Prevalence and Factors Associated with Risky Sexual Behaviors among Patients with Severe Mental Illness in Uganda: A Descriptive Cross Sectional Study

Background: Persons living with severe mental illness (SMI) which includes schizophrenia, bipolar affective disorder and recurrent major depressive disorder are predisposed to risky sexual behavior (RSB). There is a paucity of data on this problem in sub-Saharan Africa and where research has been undertaken, only a limited range of risk factors have been considered and most of it was undertaken before antiretroviral therapy was universally available, hence the need for this study. The objective of this study was to determine the prevalence and factors associated with risky sexual behavior among individuals with SMI attending care in central and south western Uganda. Methods: A cross sectional study was conducted among 393 persons with SMI attending two psychiatric out-patient facilities in Uganda. Psychiatric disorder was confirmed using MINI International Neuropsychiatric Interview version 7.2. RSB was defined as engaging in at least one of four risky sexual behaviours that have been associated with HIV infection in the Ugandan psychosocial environment in the last three months using an 8 item RSB questionnaire used for assessment of RSB. Prevalence of RSB and associated logistic Results: The of RSB The physical abuse) current psychosocial challenges (mental health stigma) and psychiatric illness factors (history of a past manic episode, current psychotic episode, severity of depressive symptoms and severity of manic symptoms). Conclusions: One quarter of clinic attending respondents with severe mental illness in Uganda practice risky sexual behaviour. Factors associated with risky sexual behaviour fall under the domains of past and present trauma, current psychosocial challenges and psychiatric illness factors. This calls for a multi-sectoral approach that includes community awareness about the nature of SMI and the rights of persons with SMI and measures to improve Psychiatric symptom management.


Background
Severe mental illness affects 3.2% of individuals worldwide [1], despite the fact that individuals living in low developed countries like Uganda being disproportionately affected by severe mental illness, there is limited data on the prevalence and correlates of severe mental illness in Uganda. Severe mental illness was defined as having any of the following disorders; which include schizophrenia, bipolar affective disorder and recurrent major depressive disorder. Patients with severe mental illness are predisposed to risky sexual behavior (RSB). Studies undertaken in middle income countries in South America and on the Indian sub-continent have reported equally high rates of RSB of between 7.0% to 70% [2] [3] [4] [5] [6]. Similarly, studies from sub-Saharan Africa have reported rates of between 7.0 to 49.1% [7]- [17]. Most of these studies from around the world have used a diversity of definitions for RSB which sometimes makes cross country comparisons difficult. Additionally, the studies that have been undertaken in sub-Saharan African have only considered a narrow band of risk factors and most were undertaken before antiretroviral therapy was universally available.

Studying RSB among individuals with SMI in HIV endemic sub-Saharan
African settings such as Uganda is important because it is the single most important risk factor for HIV [16]. In Uganda, HIV infection rates among people with SMI have been estimated to be approximately 25.5%, much higher than the rates in the general population currently estimated to be 6.2% [17] [18] [19] [20]. Persons living with SMIs have been reported to be at risk for RSB for the following reasons: increased risk of experiencing exploitative sex and interpartner violence [20]- [27]; the association of active SMI illness with impulsivity, altered judgment, increased sensitivity to personal rejection, low self-esteem, impaired reality testing [28] [29] [30] [31]; poor social support [28] [29] [30] [31]; perceived internal stigma [32] [33]; and use of substances of abuse [34] [ 35]. In this paper, we report on the prevalence of RSB among persons living

Measures
Consented study respondents were consecutively enrolled into this study and assessed using a structured, standardized, locally translated psychosocial instru- socio-economic status, and marital status); 2) Psychosocial factors (social support, mental health stigma, childhood physical abuse, childhood sexual abuse, physical abuse in adulthood and sexual abuse in adulthood); 3) Psychiatric illness factors (family history of psychiatric illness, past depressive episode, past manic episode, past psychotic episode, risky sexual behavior was the outcome variable. The tools used to assess these variables are described in Table 1. In this study the dependent variable was the proportion of respondents with "risky sexual behavior" (RSB) whose prevalence was estimated together with exact binomial 95% confidence limits. To investigate the correlates of RSB, a con-  Figure 1)). Using this model, variables were categorized as general risk factors (socio-demographic factors); psychosocial risk and protective factors (psychiatric illness factors, severity of psychiatric symptomatology, physical/sexual abuse, social support); maladaptive behaviors (alcohol/drug use, suicidal ideation/behaviour, poor adherence to medications); risk perception factors (beliefs about personal risk for HIV); and risk outcomes (which in this case is risky sexual behavior).

Statistical Analysis
Data was entered into Open Clinica and then exported to STATA® version 15 for data cleaning and analysis. Data analysis was guided by the above described  To ascertain the association between RSB and other variables binary logistic regression was performed. Variables with p value less than or equal to 0.20 at binary analysis were entered into multiple logistic regression analysis. Hosmer Lemeshow goodness of fit with backward elimination was used to test for model fitness. Variables with p value of < 0.05 at multiple regression were considered as statistically significant. The results were expressed as Odds Ratio (OR) and adjusted odds ratios (AOR).

The Prevalence of Risk Sexual Behaviour (RSB)
Overall, those who reported at least one risky sexual behaviour in the last 3 months was 24.2% (95% CI: 20.2 -28.7), with higher rates 28  Practiced at least one of the items above and never used a condom regularly (n = 95) Yes 29 (30.5%) (Table 3) On the specific risky sexual behaviours practiced over the last 3 months, the following was reported: ever had sex with anyone other than their regular partner (n = 49, 12.5%); sex in exchange for money/gifts (n = 12, 3.0%); forced sex including rape were (n = 5, 1.3%); sex with someone much older or younger (n = 17, 4.3%); ever suffered a sexually transmitted disease (n = 10, 2.5%); sex under the influence of alcohol and other substances of abuse (n = 8, 2.0%). Only about a third of individuals who practiced RSB (n = 29, 30.5%) had never used a condom in the last 3 months.
( Table 6) Individuals with severe mental illness who used alcohol were more likely to have RSB (1.31; 4.57), p = 0.005. on the association between RSB and maladaptive behaviour, perception of severe risk of getting HIV was significantly associated with RSB (AOR = 3.08, 95% CI: 1.43; 6.66) compared with low risk. Open Journal of Psychiatry

Discussion
Risky sexual behavior is the most common behavioral disorder and public health is important at the global level. The study found that a number of participants met criteria for risky sexual behavior. In this study, we assessed the prevalence of RSB and factors associated to RSB among individuals with severe mental illness receiving care at Butabika national referral mental hospital and Masaka regional referral hospital in Uganda.  [57]. The possible reason for the difference might be differences in study design, sample size, data collection tool and cultural differences in study population. In our study, RSB was more common among urban individuals, a factor which was significant in a study carried out in Ethiopia [52].

Prevalence of RSB
The odds of having risky sexual behavior was higher among clients living in urban than living in rural area. It might be due to prostitution, being homeless which is more common in urban than rural area This may be because urban dwellers usually have to buy literally everything in order to live including paying for their accommodation yet in rural settings accommodation is generally free and food is cheap, which can be obtained in one's garden [54].
In this study sample, 12.5% reported ever having had sexual exposure with anyone other than their regular partners. This is lower than a study done in Ethiopia [55]. Those who had engaged in sex in exchange for money/gifts was C. Birungi et al.
3.0%, those who had engaged in forced sex including rape was 1.3%; those who had engaged in sex with someone much older/younger (4.3%); those who had suffered a sexually transmitted disease (2.5%). About 2.0%, were reported to have had sex under the influence of alcohol and other substances of abuse, this is lower than findings from a study done in Ethiopia among patients with Bipolar affective disorder, where the prevalence was 49.1% [16]. Alcohol use prevalence was comparable to a study done in Nigeria [14], among patients with severe mental illness. Overall, 24.2%, practiced at least one of the above behaviors and 30.5%, never used a condom in the last 3 months. This is lower than studies done elsewhere [55] [56]. But these studies were done among the youth. However, our prevalence is still high may be due to low self-esteem and high internal stigmatization in individuals with severe mental illness younger adults with mental illness may result in a failure to provide healthier romantic relationship and are associated with failure to advocate for safer sex.

Factors Associated with Risky Sexual Behaviour
In this study, physical abuse predicted unprotected sex and mediated effects of emotional maltreatment on unprotected sex and on assertiveness in sexual refusal and the effects of sexual abuse on unprotected sex. These findings are comparable to studies done elsewhere [57] [58] [59] [61]. Both past and current sexual and physical abuse emerged as important factors in risky sexual behavior. The two past events appear to be an important pathway by which maltreatment confers risk for risky sexual behavior. Interventions to reduce risky sexual behavior should include assessment and treatment for trauma symptoms and for history of child maltreatment in all its forms [61]. Our findings underscore the need to better understand the mechanisms underlying the association between childhood sexual abuse and long-term outcomes. Some existing research points to biological mechanisms through which childhood abuse increases the risk of psychopathology. For example, the effect of childhood sexual abuse on obesity might be due to depression [32]

Conclusion
The prevalence of risky sexual behavior among individuals with severe mental illness was found to be high compared to the general population. individuals with severe mental illness who are female, in urban areas, who are current alcohol users, who have a history of physical and sexual abuse, those with past manic episodes, current psychotic episodes and those present with severe symptoms of depressive and manic disorders should be assessed for risky sexual behaviour. Interventions should contain widespread sexual and reproductive health awareness on issues such as sexual education safe sex and sexually transmitted infections for individuals with severe mental illness in various health care facilities.

Limitations of the Study
In general, the present study reported the burden of RSBs and associating factors among individuals with severe mental illness. The sensitive nature of sexual behaviors on face-to-face interview could have a social desirability bias. Moreover, this study is institution-based study. Therefore, the findings cannot be generalized to those who remain undiagnosed or untreated in the community.

Declarations Acknowledgements
All individuals who participated in this study, my supervisors, research assistants and the members of Butabika National mental hospital and Masaka regional referral hospital.

Availability of Data and Materials
The data that support the findings of this study are available from Medical Research Council/Uganda Virus Research Institute & London School of Hygiene and Tropical Medicine, Uganda research unit (MRC/UVRI & LSHTM) but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of MRC/ UVRI & LSHTM.

Authors' Contributions
BC designed the study, analyzed and interpreted the data. WS guided the data analysis, NK reviewed data analysis results, NN reviewed the literature review and implementation of study. EK reviewed the study, analysis and interpretation of the data. EK contributed in reviewing the manuscript. EK sought and obtained funding. All authors read and approved the final manuscript.

Ethics Approval and Consent to Participate
I confirm that the proposal, research tools were approved before the study commenced and to conduct the study was obtained from the Higher Degrees Research Ethics Committee (HDREC), the Uganda National Council of Science and Technology (HS 2337), the Uganda Virus Research Institute's Research and Ethics Committee (GC/127/19/10/612) and Permission was obtained from Directors of Butabika National Referral Mental and Masaka Regional Referral Hospital. Written informed consent was obtained from all participants. I confirm that all methods were carried out in accordance with relevant guidelines and regulations (HDREC, UVRI and UNCST). Confidentiality was ensured by using de-identified codes.

Funding
The Medical Research Council/Uganda Virus Research Institute/London School Open Journal of Psychiatry of Hygiene and Tropical Medicine, Uganda for having funded the data collection and analysis of the study.
Makerere University, College of Health Sciences, Department of Psychiatry for having provided part of the tuition fees needed for the PhD.

Conflicts of Interest
The authors declare that they have no competing interests.