Provision of the levonorgestrel intrauterine system in Nigeria: Provider perspectives and service delivery costs

Setting

In Nigeria, modern contraceptive use among all women stagnated around 11% between 2008 and 2013^18,19. Data from the 2018 Performance Monitoring & Accountability (PMA) 2020 survey found a modern contraceptive prevalence of 18%, suggesting some gains may have been made more recently, although the survey only spanned seven of Nigeria’s 36 states²⁰. Available estimates also highlight regional disparities, with more limited use of modern contraception in the North of the country^19,20.

In recent years, several implementing partners have successfully increased provision of implants and copper IUDs in Nigeria as part of a broader effort by the government and non-profit partners to increase access to long-acting reversible contraceptives (LARCs) and overall contraceptive prevalence²¹. In the PMA2020 survey, implants and IUDs accounted for 24% and 6% of the method mix among married women, respectively, but less than 3% of unmarried sexually active contraceptive users used each of these methods²⁰. Most women receive their contraceptive method from the private sector, although the public sector supplies the majority of LARCs and injectables¹⁹. Use of the private sector appears to be more common in the South than in the North²⁰.

Qualitative assessment

We conducted in-depth interviews (IDIs) with four providers in each of the five programs currently piloting LNG-IUS introduction in Nigeria (Table 2), for a total of 20 providers. For each program, we obtained lists of sites/providers from program managers and purposively selected providers from two regions, except for UCHI which has a single site. Regions were selected to provide representation from the North and the South of Nigeria while keeping in mind practical constraints linked to the coverage of the different programs and to visiting sites for interviews.

Program managers directly informed providers or provided letters of introduction to facilitate contacts. Potential participants were approached by phone to schedule interviews. Three local, Masters-level, female research assistants hired as consultants interviewed providers in English individually and in private at health facilities in July 2018. Each provider was interviewed once, using topic guides covering training, clinical experience with the LNG-IUS, perceived advantages and disadvantages of the method, and resources and activities for supporting service delivery (see Extended data)²². At the end of the interview, the research assistants used structured questions to elicit time estimates for completing certain tasks. IDIs were audio-recorded, then transcribed and uploaded to NVivo 12 for coding and applied thematic analysis. Transcripts were verified against audio-recordings by a supervisor for accuracy and completeness and were not returned to participants for validation. Three analysts shared transcripts for coding, running periodic checks for intercoder agreement on 25% of transcripts. Codes included a priori codes identified based on informational needs and data-driven codes that emerged from the initial reading of transcripts. Analytic memos were prepared to explore patterns in the data, after which we used Excel matrices to summarize and compare the prevalence of key themes across the five programs.

Costing assessment

The service delivery costing assessment was based on the introduction of the LNG-IUS with 71 trained providers in 40 social franchise clinics across 18 states in the SFH program between May 2017 and July 2019. The SFH program was selected for the costing exercise partly because of convenience (SFH is a partner on the broader research project that funded this assessment) and also because SFH is looking to introduce a new LNG-IUS product, Avibela, in 2020 and insights could help inform a broader introduction strategy.

Calculation of incremental costs of introducing LNG-IUS in SFH pilot settings

To estimate the costs of service delivery in SFH-supported clinics during the pilot phase, we used an Excel-based costing template (see Extended data)²² to collect input on the resources required for method provision, and on the associated unit and total costs. We included costs of direct labor from health care perspective for counseling, insertions and removals; costs of consumable supplies (except where costs were negligible including for antiseptic, soap, and iodine); costs of insertion/removal kits; and costs for provider training. Labor costs for method provision were informed by time estimates for counseling, insertion and removal collected through a questionnaire administered to all SFH providers (n=39) participating in a refresher training on the LNG-IUS in April 2018 (see Table 3).

The IUD/IUS kits were comprised of 13 re-usable instruments required for insertion and removal of either method; based on input from clinical experts, we assumed that each kit could be used for 500 insertions/removals (Table 4 and Table 5). Given that the LNG-IUS is currently donated to the SFH program, we did not include any associated cost for the LNG-IUS commodities for this component of the analysis.

SFH staff calculated costs for provider trainings (training of master trainers, initial cascade trainings with providers, and refresher training with providers) based on program expenditure reports. For demand creation, SFH used a provider-initiated awareness generation model. This involved the provider carrying out talks about contraception with women in the facilities who came for post-natal or child wellness visits. In these contexts, the provider would include the LNG-IUS in the context of a full method mix. As such, there was no incremental cost for demand creation as part of routine service delivery. Likewise, routine clinical supervision with LNG-IUS providers was conducted as part of supportive supervision that would have occurred anyway, so there was no incremental cost associated for ongoing clinical supervision for the LNG-IUS. Additional information about training and demand-generation activities are described in Table 6.

SFH staff provided service statistics on the numbers of LNG-IUS inserted and removed during the assessment timeframe (May 2017 through July 2019). We then calculated the total incremental direct cost of LNG-IUS introduction and divided by the total number of LNG-IUS insertions (1,949) to obtain a cost per LNG-IUS insertion.

Calculation of direct cost per CYP for all family planning methods

For the broader cost per CYP analysis comparing the direct service delivery costs of the LNG-IUS to those of other family planning methods in Nigeria, we used commodity costs to international procurers for all methods (taken from the UNFPA catalogue or, for the LNG-IUS, provided by Medicines360, the supplier of Avibela, which is the new, more affordable commercial product that is being registered by SFH in Nigeria). Although many family planning programs in Nigeria including SFH receive most of their contraceptive commodities for free by donation (typically from UNFPA or USAID), we wanted to include the commodity prices in this part of the analysis to better represent the economic cost to the health system.

To calculate direct service delivery costs per CYP, we followed a similar approach to that described for an earlier analysis in Kenya²³. Briefly, the analysis included commodity costs as described in Table 7, as well as costs of consumable supplies, estimated costs of instruments per client visit, and costs of direct labor for counseling, insertion, removal, and resupply if required for each method. We then used standard CYP conversion factors for each method²⁴. For short- and mid-acting methods, we aggregated costs of visits made throughout the year to achieve one CYP. For long-acting methods, we divided the costs by the appropriate conversion factor. Following current guidance, we used a conversion factor of 3.3 years for the LNG-IUS (assuming the method is labeled for five years of use). Consistent with previous analyses, the direct cost per CYP calculation does not include provider training costs because it assumes a steady state once providers have been oriented and trained on all family planning methods.

Ethical statement

The National Health Research Ethics Committee of Nigeria approved this assessment (NHREC/01/01/2007). This activity was also reviewed by FHI 360’s Protection of Human Subject Committee (PHSC) in the United States and deemed to be exempt from ethical approval because it was not human subjects research (1192089). Written informed consent to participate and to audio-record the interview was obtained from all providers.

Qualitative assessment

Provider IDIs lasted 27 minutes on average. Four MSION providers were replaced from the original pool of selected providers because facilities from the initial list were no longer in the MSION franchise network. Four replacements were also selected for DKT because providers had not yet offered the LNG-IUS product, and one because the trained provider had left the facility. Table 8 shows provider characteristics. Most providers were women, although all DKT providers were men. All providers had several years of experience providing the copper IUD. Average experience offering the LNG-IUS ranged from six months for UCHI providers to over five years for DKT providers (which likely refers to earlier experience with Mirena). Providers had varying levels of experience inserting the LNG-IUS. About half of all providers had removed an LNG-IUS at some point.

Findings from interviews with providers are presented in three main areas: perspectives on the LNG-IUS; capacity and resources; and other aspects of product introduction.

Providers’ perspectives on the LNG-IUS

Providers were aware of the differences between the LNG-IUS and the copper IUD, including different durations of action, different mechanisms of action, and different bleeding profiles. Most providers highlighted the fact that the LNG-IUS did not cause heavy bleeding as an advantage of the method over the copper IUD, with several providers also mentioning reduced cramping. Two providers explicitly said that the LNG-IUS was a better method than the copper IUD.

Approximately two-thirds of providers appreciated the clinical benefits of the LNG-IUS, especially for women with heavy bleeding or fibroids but also for endometriosis and anemia. Several providers mentioned that their LNG-IUS clients did not experience any side effects, and two providers explicitly noted minimal side effects as an advantage over other hormonal methods.

Most providers commented favorably on reduced bleeding and amenorrhea as consequences of method use. While several providers reported that their clients generally accepted bleeding changes with proper counseling, some providers noted mixed acceptability of amenorrhea among users. Several providers mentioned lifestyle benefits associated with reduced bleeding, including buying and using less sanitary pads and minimal interference with sexual activity.

Providers themselves did not appear to have concerns about the LNG-IUS as a method, and about half said that they typically did not receive any complaints from LNG-IUS users. However, several providers noted complaints by clients about initial spotting and at least two providers said they conducted pregnancy tests to reassure amenorrhoeic clients. A few providers also reported contraindication for women with active uterine infections as a disadvantage.

Some providers mentioned concerns around privacy and safety of uterine placement among women, and fears inherited from earlier myths about the copper IUD. A couple of providers in one program reported their clients’ general aversion to hormonal contraception.

With the exception of one provider who was not asked this question, all providers said they wanted to continue offering the LNG-IUS, although one had some reservations due to the amount of effort required at counseling to overcome fears and misperceptions. Reasons for wanting to continue offering the LNG-IUS included the non-contraceptive health benefits of the method, women liking the LNG-IUS, and expanding the method mix.

Capacity and resources

While all providers were experienced offering the copper IUD, they had varying exposure to the LNG-IUS prior to its introduction in each program. For example, one female provider said she had been an LNG-IUS user herself for the past 25 years, a few providers had been introduced to Mirena in earlier trainings, and a couple of providers reported prior experience inserting LNG-IUS products with their clients without receiving formal training. In contrast, a few providers had only heard about the method in books and one provider said it was new to them when they were trained.

Many providers indicated that the steps for inserting and removing the LNG-IUS were generally similar to those for the copper IUD, although they acknowledged small differences in the insertion process. One quarter of providers mentioned insertion challenges, including loading the first time and inserting in women with fibroids. Others reported no challenges, and one third of providers said they found loading easier with the LNG-IUS than with the copper IUD. Some providers found device placement more “technical” with the LNG-IUS because of the need to wait for the arms to open before pushing to the fundus.

When asked to estimate insertion times, 16 providers gave the same amount of time for the LNG-IUS as the copper IUD (around 11 minutes), three said that inserting the LNG-IUS was faster, and one that it took longer (results not shown but similar to those provided in Table 3). Most providers estimated that it took them more or the same amount of time to insert the LNG-IUS than a two-rod implant, and typically longer than to insert a one-rod implant. All providers who had already removed an LNG-IUS gave the same time estimate for removing the LNG-IUS and the copper IUD (six minutes). Removing the LNG-IUS took the same amount of time or less than removing a one-rod implant but was faster than removing a two-rod implant for most providers.

Over half of providers reported having sufficient equipment for LNG-IUS provision, while several other providers said they could manage but would benefit from additional equipment, and a couple of providers in one program reported that their instruments needed replacing due to repeat usage and decontamination. At least one provider in each of four programs indicated needing more consumables, and a few went on to explain that the costs of procuring consumables themselves were passed on to clients.

All providers described having protocols in place for infection prevention, with the main examples provided being disinfecting and sterilizing equipment, but also handwashing and wearing gloves, generally maintaining an aseptic environment, and properly disposing of waste. All providers in one program emphasized screening women for active pelvic infections. Although providers were generally confident about their protocols, some reported challenges including time requirements for autoclaving and procurement of consumables.

Other aspects of product introduction: Gaps in demand-generation, availability and pricing

Although providers typically talked about including all methods in their counseling to support informed choice by women, several providers said they recommended the LNG-IUS to women with heavy bleeding or fibroids, and a few others that they generally promoted the LNG-IUS more actively than the copper IUD, or in one case, implants, to their clients. One female provider who was an LNG-IUS user volunteered that she was using herself as an example with her clients to show the method did not cause any problems, while another provider said they showed empty LNG-IUS packs to prospective clients to convince them that other women were using the method.

Overall, providers observed that uptake of the LNG-IUS was limited to date, with a few providers saying that uptake of IUDs in general was low and several others noting that the LNG-IUS was currently less popular than the copper IUD. Most providers highlighted a general lack of awareness of the LNG-IUS. All providers in one program described an organized community outreach model that is used to generate demand and present the range of methods. Providers in other programs did not regularly engage in any regular outreach or education, with some reporting resource and staff constraints.

Recommended strategies to address the lack of awareness of the LNG-IUS included using the radio, posters and flyers or conducting health education in communities. Many providers also advised improving access by making the LNG-IUS available at more service delivery points, with a few providers noting that other places besides their organization or network that offered the method typically used Mirena, which was priced too high for women in their community.

Pricing strategies varied across programs (see Table 2). Some providers reported charging a minimal fee for consumables or client cards in addition to the recommended price, and a couple of providers revealed they sometimes charged wealthier clients more in order to lower the price for poorer women. Many providers in programs charging a product fee felt that cost remained a barrier, despite being lower than the price for Mirena (the price of Mirena was cited by some providers at around 40,000-45,000 naira or 110-125 USD).

Costing assessment

The estimated cost of LNG-IUS provision from May 2017 through July 2019 including training of 71 providers across 40 facilities and direct costs for labor, commodities and equipment was USD 66,300 with the LNG-IUS commodity donated for free by the ICA Foundation. During that period, 1,949 LNG-IUS units were inserted in the 40 clinics involved in the pilot; as such, the average per unit direct cost was USD 34 (see Table 9). Of the total service delivery costs, 92% (USD 61,530) was comprised of the training inclusive of the original training of trainers, cascade trainings to the 40 clinics and refresher trainings among previously trained and newly trained staff.

The direct delivery costs per CYP are presented in Figure 1. The cost per CYP of the Avibela LNG-IUS (USD 5.64) is slightly higher than that of the one-rod implant, Nexplanon (USD 5.23), and is slightly lower cost per CYP than injectable methods (USD 6.16 for Sayana Press and USD 6.59 for Depo Provera). Injectables are currently the most commonly used method in Nigeria.

Figure 1. Direct service delivery costs per couple years of protection (CYP) in USD for various family planning methods.

Underlying data

Full qualitative transcripts are not available for ethical reasons because even after removing directly identifiable information such as names and addresses, participant identity may be difficult to fully conceal, and research locations may remain potentially identifiable, presenting a risk of deductive disclosure. However, topic guides, codebooks and relevant excerpts of transcripts are available from the authors on reasonable request. Requests should be sent to the corresponding author at krademacher@fhi360.org or to familyplanning@fhi360.org. Requests will be granted to researchers for the purposes of comparative analysis, upon approval from relevant ethics committees.

Extended data

Harvard Dataverse: Provision of the levonorgestrel intrauterine system in Nigeria. https://doi.org/10.7910/DVN/7JMETP²²

This project contains the following extended data:

Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).