Prior to the EVD epidemic, the Demographic Health Survey conducted in 2013 in Sierra Leone showed that condom use was low, with 5% of female and 13% of male respondents reporting having used a condom in the past 12 months. Twenty-five percent of men and 6% of women reported having two or more sex partners in the last 12 months, and 23% of men and 5% of women reported concurrent sexual partnerships. Multiple and concurrent sexual partnerships were highest among older married men with low education [9]. Human Immunodeficiency Virus (HIV) prevalence was 1.5% among adults aged 15–49 years [9]. There is limited information regarding the impact of the EVD epidemic on sexual risk behavior, but adolescent girls in Sierra Leone reported more unplanned pregnancies and engagement in transactional sex in the nine months during the epidemic when public schools were closed [10]. In addition, EVD survivors reported stigma and feelings of bereavement similar to those experienced by people living with HIV, possibly negatively affecting both their quality of life and intimate relationships, and their motivation to seek healthcare and other services [11–13]. The impact of the EVD epidemic on the sexual behavior of EVD survivors has not yet been fully explored.

In May 2015, the Sierra Leone Ebola Virus Persistence Study was launched to investigate EBOV persistence in the body fluids of EVD survivors in Sierra Leone [5]. The study consisted of two phases. The first phase assessed EBOV persistence in semen of 100 adult male EVD survivors, and the second phase assessed EBOV persistence in semen and additional body fluids (vaginal fluid, menstrual blood, urine, rectal fluid, sweat, saliva, tears, and breast milk as applicable by sex) in 120 male and 120 female EVD survivors. Male and female EVD survivors living with HIV were also invited to participate in the study to characterize EBOV persistence among EVD survivors living with HIV. The study took place in two sites: Military Hospital 34 (an urban facility in Freetown, Western District) and Lungi Government Hospital (a semi-rural facility in Lungi, Port Loko District). In this paper, we discuss the development and implementation of the Ebola Virus Persistence Risk Reduction Behavioral Counseling Protocol (henceforth referred to as the behavioral counseling protocol) used in the study. The test results from this study will be published separately and the overall study design has been described elsewhere [5].

Although the authors note the importance of acquiring information on the persistence of EBOV in various body fluids within the pediatric population, for ethical reasons, this study was limited to adults aged 18 years or older. All participants provided written informed consent at the first study visit. The study was approved by the Sierra Leone Ethics and Scientific Review Committee and the WHO Ethical Review Committee (No. RPC736).

We sought to reduce sexual exposure to EBOV by behavior change strategies such as abstinence, increased condom use, and choice of partners. To do this, we developed a behavioral counseling protocol adapted from the Project RESPECT Brief Counseling intervention, an individual face-to-face counseling model that has been proven effective at reducing both new sexually transmitted infections (STIs) and risky behaviors in randomized controlled trials [14]. RESPECT’s Brief Counseling was implemented in the context of Human Immunodeficiency Virus (HIV) testing, and included two 20-minute counseling sessions (i.e. pretest/posttest) in which the counselor supported risk reduction behaviors by increasing the client’s perception of personal risks, emphasizing self-efficacy and personalized goal setting through identifying concrete, incremental and achievable risk-reduction steps that limited sexual HIV/STI exposure. STI clinic patients were asked to describe their own sexual risk behaviors, and misconceptions about risk were clarified. Counselors supported clients in identifying personal barriers to risk reduction and possible ways to overcome them, and helped clients to identify and negotiate behavioral risk reduction steps relevant to their personal risk behaviors. RESPECT’s Brief Intervention model has been previously adapted in behavior change interventions aimed at reducing STIs and pregnancy in vulnerable populations residing in high HIV prevalence settings [15, 16, 17, 18]. The intervention appears to be particularly effective among those with limited exposure to HIV/STI prevention guidance [15, 16, 17, 18]. In sum, the rationale for using a RESPECT model for reducing sexual EBOV exposure was that the approach (1) has demonstrated efficacy at reducing behaviors relevant to sexual EBOV transmission (e.g., increasing abstinence, increasing condom use, reducing risky sexual partnerships), (2) uses a flexible approach that meets the client at his or her own understanding, (3) has been shown to be effective using counselors that are trained in adherence to the model but do not have advanced degrees in counseling, and (4) has been successfully adapted in many international settings with varying cultural contexts.

The adaptation of Project RESPECT for male and female EVD survivors enrolled in the Sierra Leone Ebola Virus Persistence Study required several changes to HIV/STI prevention behavioral guidance. These changes were made in consultation with HIV service providers operating in Sierra Leone [e.g. Sierra Leone National AIDS Control Programme, National AIDS Secretariat, Dignity Association, Joint United Nations Programme on HIV/AIDS (UNAIDS)].

The standard operating procedure for the behavioral counseling protocol included EBOV persistence pre-and post-test counseling as well as HIV pre-and post-test counseling. EBOV persistence pre-test counseling is an introduction to EBOV testing and risk reduction advice prior to testing, while post-test counseling is a delivery of tailored guidance based on qRT-PCR test results. Fig 1 shows the typical participant visit flow.

Table 1 shows the behavioral risk reduction guidance corresponding to each body fluid tested. In the event of a positive qRT-PCR test result, relevant guidance was delivered to the participant in order to reduce transmission risk. All standard operating procedures associated with the delivery of guidance at pre-test and post-test counseling can be found in the full Ebola virus persistence study behavioral counseling protocol here: S1 Protocol.

As prior evidence showed that EBOV could persist in semen [2], male EVD survivors were encouraged at pre-test counseling to use condoms or engage in abstinence for the prevention of EBOV transmission as well as prevention of HIV/STI transmission and unwanted pregnancy. Given limited or no prior evidence for EBOV persistence in vaginal fluids, menstrual blood, urine, rectal fluids, sweat, saliva, tears, and breast milk, precautionary guidance was not provided at pre-test counseling for any of these body fluids. Women were advised at pre-test counseling to use condoms or abstain from sexual intercourse to prevent HIV/STI acquisition and unwanted pregnancy [2]. Female participants were given sexual risk reduction guidance to prevent EBOV transmission only in the event of a positive qRT-PCR test result.

Sexual risk reduction behavioral guidance was tailored for each type of body fluid. Participants were given different guidance for fluids that posed a greater risk for sexual transmission of EBOV due to contact during sexual activity (semen, vaginal fluids, menstrual blood, rectal fluid, urine, sweat, saliva) as compared to fluids for which there was likely to be less contact during sexual activity (breast milk, tears) [2]. Sexual risk reduction guidance was tailored to participants’ religious and cultural practices, particularly around genital washing. For example, counselors reported that Muslim participants preferred to wash the genital area or entire body after sexual intercourse, while Christian participants preferred to wipe genital areas with tissues after sexual intercourse.

For non-sexual risk reduction behavioral guidance, the study team reviewed general infection prevention and control (IPC) guidelines for other pathogens that are spread through contact with infectious body fluids such as viral meningitis and hepatitis B (http://www.cdc.gov/meningitis/viral.html). This guidance was modified for a low resource setting, taking into consideration limited access to clean water and sustained power sources, and lack of flushing toilets, bleach, or other materials and practices used in standard IPC protocols.

Special consideration was given to shaping guidance for female EVD survivors. A focus group with female EVD survivors was held prior to the enrollment of women in the second phase of the study. Participants discussed experience with stigma due to their status as an EVD survivor, fear of resuming sexual activity, and concerns that their partner would not accept condom use. Information from this focus group was shared with study counselors as preparation for discussions with participants regarding specimen collection, sexual activity, and condom use among female participants. Given high rates of sexual violence among women in Sierra Leone [9], a referral pathway for intimate partner violence (IPV) was established in accordance with IPV service provision guidelines from the government of Sierra Leone. Additionally, if a breast milk specimen of a breastfeeding female survivor were to test positive for EBOV, she would be immediately asked to switch to replacement feeding using ready-to-use infant formula provided free of charge with support from UNICEF, and receive ongoing counseling and support from Sierra Leone MoHS Food and Nutrition Directorate staff or a trained nurse.

Given the complex nature of the behavioral guidance for all body fluids, particularly for semen and other intimate fluids, counselors with a background in HIV/STI and/or mental health nursing with previous experience with the EVD epidemic were recruited for the study. At Military 34 Hospital, one female HIV/STI nurse and one male mental health nurse were chosen. At Lungi Government Hospital, one woman with a background in HIV/STI nursing and one woman with a background in midwifery were chosen.

For the first phase of the study, we developed a four-day training package that focused on the science of EBOV transmission and diagnostic testing methods, how to effectively discuss sexual behavior and risk reduction, and common mental health/psychosocial issues reported by EVD survivors. We held an additional four day-training focused on HIV testing and counseling. For the second phase of the study, we added a four-day training to address behavioral guidance for additional body fluids, and to assess for IPV among female study participants in accordance with national and international guidelines [20]. These trainings were primarily geared towards study counselors but all study staff were invited to attend trainings relevant to their roles.

The counselors used role-playing to practice using counseling scripts that had been translated into Krio language and then back-translated for the first phase of the study. The Sierra Leone National AIDS Control Programme, Dignity Association, and UNAIDS provided training on HIV testing and counseling, condom demonstrations, and stigma and discrimination. During and after the training sessions, suggestions from the counselors were incorporated in the counseling scripts and behavioral guidance.

We observed great receptivity to condom demonstrations, with many participants informing research staff that they had not participated in a condom demonstration prior to joining the study. Given the low national rates of condom use among men and women in Sierra Leone [9], exposure to condoms via the behavioral counseling protocol may eventually be associated with gains in reproductive health and STI prevention in participants.

Communicating the risk of sexual transmission of EBOV to study participants was challenging as there was limited scientific evidence regarding the length of persistence and the infectiousness of positive qRT-PCR body fluids. In this study, we observed that communicating uncertainty in a transparent way appeared to facilitate trust between participants and the research study. For example, following the report of an EVD survivor in the United Kingdom who relapsed with EVD meningitis [22], the behavioral counseling protocol was adapted to ensure that participants were aware of possible severe relapses. Future development of similar counseling protocols may consider embracing scientific transparency as a trust-building communication tool.

In the course of the study, we identified a need to clearly and simply explain the detection method of the qRT-PCR assay to participants. We developed a “mango tree” analogy comparing detection of EBOV-specific RNA primers (short target sequences) in the body fluids of Ebola survivors via qRT-PCR testing to trying to detect a whole mango tree (the intact, viable Ebola virus) by being able to detect only the mango fruits or the leaves (the target RNA primers for EBOV). Detecting only fruits or leaves, one cannot determine whether the tree is indeed intact and alive just as qRT-PCR testing can only determine whether the target RNA primers are detected or undetected in a body fluid specimen, and not whether the virus is infectious. Detecting only one of the two target RNA sequences (an indeterminate test result), could also be explained using this analogy, and also to explain possible variation in test results using different qRT-PCR assays with differing target sequences. This addition to the counseling script was very helpful for study participants to conceptualize the test, the different results they encountered, and the associated counseling messages.

A flexible protocol was also instrumental in optimizing participant flow during the study. For example, some participants preferred to visit the counselor prior to specimen collection for more detailed instructions on specimen collection processes or for extra encouragement to continue with the testing process. Talking points were developed as reference materials for study counselors as the original scripts were lengthy. Because we believed that participants should receive their previous test results before deciding whether to donate specimens for testing at the current visit, there was a delay between delivery of test results and post-test counseling and explanation of guidance. Future iterations of similar counseling protocols may be able to concurrently deliver testing results and post-test counseling if a rapid diagnostic test for detection of Ebola RNA in the body fluids of Ebola survivors becomes available.

The EVD epidemic had a devastating impact on the economic and social ties of communities across Sierra Leone [12, 21, 24]. EVD survivors commonly reported medical and mental health issues post-convalescence for which there was limited assistance in Sierra Leone [25]. These needs were far greater than the services the study could offer. This was particularly true for semen testing, as there were far more EVD survivors in Sierra Leone than could be enrolled in the EBOV persistence study. To this end, the Government of Sierra Leone initiated a comprehensive program for EVD survivors (CPES) in October 2015 that aimed to provide services for EVD survivors in Sierra Leone including counseling, semen testing, eye care, myalgia, and treatment for other EVD sequelae [26].

The mental health infrastructure of Sierra Leone is underdeveloped. There is limited capacity to identify and treat mental health disorders in the general public or EBOV-affected populations [12]. When the study was initiated, counseling staff were less familiar with behavioral counseling methods that emphasized dialogue regarding participants’ experiences with EVD survivorship, stigma, and intimate relationships. Learning these skills helped study counselors better assist participants who received multiple positive qRT-PCR test results to maintain participation in semen testing and openly discuss feelings of frustration and anxiety.

After observing initial high rates of HIV testing uptake in the first phase of the study, acceptance of HIV testing declined among study participants during the first visit. We hypothesize that participants may have felt anxious about their EBOV qRT-PCR tests and did not want to learn their HIV test results at the same time. We re-trained study counselors on HIV testing and counseling procedures and encouraged them to offer HIV tests at follow-up study visits and observed a rise in uptake of HIV testing. More formative research is needed to determine how best to incorporate both HIV and EBOV testing and counseling in future EVD epidemics.

One limitation of our behavioral counseling protocol is that, due to time constraints during study visits, we did not employ couples’ counseling. Future development of similar EBOV persistence behavioral counseling protocols may consider the inclusion of couples’ counseling to facilitate correct and consistent condom use, particularly for those participants who repeatedly test positive [27]. Consideration could also be given on how best to recruit, enroll, and counsel EVD survivors who are men who have sex with men (MSM), and who may be reluctant to discuss their sexual behavior in an environment where same sex behavior is criminalized [9], as well as survivors who may object to body fluid testing due to religious objections to masturbation or other specimen donation procedure.

Another limitation of the protocol is that we developed and implemented risk reduction guidance when relatively little was known about the persistence of Ebola RNA in the body fluids of survivors, or how qRT-PCR results related to the risk of transmitting Ebola to others. Since the implementation of this behavioral counseling protocol, additional data on the persistence of Ebola RNA in semen has become available from multiple Ebola-affected countries [28–31]. These data show Ebola RNA persisting in semen for lengthy periods of time for some Ebola survivors, and further illustrate the critical role that semen testing and behavioral counseling can play in Ebola epidemic control [32]. Future development of similar behavioral counseling protocols should consider these additional data in adapting behavioral guidance for Ebola survivors, such as collecting body fluid specimens more frequently and increasing the number of consecutive negative tests needed before testing ceases following the receipt of positive test results.

A behavioral counseling protocol that pairs test results with risk reduction behavioral guidance might help mitigate transmission risks associated with body fluids of EVD survivors in which EBOV has been detected. Risk reduction behavioral counseling is rapidly becoming an integral part of addressing the sexual transmission risk in this EVD epidemic and in filovirus outbreaks moving forward; its utility will likely also be swiftly demonstrated for other pathogens where virus persistence in body fluids may pose a risk for continued transmission from survivors [32]. A qualitative evaluation assessing the impact of the behavioral counseling protocol on study participants as well as staff perceptions of the protocol has been performed to add to lessons learned presented in this manuscript.

As of July 2016, the Ebola Virus Persistence Risk Reduction Behavioral Counseling Protocol has been used for more than 220 male and 120 female participants. The protocol has been adapted for use by other body fluid testing programs for EVD survivors, including the Government of Sierra Leone’s CPES (Alpren, C., personal communication) and Liberia’s Men’s Health Screening Program [29]. Lessons learned from implementation of the behavioral counseling protocol in the Sierra Leone Ebola Virus Persistence Study have also been shared to advise the operations of these semen testing programs [29]. We hope that our experience implementing this behavioral counseling protocol in the midst of an EVD epidemic in Sierra Leone can inform similar future efforts so that robust and effective services can be provided to EVD survivors.