Delivering integrated care after sexual violence in the Democratic Republic of the Congo

Setting

The programme was implemented in the South Kivu Province in eastern DRC at Panzi Hospital and a subset of 12 rural clinics that refer GBV survivors to Panzi Hospital. The South Kivu Province, home to 4.614 million people, has been an epicentre of armed conflict and political instability.2 16 Since its inception in 1999, Panzi Hospital, an internationally recognised centre for the treatment of rape survivors, has treated more than 85 864 girls and women with gynaecological injuries, more than 48 000 of whom are survivors of sexual violence.17 Panzi Hospital provides comprehensive GBV care including post-rape medical care, gynaecological surgery, psychosocial and psychiatric care, and referral to safe havens and legal programmes. As a referral facility for the entire South Kivu Province, Panzi Hospital receives GBV survivors days, weeks and even months after violence, with many care-seekers arriving outside of the time window for effective emergency contraception and HIV post-exposure prophylaxis. The 12 rural sites were selected by Panzi Hospital based on the following characteristics: (1) history of GBV referrals to Panzi Hospital, (2) accessibility by vehicle and (3) known absence of post-rape medical kits.

Community sensitisation

Panzi Hospital staff trained community health workers to increase community-level awareness of the importance, time sensitivity and availability of emergency contraception, HIV prophylaxis and STI treatment after sexual violence. In addition, radio journalists broadcast public service announcements with similar messaging that were recorded by programme staff.

The Prevention Pack

We designed the Prevention Pack (figure 1), a pre-packaged post-rape medical kit. The following medications were included in the carton: (1) emtricitabine 200 mg+tenofovir disoproxil fumarate 300 mg to prevent HIV infection, 30 tablets, to be taken orally once daily for 30 days; (2) azithromycin to treat STIs, 1 g tablet, one-time oral dose; and (3) levonorgestrel to prevent pregnancy, a single package of two 0.75 mg tablets for one-time oral dosing. The emtricitabine/tenofovir regimen is an accepted antiretroviral regimen within the WHO guidelines and was selected for its tolerability.18 Azithromycin was selected as a single-dose first-line prophylactic agent against Chlamydia trachomatis; while no longer recommended as a first-line treatment, azithromycin is generally active against Neisseria gonorrhoeae.19 The medications were pre-packaged in a carton to eliminate the need for front-line providers to assemble the multidrug prescription. Each carton contained a punch-through calendar card for tracking the emtricitabine/tenofovir disoproxil fumarate doses. Global Strategies designed the packaging and purchased the azithromycin and levonorgestrel. Gilead Pharmaceuticals provided 4000 treatment courses (30 days) of emtricitabine/tenofovir disoproxil fumarate. The Prevention Packs were assembled by Panzi Hospital pharmacy staff and stored in the Panzi main pharmacy.

Inventory management system

Logistimo (www.logistimo.com) is a cloud-based inventory management software that allows users to create a map of locations that can receive and dispense inventory. Logistimo also allows data managers to pre-select stock levels that trigger text message alerts to project personnel. The Logistimo platform is a general inventory software and not specific to GBV. We adapted Logistimo to track both medication inventory and the number of rape survivors received at each clinic.

Training

In collaboration with Panzi Hospital’s GBV division, we developed a training curriculum that complemented Panzi Hospital’s existing GBV model. The Panzi Model is a holistic person-centred care model that encompasses four pillars: medical, psychosocial, legal and socioeconomic.20 As the rural sites did not previously have access to post-rape medical kits, we added modules to the Panzi curriculum covering the following areas: HIV prevention, infection, diagnosis and treatment; HIV post-exposure prophylaxis, prophylactic treatment of other STIs and emergency contraception; how to safely administer antiretroviral drugs; side effects of post-rape medications; sensitivity to issues of violence, confidentiality, record keeping, drug security, the Prevention Pack and inventory management.

Training was conducted by five medical doctors, a psychologist from Panzi Hospital and Congolese Prevention Pack Program personnel. The training team used didactics and role-playing exercises to teach the curriculum. Participants were introduced to the data manager responsible for the daily inventory phone calls and a representative from each site was designated to receive the calls.

Between February 2013 and February 2017, 40 community healthcare workers, 61 nurses, 6 physicians, 3 laboratory technicians and 3 journalists from local radio stations received a 3-day comprehensive training in GBV and specific information related to the Prevention Pack Program. Ongoing refresher training, as well as training for new staff members, was provided as needed.

Provision of Prevention Pack

Clinic staff evaluated individuals who presented for post-rape care. Individuals were considered eligible for, and were offered, the Prevention Pack if the rape had occurred within the last 72 hours, the victim was at least 12 years old, was not HIV infected and did not have a history of adverse reactions to the medications in the Prevention Pack. HIV testing was offered at the time of presentation. For visibly pregnant women, women who self-reported pregnancy and men, the kit was provided, but levonorgestrel was removed. Pregnancy testing was not required before Prevention Pack administration as a single dose of levonorgestrel is considered safe during pregnancy.21 Caregivers of children younger than 12 years presenting at rural clinics were given a referral note and directed to the nearest hospital. Though not required, it was strongly recommended that the first doses of medication be taken in the clinic under direct observation by the nurse. All individuals were instructed to return for follow-up care in 1 week and at 1 month.

Individuals with HIV who presented within 72 hours were offered the individual components, azithromycin and levonorgestrel. HIV-infected individuals not reporting to be receiving antiretroviral treatment were provided with a referral note and directed to the nearest HIV clinic.

Individuals presenting more than 72 hours after rape received HIV testing, azithromycin and referral to additional services as needed. To reduce the possibility that a Prevention Pack would be opened to access any element of its contents for indications other than rape within the last 72 hours, participating sites were provided with additional doses of azithromycin and levonorgestrel.

Assessment of the demand and consumption of Prevention Pack

To assess the demand and consumption of Prevention Pack inventory, we tracked the number of eligible survivors, ineligible survivors and kit inventory using Logistimo (figure 2). The data manager entered each of the 13 sites into Logistimo along with its Global Positioning System (GPS) coordinates. These sites were configured as ‘entities’ that could receive and dispense inventory. Participating sites received an initial stock of between 10 and 20 kits based on the self-reported demand for post-rape care medications at each clinic at the initiation of the programme. In addition to the clinics, the Panzi Hospital main pharmacy was entered into Logistimo as the inventory origin point. The Logistimo system is password protected and does not include individual identifiers.

The project data manager phoned nurses daily at each of the 13 sites. We selected phone communication because it allowed for identification of staff turnover and the ability to limit costs to the call initiator. Nurses at each site reported on the number of individuals presenting for post-rape care since the last phone call. No identifying information was shared.

Individuals were considered eligible for treatment if they presented within 72 hours after rape, were more than 11 years old and were HIV negative. Ineligible individuals either presented after 72 hours, were less than 12 years old, tested positive for HIV or self-reported being HIV infected. The specific cause of ineligibility was not recorded. Nurses also reported the number of Prevention Packs administered and any doses of azithromycin or levonorgestrel that were administered from the separate stock of those medications. The project data manager then entered the data into Logistimo.

Approximately every 3 months, project personnel conducted field visits and restocked each site. Project personnel tailored restocking quantity based on current stock level and consumption rate. Two methods were used to monitor inventory between field visits. Key project personnel received text messages through Logistimo when any individual clinic’s stock went below five Prevention Packs. In addition, project personnel monitored consumption patterns to analyse for mass violence events and increasing consumption rates that might precipitate stockouts. Project personnel responded to stock concerns by transporting inventory from the Panzi Hospital main pharmacy or authorising the transfer of inventory from neighbouring clinics. The decision between the two options was based on several factors including time to next scheduled field visit, current inventory level, kit consumption rate, proximity to other sites, security and weather/road conditions.

Uptake of post-rape medical kits in rural areas

Our results demonstrate that there is a high demand for post-rape medical kits in remote rural settings in the eastern Congo. While the WHO guidelines provide a framework for selecting medications for post-rape medical kits, there is little guidance or research regarding the optimal geographical distribution of pharmaceutical resources. Our work demonstrates the limitation of restricting post-rape care medications to urban specialty centres that require rape survivors to travel significant distances.

Streamlined and simplified care

Studies have demonstrated low percentages of rape survivors receiving HIV post-exposure prophylaxis as a result of poor provider training.22 In our study, a 3-day training, followed by intermittent refreshers and onboarding of new staff, resulted in the continuous prescription of post-rape medical kits to all eligible survivors over a 4-year period in remote primary healthcare clinics. There are a number of possible explanations for this result. First, the Prevention Pack is pre-bundled. This low-cost intervention should be considered to reduce barriers to prescription and facilitate inventory tracking. Second, our study used eligibility criteria for HIV post-exposure prophylaxis consistent with the DRC and WHO guidelines that emphasise the treatment of all rape survivors who are HIV negative or unknown.23 24 While there is agreement that HIV post-exposure prophylaxis should be offered to survivors of sexual violence, there is variation in the extent to which different guidelines emphasise treating all eligible survivors versus avoiding treatment of ineligible survivors. These subtle variations in how eligibility criteria are presented may result in variations in prescribing HIV post-exposure prophylaxis.

Near real-time tracking of inventory and post-rape care demand

The widespread stockouts of medications may be the single largest barrier to the timely delivery of emergency contraception, HIV prophylaxis and STI treatment to survivors of sexual violence.13 Our work demonstrates that stockouts can be prevented, even in remote rural areas. One possible explanation is that the Prevention Pack Program uses a push system of proactive inventory phone calls rather than a pull system that relies on rural clinics to report inventory problems. In our experience, we found that a push system facilitates implementation by removing the requirement that busy providers know they have stocked out of an infrequently used or recently introduced medical supply. Additionally, the Logistimo software allows a centralised data manager to respond to low stock levels by rapidly identifying nearby stock and facilitating transfers between clinics. Scaling this approach will require reducing implementation costs, while demonstrating the value of the system in reducing restocking visits, minimising medication waste and capturing data on demand for post-rape care.

Limitations

We acknowledge several limitations of our programme. The high number of survivors presenting at Panzi Hospital ineligible for Prevention Pack likely reflects the limited reach of our intervention: only a small subset of the clinics that refer to Panzi Hospital were incorporated into the Prevention Pack Program. Panzi Hospital is a referral centre and survivors of GBV often travel for multiple days to reach the facility. We were not able to differentiate the causes of ineligibility with current data collection mechanisms. The timing of presentation for individuals presenting after 72 hours was also not collected. Both of these data points would better delineate the ineligible population. While individuals who received Prevention Pack were instructed to return for follow-up, data were not collected to determine if they became pregnant, HIV infected or developed STIs. It was not feasible to collect this information. Though the medication regimen was designed to be more tolerable based on known side effect profiles,25 26 it was not feasible to assess the tolerability and adherence to the full course of medication treatment. The true denominator of rapes in this population is unknown, as is the number of persons seeking care after sexual violence at participating sites prior to this intervention.