King, Cynthia MD, MPH*; Feldman, Jacqueline MD†; Waithaka, Yvonne MPH*; Aban, Inmaaculada PhD§; Hu, Jianfang MS§; Zhang, Sijian MS§; Hook, Edward III MD*‡∥; Bachmann, Laura H. MD, MPH*‡¶
APPROXIMATELY 20% OF THE ADULT American population suffers from some mental illness each year and nearly 3% of all American adults annually experience severe and persistent mental illness (SPMI) (e.g., schizophrenia, bipolar disorder, panic disorder, severe depression, or obsessive-compulsive disorder).1,2
Several studies have shown that behaviors that place them at risk for sexually transmitted infection (STI) are relatively common among persons with SPMI.3 For instance, although the prevalence of human immunodeficiency virus (HIV) in the general population is 0.3%, approximately 8% of those with SPMI are infected.4 Factors contributing to increased vulnerability for STIs, including HIV, among those with mental illness include impaired autonomy, increased impulsivity, and increased susceptibility to coerced sex.5 At the same time, much of the currently available data on the convergence of mental illness and STI risk are based on self-report or on studies evaluating the prevalence for mental illness amongst those with the STIs of interest.
Most care for persons with mental illness is provided in outpatient settings which focus upon their clients’ mental health and, in general, tend to provide only limited general medical, or specifically, reproductive health care.6 Although there is awareness that mental illness is relatively common among persons with STIs, there are few data which describe the prevalence of high risk behaviors or STIs among persons receiving care for mental illness. To address these deficits, the objectives of this study were two-fold: 1) to quantitate STI/HIV risk among patients with SPMI receiving care at a university-based clinic, and 2) to determine the prevalence of Neisseria gonorrhoeae, Chlamydia trachomatis, and Trichomonas vaginalis in this group.
Between July 2003 and August 2004 all patients between 19 and 60 years of age who were attending an outpatient university-based psychiatric clinic for new or established care were approached for study participation. All patients were considered to be eligible for participation unless the following exclusion criteria were met: 1) the patient declined participation; 2) inability to give informed consent as judged by clinic staff (a declaration of legal incompetence, and/or IQ ≤45); 3) receipt of antimicrobials active against N. gonorrhoeae, C. trachomatis, and/or T. vaginalis within the prior 30 days; or 4) age greater than 60.
After receipt of written informed consent, participants underwent an interviewer-administered survey to elicit data on: history of sexual activity, sexual partner number, use and consistency of barrier protection, exchange of sex for drugs or money, history of STI and substance use, and psychiatric diagnosis. Interviewers also performed chart abstraction to document the patient’s history of incarceration, emergency department visits, psychiatric hospitalization, clinic attendance, clinician’s thoughts on patient progress or regression, and the patient’s official psychiatric diagnoses (up to 4 diagnoses were given per patient). Psychiatric diagnoses were made by board-certified psychiatrists using DSM-IV criteria.7
Upon completing the interview, male patients were asked to submit 30 mL of first-void urine to test for N. gonorrhoeae and C. trachomatis using Gen-Probe APTIMA Combo 2® Assay transcription mediated amplification. Female patients were asked to submit 2 self-obtained vaginal swabs, 1 for APTIMA Combo 2 testing for N. gonorrhoeae and C. trachomatis, and 1 for inoculation into an InPouch™ T. vaginalis culture system (BioMed Diagnostics, Inc., Santa Clara, CA). Males were not tested for T. vaginalis. Test results were made available to participants within 6 days of sample submission. Participants were asked to call a dedicated phone line for results. Those who did not call but had a positive result were contacted by project personnel. Those with positive test results were treated at no cost to the participant in accordance with the 2002 CDC STD Treatment Guidelines.8 As mandated by public health law, patients diagnosed with gonorrhea or chlamydial infection were reported to the Alabama Department of Public Health. All study procedures were approved by the University of Alabama at Birmingham Institutional Review Board for Human Use.
Urine and vaginal swabs were collected for testing for C. trachomatis and N. gonorrhoeae and processed within 48 hours of receipt in the laboratory. Gen-Probe APTIMA Combo 2® transcription mediated amplification assays were performed according to the manufacturer’s instructions (Gen-Probe Incorporated, San Diego, CA). All positive tests were repeated for confirmation. For women, a vaginal swab specimen was inoculated into an InPouch™ T. vaginalis culture system (BioMed Diagnostics, Inc., Santa Clara, CA) and incubated at 37°C for up to 5 days. The presence of motile trichomonads was considered positive.9
Data Management and Statistical Analysis
Questionnaire forms were directly scanned into a study database using the TELEform® Version 6 program (Cardiff Software, Inc., San Marcos, CA). Data were converted to SAS Version 8.2 (the SAS Institute, Cary, NC) format for statistical analysis. Continuous variables are summarized as means ± standard deviations or medians and ranges and tested using t test or Wilcoxon test (nonparametric analog of the t test). Categorical variables are presented as proportions and compared using Fisher exact test. Psychiatric diagnoses were categorized by shared global diagnostic criteria. A patient was considered as having a specific psychiatric diagnosis if the diagnosis was listed as any 1 of the 4 diagnoses that a patient may be given. The STI composite variable was categorized by including any participant who had a positive test for gonorrhea, chlamydia, and/or trichomoniasis. Associations between specific variables and the presence of gonoccocal, chlamydial, or trichomonal infection were determined using χ2 or Fisher exact tests as appropriate. Binomial test was used to compare the sample proportion observed to the population prevalence. Logistic regression was conducted to assess the risk factors simultaneously and backward selection was used to select the final model. For all analyses a P value of <0.05 was deemed statistically significant. No adjustments were done on the P values presented here to account for multiple testing since this article is exploratory in nature.
From July 2003 to August 2004, 629 patients were approached for study participation and 400 (63.6%) were enrolled (189 females and 211 males). Those who refused enrollment were not significantly different from those who participated in the study in terms of gender (P = 0.30), age (P = 0.07) or race (P = 0.19). Interestingly, among the 229 refusals, 106 (46.3%) cited recent STI screening as their primary reason for refusal. Table 1 demonstrates the general characteristics of the study population. The median age was 40 years (range, 19–57) and the majority of patients were black (65%) and male (53%). Most patients were single and had never been married (56.8%) and 69.3% had 12 or fewer years of education (mode, 12 years). The most common psychiatric diagnoses were psychotic disorders (61.9%) that include schizophrenia, schizoaffective disorder, other psychotic disorders, and delusional disorders. Affective disorders were the next most common diagnoses (39.9%) and included depression, dysthymia, bipolar disorder, conduct disorder, and personality disorder. Thirty-four percent of the study population had more than 1 diagnosis, of which 61% were substance use disorders (data not shown).
When queried about nonpsychiatric medical care 60% of participants (N = 240) reported receipt of care with women being more likely than men to report nonpsychiatric medical care. This was most often care from a private physician though 3.3% reported routinely receiving care in the Emergency Department. The majority reported some type of health coverage, most often Medicaid.
Table 2 presents the sexual behaviors reported by study participants. Ninety-five percent of participants stated that they had engaged in sexual activity within their lifetime. Although 95.2% of women and 90.3% of men reported heterosexual activity, same sex or bisexual activity was not uncommon (9.6% of males and 4.8% of females). Forty-three percent of participants reported sex in the past 30 days (36.7% and 50.8% of men and women, respectively). Nearly 21% of men and 11% of women had sex with at least one new partner during this time frame (range 1–10 new partners) with 8% of individuals reporting that one or more of their partners over the previous 30 days were anonymous. The median partner number for men within the previous 6 months was 1 (range 0–70) and for women it was 1 (range 0–40) (data not shown). Overall, 6.7% of males and 11.2% of females reported exchanging sex for drugs at some point in their lives. Approximately 40% of participants reported a prior history of an STI with the most commonly cited STIs being gonorrhea (16%) for men and trichomoniasis (11%) for women (data not shown). When asked about their last 5 sexual encounters, over half (52.8%) of participants reported no condom use.
Table 2 also demonstrates substance use history and current patterns of use in the population. Over half of those surveyed reported alcohol use in the past 6 months. Almost a quarter (24%) consumed 5 or more drinks in 1 sitting at least some of the time if not more often when they did drink alcoholic beverages in the prior 30 days. A substantial proportion of the population (65%) had used some type of drug in the previous 6 months. Twenty-four percent had used a sedative, 18% had used marijuana, and 15% had used some form of cocaine and/or heroin.
The prevalence of N. gonorrhoeae and C. trachomatis were relatively low (N. gonorrhoeae −1% and C. trachomatis −3.3%). In contrast, 15.7% of women were found to have trichomoniasis. Three participants were infected with more than 1 organism, one of whom was infected with Gonorrhea, Chlamydia, and Trichomonas.
Reflecting the high prevalence of trichomoniasis in the female participants there was a statistically significant relationship between gender and the presence of an STI (P <0.0001) (Table 3). There was also a statistically significant relationship between an STI diagnosis and race (P = 0.0013). These associations were no longer present if trichomoniasis was excluded from the STI composite variable (P = 0.122 for gender and P = 0.5895 for race). Psychiatric diagnosis did not predict the presence of an STI (P = 0.1857) (data not shown). A statistically significant relationship between smoking/snorting cocaine and/or heroin (P = 0.03) in the prior 6 months and having an STI was noted. Exchanging sex for drugs was also found to have a significant association with having an STI (P = 0.0006). Participants who were sexually active in the past 6 months were more likely to have an STI (P = 0.0006) particularly if they had more than 1 sexual partner during this time (P = 0.0220) (data not shown). Finally, although there was no significant relationship between psychiatric diagnosis and the presence of an STI, there were significant associations between certain psychiatric disorders and risk markers. In this case, each disorder was paired separately with a risk marker to investigate the association. For instance, patients with psychotic disorders were less likely to report a prior history of STI or to have used drugs in the previous 6 months (P = 0.02 and P = 0.015, respectively). On the other hand, patients with affective disorder were more likely to have had STI and to have used drugs in the previous 6 months (P = 0.047 and P = 0.016, respectively). Patients with a substance abuse diagnosis were more likely to report exchanging sex for drugs or money (P = 0.0073) and reported higher numbers of sexual partners (P = 0.0045) and less condom use (data not shown) than patients not diagnosed with substance abuse. Patients with cognitive disorder reported fewer sexual partners than those without cognitive disorder (P = 0.022). However, as stated above, none of these differences in risk behavior translated into differences in STI prevalence amongst the different diagnostic groups.
Multiple logistic regression analysis was performed using any STI as the outcome. Gender, race, and exchanging sex for drugs were found to be significant predictors of the presence of an STI with P values all <0.05 (Table 4). The odds of having an STI for a female was approximately 5 times the odds for men (95% CI: 2.29–10.90); whereas the odds of having an STI for blacks was about 4.5 times the odds for white participants (95% CI: 1.5–13.3). Participants who exchanged sex for drugs had odds of having an STI that are 3.3 times higher than those who did not (95% CI: 1.38–7.81). Because men were not screened for trichomoniasis, the most common STI in the study, the decision was made to perform analyses separately for men and women. However, multiple logistic regression was unable to be performed separately for men because of the small number of men with an STI. When male participants were excluded, exchange of sex for drugs was the only significant predictor of an STI in women.
Self-reported sexual risk behaviors were heterogeneous but relatively common in this population of patients receiving chronic care for mental illness. For instance, although almost 6% of the population stated that they had never had sex in their lifetime and over half (56.6%) stated that they had not been active over the previous 30 days, of the 43.3% who were active, 8% reported 1 or more anonymous partners during this time frame. Furthermore, nearly 1 in 10 patients reported exchanging sex for drugs or money (with no significant difference by gender). This contrasts with population-based data from the 2002 National Survey of Family Growth, in which 2.3% of people aged 15 to 44 years of age reported exchanging sex for drugs or money.10 Despite the high prevalence of risky exposure, protective measures were rather uncommon—over half the population did not use a condom during any of their last 5 sex acts. Interestingly, STI prevalence was low to moderate, with the exception of trichomoniasis, which was present in 15.7% of the female participants. Therefore, T. vaginalis prevalence in our study participants was significantly higher (P <0.0001) than that seen in a recent population-based study, which found a prevalence of 3.1% among women aged 14 to 49 years (ranging from 1.3% in non-Hispanic whites to 13.3% in non-Hispanic blacks).11
Substance use, an important cofactor for STI acquisition, was also common in these patients.12,13 This is not surprising given that substance use (including alcohol use) has been described as a method of “self-medication” in this population.14,15 Among those reporting alcohol consumption within the previous 6 months, almost a quarter reported having 5 or more alcoholic beverages during 1 drinking session. Binge drinking,16 a behavior that may cloud the judgment of the drinker, may make the individual more likely to engage in risky sexual activity. The same could be said for illicit drug use in the population. In the past 6 months, sedatives were most often used on a regular basis but marijuana and smoked or snorted cocaine or heroin was used by 15% to 20%. Although a statistically significant relationship between smoking/snorting cocaine and/or heroin (P = 0.03) in the prior 6 months and having an STI was noted in the univariate analysis, this relationship was not significant in the multivariate analysis.
Risky sexual behavior in the mentally ill population has been well documented by other investigators. Hutton et al.17 determined that depressed patients were more likely to have exchanged sex for money, to have had sex with an intravenously drug user, to have had sex when “high” on alcohol and/or drugs, to have a greater number of lifetime sexual partners and to abuse alcohol and/or drugs more often than nondepressed patients. Coverdale and Bayer’s18 focus on men with active mental disorders found that they were significantly more likely than controls to have known their partners for less than a day and to have been pressured into unwanted sexual intercourse. Because many persons, mainly women, with SPMI cannot meet basic needs of food, shelter, money, etc., they may use sex in exchange for these needs.19 Surprisingly, in this study, although diagnostic category was associated with some risk markers, the individual’s psychiatric diagnosis(es) did not seem to predict the presence of an STI.20 One would hypothesize that certain disorders (for instance bipolar disorder is associated with hypersexuality in some patients) may be more likely than others to result in practice of risky behaviors with increased STI acquisition.21 However, patients had up to 4 psychiatric diagnoses each and there were 25 distinct psychiatric diagnoses in this population. Although we aggregated psychiatric diagnoses by global diagnostic criteria, our sample size was likely too small to tease out differences between diagnoses and groups of diagnoses as they relate to STI risk. Limitations of this study include the relatively small sample size and the fact that men were not tested for trichomoniasis. The fact that almost half (46.3%) of those who refused to participate in the study cited having received recent STI testing as the reason for nonparticipation speaks to the possibility that patients at higher risk may have been excluded (though fortunately they were receiving STI screening). Finally, it is likely that individuals with SPMI who are not in care are practicing more risky behaviors than those who seek routine mental health care and the current study is unable to address this.
The risky behaviors found in this population of chronically mentally ill patients bolsters the need for regularly obtaining a detailed sexual history from patients. Although taking a sexual history is deemed important by many health care providers,22,23 often the questions are not asked because there is a perceived belief that the patients do not want to discuss sexual health matters. However, studies have shown that psychiatric outpatients are comfortable discussing sex and expect to be asked some questions regarding their sexual practices.24 Although asking patients their sexual and STI history is appropriate, needed, and expected by patients, questionnaires and self-report have been the most commonly used methods to gather this information. Obtaining information this way can often be difficult because: 1) most STIs are asymptomatic and detection would be highly dependent on screening; 2) availability of routine STI screening is highly variable and dependent upon the resources and health care access of the population; and 3) patients who have been diagnosed with an STI must be willing to divulge their STI history, a process that is related to the comfort level of the patient.
In our experience, the utilization of someone comfortable and experienced in taking a sexual and STI history could make the process more successful as this could ensure that all questions are asked and it would allow for the patient to ask or clarify questions if needed. Once a history is obtained, based on the high-risk behaviors found in our population, routine STI and HIV screening would be recommended for those patients who have been sexually active in the past 6 months, sexually active with more than 1 partner, admit to exchanging sex for drugs, or actively using drugs (particularly cocaine and/or heroin). Taking a sexual and STI history along with screening for STI/HIV will allow for the detection and treatment of STI/HIV as well as provide an opportunity for preventive education regarding STI/HIV which seems to be an ongoing need in this vulnerable population.
2. Kessler RC, Berglund PA, Zhao S, et al. The 12-month prevalence and correlates of serious mental illness (SMI). In: Manderscheid RW, Sonnenschein MA, eds. Mental Health, United States. Washington, DC: U. S. Government Printing Office, 1996:59–70.
3. Carey MP, Carey KB, Kalichman SC. Risk for human immunodeficiency virus (HIV) infection among persons with severe mental illnesses. Clin Psychol Rev 1997; 17:271–291.
4. Cournos F, McKinnon K. HIV seroprevalence among people with severe mental illness in the United States: A critical review. Clin Psychol Rev 1997; 17:259–269.
5. Coverdale JH, Bayer TL, McCullough LB, et al. Sexually transmitted disease prevention services for female chronically mentally ill patients. Community Ment Health J 1995; 31:303–315.
6. Roca RP, Breakey WR, Fischer PJ. Medical care of chronic psychiatric outpatients. Hosp Community Psychiatry 1987; 38:741–745.
7. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association, 1994.
8. Centers for Diseases Control and Prevention. 2002 Guidelines for the treatment of sexually transmitted diseases. MMWR Recomm Rep 2002; 51:1–78.
9. Borchardt KA, Hernandez V, Miller S, et al. A clinical evaluation of trichomoniasis in San Jose, Costa Rica using the InPouch™ TV test. Genitourin Med 1992; 68:328–330.
10. Anderson JE, Mosher WD, Chandra A. Measuring HIV risk in the U.S. population, ages 15–44: Results of cycle 6 (2002) of the National Survey of Family Growth. Advance data from vital and health statistics. Hyattsville, MD: National Center for Health Statistics, 2006.
11. Sutton M, Sternberg M, Koumans EH, et al. The prevalence of Trichomonas vaginalis infection among reproductive-age women in the United States, 2001–2004. Clin Infect Dis 2007; 45:1319–1326.
12. Bachmann LH, Lewis I, Allen R, et al. Risk and prevalence of treatable sexually transmitted diseases at a Birmingham substance abuse treatment facility. Am J Public Health 2000; 90:1615–1618.
13. Siegal HA, Falck RS, Wang J, et al. History of sexually transmitted diseases infection, drug-sex behaviors, and the use of condoms among Midwestern users of injection drugs and crack cocaine. Sex Transm Dis 1996; 23:277–282.
14. Bolton J, Cox B, Clara I, et al. Use of alcohol and drugs to self-medicate anxiety disorders in a nationally representative sample. J Nerv Ment Dis 2006; 194:818–825.
15. Buckley PF. Prevalence and consequences of the dual diagnosis of substance abuse and severe mental illness. J Clin Psychiatry 2006; 67(suppl7):5–9.
16. Naimi TS, Brewer RD, Mokdad A, et al. Binge drinking among US adults. JAMA 2003; 289:70–75.
17. Hutton HE, Lyketsos CG, Zenilman JM, et al. Depression and HIV risk behaviors among patients in a sexually transmitted disease clinic. Am J Psychiatry 2004; 161:912–914.
18. Coverdale JH, Turbott SH. Risk behaviors for sexually transmitted infections among men with mental disorders. Psychiatr Serv 2000; 51:234–238.
19. Weinhardt LS, Bickham NL, Carey MP. Sexual coercion among women living with a severe and persistent mental illness. Aggress Violent Behav 1999; 4:307–317.
20. Erbelding EJ, Hutton HE, Zenilman JM, et al. The prevalence of psychiatric disorders in sexually transmitted disease clinic patients and their association with sexually transmitted disease risk. Sex Transm Dis 2004; 31:8–12.
21. Keck PE Jr, McElroy SL, Arnold LM. Bipolar disorder. Med Clin North Am 2001; 85:645–661, ix.
22. Jennings PR, Bachmann L. Yes, the sexual history is important. JAAPA 2005; 18:70–72.
23. Nussbaum MP, Shenker IR, Feldman JG. Attitudes versus performance in providing gynecologic care to adolescents by pediatricians. J Adolesc Health Care 1989; 10:203–208.
24. Crawford MJ, Shaw T. Psychiatric out-patients’ views on talking about sex. Psychiatr Bull 1998; 22:365–367.