. The causal nature of sexual difficulties on quality of life is unclear, given that these variables can plausibly be argued to be bi-directional in influence. However, there appears to be a strong correlation between impaired sexual function and both emotional and physical satisfaction in relationships, as well as with indices of general life satisfaction and happiness 34.

There is continued need for additional investigation into the area of sexual function among HIV-positive women. This would include an increased understanding of sexual function measures in the context of infectious diseases such as HIV/AIDS, particularly in regard to the role of intentional abstinence. While definitive conclusion awaits replication of our findings in other cohorts, our study shows a clear link between HIV infection and sexual problems among women. We propose therefore that there is a role for assessment of sexual problems in the overall care of women with HIV infection, particularly those classified as having AIDS. Additional areas of investigation would include methods for accurate and feasibly implemented assessment of sexual problems in the context of HIV care, as well as an examination of intervention approaches that could be effectively and reasonably implemented in this or similar settings.

Acknowledgements

Data in this manuscript were collected by the Women's Interagency HIV Study (WIHS) Collaborative Study Group with centers (Principal Investigators) at New York City/Bronx Consortium (Kathryn Anastos); Brooklyn, NY (Howard Minkoff); Washington DC Metropolitan Consortium (Mary Young); The Connie Wofsy Study Consortium of Northern California (Ruth Greenblatt); Los Angeles County/Southern California Consortium (Alexandra Levine); Chicago Consortium (Mardge Cohen); Data Coordinating Center (Stephen Gange). The WIHS is funded by the National Institute of Allergy and Infectious Diseases (UO1-AI-35004, UO1-AI-31834, UO1-AI-34994, UO1-AI-34989, UO1-AI-34993, and UO1-AI-42590) and by the National Institute of Child Health and Human Development (UO1-HD-32632). The study is co- funded by the National Cancer Institute, the National Institute on Drug Abuse, and the National Institute on Deafness and Other Communication Disorders. Funding is also provided by the National Center for Research Resources (UCSF-CTSI Grant Number UL1 RR024131). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health. Stacy Lindau’s effort was supported in part by the Chicago Core on Biomeasures in Population-Based Health and Aging Research at the University of Chicago and NORC Center on Demography and Economics of Aging (5P30 AG 012857). We thank members of the WIHS National Community Advisory Group for bringing the issue of sexual function into the WIHS research agenda.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Supported by National Institute of Allergy and Infectious Diseases (UO1-AI-35004, UO1-AI-31834, UO1-AI-34994, UO1-AI-34989, UO1-AI-34993, and UO1-AI-42590) and by the National Institute of Child Health and Human Development (UO1-HD-32632).

References

  • 1.Weinhardt LS, Kelly JA, Brondino MJ, et al. HIV transmission risk behavior among men and women living with HIV in 4 cities in the United States. J Acquir Immune Defic Syndr. 2004 Aug 15;36(5):1057–1066. doi: 10.1097/00126334-200408150-00009. [DOI] [PubMed] [Google Scholar]
  • 2.Aidala AA, Lee G, Garbers S, Chiasson MA. Sexual behaviors and sexual risk in a prospective cohort of HIV-positive men and women in New York City, 1994–2002: implications for prevention. AIDS Educ Prev. 2006 Feb;18(1):12–32. doi: 10.1521/aeap.2006.18.1.12. [DOI] [PubMed] [Google Scholar]
  • 3.Lang DL, Salazar LF, Wingood GM, DiClemente RJ, Mikhail I. Associations between recent gender-based violence and pregnancy, sexually transmitted infections, condom use practices, and negotiation of sexual practices among HIV-positive women. J Acquir Immune Defic Syndr. 2007 Oct 1;46(2):216–221. doi: 10.1097/QAI.0b013e31814d4dad. [DOI] [PubMed] [Google Scholar]
  • 4.Crepaz N, Marks G. Serostatus disclosure, sexual communication and safer sex in HIV-positive men. AIDS Care. 2003 Jun;15(3):379–387. doi: 10.1080/0954012031000105432. [DOI] [PubMed] [Google Scholar]
  • 5.Crepaz N, Marks G. Towards an understanding of sexual risk behavior in people living with HIV: a review of social, psychological, and medical findings. AIDS. 2002 Jan 25;16(2):135–149. doi: 10.1097/00002030-200201250-00002. [DOI] [PubMed] [Google Scholar]
  • 6.Courtenay-Quirk C, Pals SL, Colfax G, McKirnan D, Gooden L, Eroglu D. Factors associated with sexual risk behavior among persons living with HIV: gender and sexual identity group differences. AIDS Behav. 2008 Sep;12(5):685–694. doi: 10.1007/s10461-007-9259-y. [DOI] [PubMed] [Google Scholar]
  • 7.Lindau ST, Schumm LP, Laumann EO, Levinson W, O'Muircheartaigh CA, Waite LJ. A study of sexuality and health among older adults in the United States. N Engl J Med. 2007 Aug 23;357(8):762–774. doi: 10.1056/NEJMoa067423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Basson R, Schultz WW. Sexual sequelae of general medical disorders. Lancet. 2007 Feb 3;369(9559):409–424. doi: 10.1016/S0140-6736(07)60197-4. [DOI] [PubMed] [Google Scholar]
  • 9.Chander G, Himelhoch S, Moore RD. Substance abuse and psychiatric disorders in HIV-positive patients: epidemiology and impact on antiretroviral therapy. Drugs. 2006;66(6):769–789. doi: 10.2165/00003495-200666060-00004. [DOI] [PubMed] [Google Scholar]
  • 10.Klinkenberg WD, Sacks S. Mental disorders and drug abuse in persons living with HIV/AIDS. AIDS Care. 2004;16 Suppl 1:S22–S42. doi: 10.1080/09540120412331315303. [DOI] [PubMed] [Google Scholar]
  • 11.Florence E, Schrooten W, Dreezen C, et al. Prevalence and factors associated with sexual dysfunction among HIV-positive women in Europe. AIDS Care. 2004 Jul;16(5):550–557. doi: 10.1080/09540120410001716333. [DOI] [PubMed] [Google Scholar]
  • 12.Schrooten W, Colebunders R, Youle M, et al. Sexual dysfunction associated with protease inhibitor containing highly active antiretroviral treatment. Aids. 2001 May 25;15(8):1019–1023. doi: 10.1097/00002030-200105250-00010. [DOI] [PubMed] [Google Scholar]
  • 13.Goldmeier D, Kocsis A, Wasserman M. Sexual dysfunction in women with HIV. Sex Transm Infect. 2005 Aug;81(4):284. doi: 10.1136/sti.2004.009266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Lambert S, Keegan A, Petrak J. Sex and relationships for HIV positive women since HAART: a quantitative study. Sex Transm Infect. 2005 Aug;81(4):333–337. doi: 10.1136/sti.2004.013516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Siegel K, Schrimshaw EW, Lekas HM. Diminished sexual activity, interest, and feelings of attractiveness among HIV-infected women in two eras of the AIDS epidemic. Arch Sex Behav. 2006 Aug;35(4):437–449. doi: 10.1007/s10508-006-9043-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bova C, Durante A. Sexual functioning among HIV-infected women. AIDS Patient Care STDS. 2003 Feb;17(2):75–83. doi: 10.1089/108729103321150809. [DOI] [PubMed] [Google Scholar]
  • 17.Trotta MP, Ammassari A, Murri R, et al. Self-reported sexual dysfunction is frequent among HIV-infected persons and is associated with suboptimal adherence to antiretrovirals. AIDS Patient Care STDS. 2008 Apr;22(4):291–299. doi: 10.1089/apc.2007.0061. [DOI] [PubMed] [Google Scholar]
  • 18.Bouhnik AD, Preau M, Schiltz MA, Obadia Y, Spire B. Sexual Difficulties in People Living with HIV in France-Results from a Large Representative Sample of Outpatients Attending French Hospitals (ANRS-EN12-VESPA) AIDS Behav. 2008 Jan 11; doi: 10.1007/s10461-007-9355-z. [DOI] [PubMed] [Google Scholar]
  • 19.Barkan SE, Melnick SL, Preston-Martin S, et al. The Women's Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology. 1998 Mar;9(2):117–125. [PubMed] [Google Scholar]
  • 20.Bacon MC, von Wyl V, Alden C, et al. The Women's Interagency HIV Study: an observational cohort brings clinical sciences to the bench. Clin Diagn Lab Immunol. 2005 Sep;12(9):1013–1019. doi: 10.1128/CDLI.12.9.1013-1019.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Rosen R, Brown C, Heiman J, et al. The Female Sexual Function Index (FSFI): a multidimensional self-report instrument for the assessment of female sexual function. J Sex Marital Ther. 2000 Apr-Jun;26(2):191–208. doi: 10.1080/009262300278597. [DOI] [PubMed] [Google Scholar]
  • 22.Meston CM. Validation of the Female Sexual Function Index (FSFI) in women with female orgasmic disorder and in women with hypoactive sexual desire disorder. J Sex Marital Ther. 2003 Jan-Feb;29(1):39–46. doi: 10.1080/713847100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wiegel M, Meston C, Rosen R. The female sexual function index (FSFI): cross-validation and development of clinical cutoff scores. J Sex Marital Ther. 2005 Jan-Feb;31(1):1–20. doi: 10.1080/00926230590475206. [DOI] [PubMed] [Google Scholar]
  • 24.Radloff L. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  • 25.Golub ET, Latka M, Hagan H, et al. Screening for depressive symptoms among HCV-infected injection drug users: examination of the utility of the CES-D and the Beck Depression Inventory. J Urban Health. 2004 Jun;81(2):278–290. doi: 10.1093/jurban/jth114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.National Institutes of Health State-of-the-Science Conference statement: management of menopause-related symptoms. Ann Intern Med. 2005 Jun 21;142(12 Pt 1):1003–1013. [PubMed] [Google Scholar]
  • 27.Utian WH. Semantics, menopause-related terminology, and the STRAW reproductive aging staging system. Menopause. 2001 Nov-Dec;8(6):398–401. doi: 10.1097/00042192-200111000-00003. [DOI] [PubMed] [Google Scholar]
  • 28.Schoenbaum EE, Hartel D, Lo Y, et al. HIV infection, drug use, and onset of natural menopause. Clin Infect Dis. 2005 Nov 15;41(10):1517–1524. doi: 10.1086/497270. [DOI] [PubMed] [Google Scholar]
  • 29.Massad LS, Evans CT, Wilson TE, et al. Impact of menopause on condom use by HIV-seropositive and comparison seronegative women. J Acquir Immune Defic Syndr. 2008 Mar 1;47(3):401–402. doi: 10.1097/qai.0b013e31815e7466. [DOI] [PubMed] [Google Scholar]
  • 30.Fantry LE, Zhan M, Taylor GH, Sill AM, Flaws JA. Age of menopause and menopausal symptoms in HIV-infected women. AIDS Patient Care STDS. 2005 Nov;19(11):703–711. doi: 10.1089/apc.2005.19.703. [DOI] [PubMed] [Google Scholar]
  • 31.DHHS/Henry J. Kaiser Family Foundation Panel on Clinical Practices for the Treatment of HIV infection. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. [Google Scholar]
  • 32.Nieuwkerk PT, Oort FJ. Self-Reported Adherence to Antiretroviral Therapy for HIV-1 Infection and Virologic Treatment Response: A Meta-Analysis. J Acquir Immune Defic Syndr. 2005 Apr 1;38(4):445–448. doi: 10.1097/01.qai.0000147522.34369.12. [DOI] [PubMed] [Google Scholar]
  • 33.Bozzette SA, Hays RD, Berry SH, Kanouse DE, Wu AW. Derivation and properties of a brief health status assessment instrument for use in HIV disease. J Acquir Immune Defic Syndr Hum Retrovirol. 1995 Mar 1;8(3):253–265. doi: 10.1097/00042560-199503010-00006. [DOI] [PubMed] [Google Scholar]
  • 34.Laumann EO, Paik A, Rosen RC. Sexual dysfunction in the United States: prevalence and predictors. JAMA. 1999 Feb 10;281(6):537–544. doi: 10.1001/jama.281.6.537. [DOI] [PubMed] [Google Scholar]
  • 35.Peugh J, Belenko S. Alcohol, drugs and sexual function: a review. J Psychoactive Drugs. 2001 Jul-Sep;33(3):223–232. doi: 10.1080/02791072.2001.10400569. [DOI] [PubMed] [Google Scholar]
  • 36.Lansky A, Thomas JC, Earp JA. Partner-specific sexual behaviors among persons with both main and other partners. Fam Plann Perspect. 1998 Mar-Apr;30(2):93–96. [PubMed] [Google Scholar]
  • 37.Wilson TE, Feldman J, Vega MY, et al. Acquisition of new sexual partners among women with HIV infection: patterns of disclosure and sexual behavior within new partnerships. AIDS Educ Prev. 2007 Apr;19(2):151–159. doi: 10.1521/aeap.2007.19.2.151. [DOI] [PubMed] [Google Scholar]
  • 38.Wilson TE, Gore ME, Greenblatt R, et al. Changes in sexual behavior among HIV-infected women after initiation of HAART. Am J Public Health. 2004 Jul;94(7):1141–1146. doi: 10.2105/ajph.94.7.1141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Bogart LM, Collins RL, Kanouse DE, et al. Patterns and correlates of deliberate abstinence among men and women with HIV/AIDS. Am J Public Health. 2006 Jun;96(6):1078–1084. doi: 10.2105/AJPH.2005.070771. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Courtenay-Quirk C, Zhang J, Wolitski RJ. Intentional Abstinence Among Homeless and Unstably Housed Persons Living with HIV/AIDS. AIDS Behav. 2008 Sep 26; doi: 10.1007/s10461-008-9461-6. [DOI] [PubMed] [Google Scholar]

RESOURCES