Introduction

Onchocerciasis is a severely debilitating, vector-borne neglected tropical disease (NTD) caused by infection with the filarial nematode Onchocerca volvulus and transmitted among humans by the bites of Simulium blackflies. Although the great majority (99%) of the cases are found in sub-Saharan Africa (SSA), the infection also occurs in Yemen and in two (Venezuela and Brazil) out of the six original endemic countries in Latin America [1]. In 2017, it was estimated that at least 220 million people required preventive chemotherapy against onchocerciasis, and according to the Global Burden of Disease (GBD) Study for that year, 14.6 million of those infected had skin disease and 1.15 million had vision loss [2]. In 2021, the GBD Study estimated that 19.6 (95% Uncertainty Interval, UI = 17.8–21.7) million people were infected, and that the disease was responsible for 1.26 (95%UI = 0.753–1.90) million disability-adjusted life-years (DALYs) [3].

Preventive chemotherapy relies on mass drug administration (MDA) delivered, in Africa, as community-directed treatment with ivermectin (CDTI), predominantly annually, but in several foci of some countries (such as Ethiopia, Ghana, Nigeria, Togo and Uganda), also 6-monthly (biannually). Ivermectin effectively kills microfilariae of O. volvulus and the rapid reduction in microfilarial load in the skin and eyes alleviates disease symptoms like skin lesions, itching, and visual impairment [4]. Ivermectin temporarily inhibits the release of new microfilariae by adult female worms (macrofilariae) leading to the transient decline in microfilarial density, however microfilaria production eventually resumes [5]. Additionally, repetitive doses of ivermectin have been known to exert a partial macrofilaricidal effect on the parasite, culminatively reducing the lifespan and fertility of adult female worm. The cumulative microfilaricidal and embryostatic effects of ivermectin on O. volvulus parasites are critical for interrupting the parasite’s life cycle and achieving onchocerciasis elimination [5]. Therefore, community-directed treatment with ivermectin (CDTI) must be sustained over many years to ensure the cumulative effects of the drug are sufficient to achieve elimination of O. volvulus transmission (EOT) [6]. With ivermectin MDA, EOT has been reported for some foci of Mali, Senegal, Nigeria and Sudan, without vector control [7]. In consequence, the World Health Organization (WHO) 2021–2030 NTD roadmap has proposed that 12 endemic countries be verified for EOT of onchocerciasis by 2030 [8].

The WHO has delineated three distinct phases in onchocerciasis elimination programmes: I) the treatment phase with CDTI, during which monitoring and evaluation are conducted periodically to ascertain progress towards transmission suppression and ultimately interruption; II) the post-treatment surveillance (PTS) phase, during which CDTI is no longer delivered but indicators of exposure and transmission are evaluated to confirm that EOT has been reached, and III) the post-elimination surveillance (PES) phase, to alert of potential resurgence of transmission or re-introduction of infection [9]. These distinct phases require different diagnostic tools that allow transitioning from one to the next. During phase I, and until Stop-MDA surveys are conducted, the gold standard for parasitological diagnosis and calculation of epidemiological indicators (microfilarial prevalence and load), has been the detection and enumeration of O. volvulus microfilariae by the so-called skin-snip microscopy method [10]. Skin-snip microscopy is highly specific provided that microfilarial morphometric characteristics are assessed, and able to detect active infection (as microfilariae are the embryonic progeny of reproductively active adult worms), but its sensitivity (based on two snips) decreases as microfilarial prevalence and load decline due to CDTI [11]. The procedure is also mildly invasive and resource-intensive, and its acceptability diminishes in communities where decades of skin-snipping have occurred [12].

Skin-snip Polymerase Chain Reaction (PCR) has been suggested to circumvent the problem of low sensitivity posed by dwindling skin microfilarial loads, to confirm the infection status of children seropositive for IgG4 antibodies against the Ov16 O. volvulus antigen, or when species-specific identification of microfilariae may be necessary [13]. However, it is not practical for large-scale community surveillance in SSA because it is expensive to purchase and operate and requires highly-skilled personnel. Serological, enzyme-linked immunosorbent assay (ELISA) tests based on the detection and quantification of IgG antibodies to the O. volvulus Ov16 recombinant antigen have shown, in longitudinal human and chimpanzee studies, that anti-Ov16 seropositivity may precede skin-snip positivity, providing a marker of exposure to early infection stages [14]. Other studies, also in chimpanzees, have indicated that IgG4 seroconversion against Ov16 coincides with the onset of patency (detectable skin microfilariae) [15], but as its duration may exceed that of patent infection, anti-Ov16 seropositivity may not yield a useful marker of active infection. Besides, ELISA-based serological tests also pose challenges including the need for laboratory analysis and the lack of a standardized, commercially-available Ov16 ELISA test, resulting in marked inter-laboratory variation [12]. A field-friendly, rapid diagnostic test (RDT) with the potential for integration into current onchocerciases surveillance activities in endemic countries, was developed and made commercially available (SD BIOLINE Onchocerciasis IgG4 rapid test) [16,17]. The performance of this Ov16 RDT has been evaluated in the field and current research priorities focus on operational and implementation research to demonstrate its usefulness, particularly in communities with low infection prevalence owing to many years of CDTI [18].

In Ghana, ivermectin MDA has been implemented for more than 30 years, having started in the late 1980s [19]. In 2009, a nationwide rapid epidemiological mapping of onchocerciasis (REMO) evaluation was conducted to provide updated information on the distribution of the disease. Areas with an infection (palpable nodule) prevalence in adult male samples above 20%, were allocated to a biannual treatment frequency, also considering a buffer zone of 20 km around these CDTI priority areas [20]. Forty-four districts were classified as meso- and hyperendemic and have received biannual MDA since 2010. Forty-one districts, which were re-classified as of low endemicity and were already on MDA before the REMO exercise, continued to receive annual MDA. From 2015 to 2020, efforts were intensified to accelerate the shift from onchocerciasis control to EOT. An operational threshold of microfilarial prevalence <1% was adopted, the implementation unit changed from community to sub-district, and the geographical and (reported) therapeutic coverage increased substantially. In 2017, an impact assessment survey was carried out using Ov16 serology in children aged <10 years and skin-snip microscopy in adults aged ≥20 years [21]. Together with historical data collected by the Onchocerciasis Control Programme in West Africa (OCP), the impact assessment helped the national NTD Programme to formulate strategies to reach EOT. Delineation epidemiological surveys were conducted in 50 hypoendemic districts that had never received ivermectin (not having been previously prioritized for control but in need of re-assessment for elimination), and treatment strategies and management decisions were redefined [20]. Currently, 137 districts require or are under biannual CDTI [21].

Previous studies in the Bono region of Ghana, in which some communities had been under biannual CDTI since 2010, revealed that O. volvulus infection and associated cutaneous and ocular morbidities persisted despite 27 years of treatment [21,22], and that despite high levels of reported coverage [22], treatment adherence was relatively poor [23]. Although such communities may not be ready to move from the treatment phase to Stop-MDA surveys or may require alternative interventions to accelerate progress [24], it is important to understand the role of different diagnostics for surveillance activities.

As Ghana transits from the treatment phase to the next stages of the EOT pathway in some endemic areas, and to evaluate progress towards suppression and interruption of transmission, the suitability of current diagnostic tools to accompany such transition needs to be ascertained. Whilst studies from other countries have investigated the acceptability, usability, and cost of the Ov16 RDT [12], no such study has been conducted in Ghana. Therefore, this study aimed to assess the diagnostic attributes, technician and community acceptability, usability, and cost of the Ov16 RDT in onchocerciasis surveillance activities in Ghana.

Materials and methods

Study area

The study was performed in six communities, namely, four communities in the Tain District (Abekwai 2, Abekwai 3, Attakrom and Kokomba) and two communities in the Wenchi Municipality (Blibor and Jonnykrom/Adamakrom), in the Bono Region of Ghana. Fig 1 shows the locations of the study villages. The communities were selected based on the following criteria informed by previous studies [21,22]: i) they had persistent endemicity for O. volvulus infection, albeit with relatively low prevalence; ii) they had received ivermectin MDA for at least 20 years and biannual CDTI for about 10, and iii) they were situated close to fast-flowing rivers (the Subin, the Tain, or their tributaries) where the blackfly vectors breed. These criteria make the study communities representative of other onchocerciasis hotspots, which continue to exhibit persistent microfilarial infections and active blackfly breeding despite decades of CDTI and vector control interventions [26].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Map of study communities in the Tain District and Wenchi Municipality, Bono Region, Ghana.

The light pink area to the right of the study area represents the Nkoranza North District. The study communities fall within the so-called transitional zone with mainly savannah type vegetation. The map data used in this document is attributable to OpenStreetMap contributors and is licensed under the Open Database License (ODbL). More information can be found at https://www.openstreetmap.org/copyright.

https://doi.org/10.1371/journal.pntd.0012191.g001

Sample size and statistical analysis

A formal sample size was not calculated a priori for the study sites; instead a convenience sampling approach was used, yielding an average sample size equivalent to 12% (range = 5%–33%) of the respective community populations (Table 1). Nonetheless, we aimed to achieve at least 30 participants per community in line with typical thresholds reported in onchocerciasis epidemiological studies [29].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Number of participating communities, their sex composition, median age, and numbers examined for each onchocerciasis diagnostic.

https://doi.org/10.1371/journal.pntd.0012191.t001

Data were entered in purposely designed MS Excel spreadsheets (one for the epidemiology results and another for the cost menus). Analyses were conducted using Jamovi Desktop (version 2.3.19) and GraphPad Prism 8 for macOS (version 8.2.1). Descriptive statistical analysis was performed first to observe the characteristics of the variables. The main statistical analysis consisted of univariable and multivariable logistic regression with crude odds ratio (AOR) and adjusted odds ratio (AOR), respectively, to identify factors associated with Ov16 seropositivity in the study communities. When reporting proportions, 95%CIs were calculated using the Wilson score method [30]. Proportions were compared by assessing their 95%CIs or by using the z-statistic where appropriate (non-paired data), in which case asymptotic 95%CIs are reported [31]. The significance level for all analyses was set at 0.05.

Results

Of the total of 254 individuals, there were 253 with recorded sex, consisting of 136 (53.8%) females and 117 (46.3%) males. Age was recorded for 242 individuals, with a median age of 31 (range = 5–83) years (Table 1).

A total of 254 individuals (100%) consented to the Ov16 RDT, 94 (37%) to skin-snip microscopy, and 254 (100%) to nodule palpation. Consequently, 94 participants underwent all three tests. Among this subset, the median age was 31.0 years (range = 5–83), 53.8% were female, 70.5% were farmers and 52.9% had lived in the community for more than one year.

The proportion of the population for whom microfilaridermia was assessed ranged from 26% to 47%.

Of the 254 individuals tested for Ov16 RDT, 60 (23.6%, 95%CI = 18.8%–29.2%) were seropositive, compared to 11 (11.7%, 95%CI = 6.7%–19.8%) microfilaria-positive individuals out of the 94 with skin-snip results (z-value = 2.5; P-value = 0.0143). For palpable nodules, 14 out of 254 individuals examined were positive (5.5%, 95%CI = 3.3%–9.0%). The positivity of palpable nodules was only marginally significantly different from that of microfilaridermia (z-value = 2, P-value = 0.0471), but highly significantly different from that of anti-Ov16 seroprevalence (z-value = 5.8, P-value <0.0001).

The distribution of Ov16 seropositivity, skin snip positivity and nodule palpation result based by dichotomised sociodemographic variables is presented in Table 2. Being female, over 30 years of age, being a farmer, living in proximity to streams and frequently visiting streams were all associated with higher odds of being anti-Ov16 seropositive.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Positivity rates of the three onchocerciasis diagnostic tests and multivariable analysis of associated sociodemographic variables.

https://doi.org/10.1371/journal.pntd.0012191.t002

Among individuals who tested anti-Ov16 seropositive, 10.5% (95%CI = 2.9%–31.4%) also had a positive skin snip, and 11.7% (95%CI = 5.8%–22.2%) presented with palpable onchocercomas.

Of the 17 participants who were anti-Ov16 positive but skin-snip negative, 13 (76.5%) were female and all 17 had taken ivermectin in the previous year. No participant tested positive by all three tests.

The results of the age and sex distribution of anti-Ov16 RDT, skin-snip and palpable nodule positivity is provided in S1 Table.

In the 5–9 age group, 3/27 children (11.1%, 95%CI = 3.9%–28.1%) were anti-Ov16 seropositive, 3/13 children (23.1%, 95%CI = 8.2%–50.3%) skin-snip positive, and 1/27 (3.7%, 95%CI = 0.7%–18.3%) nodule positive.

Cost analysis for Ov16 RDT and skin-snip microscopy

Time considerations. The training of nine technicians for the Ov16 RDT took a maximum of two hours (120 minutes) in contrast to 14 days (6,720 minutes for an 8-hour daily schedule) for the training of five technicians on skin-snip microscopy. The Ov16 RDT took approximately 10 minutes to take a sample from each participant (labelling the cassette, selecting, and cleaning a finger, pricking the finger and placing the blood and the assay diluent in the cassette) and 30–40 minutes to get and communicate the results (30 minutes to read the results from the cassette and 10–20 minutes to record such results and convey them to participants). By contrast, skin-snip microscopy took about 30 minutes for sampling from each participant (cleaning left and right iliac crests with alcohol swabs, taking the skin snips with disinfected punches, placing the snips in labelled microtitre plate wells containing incubation medium, recording participant details), 24 hours for the incubation of the snips and about 20 minutes to pipette and observe the incubation medium under a microscope under two magnifications, for a total of 24 hours and 50 minutes (1,490 minutes) to obtain the results. A common concern was that the participants of skin-snip microscopy typically never get to know their results or at best get to know them only during the next field visit, which can take weeks or months. These have cost implications for both Ov16 RDT and skin-snip microscopy tests.

Cost considerations. The total cost for the activities targeting 400 participants was estimated at US$32,156. The cost was apportioned to Ov16 RDT, skin-snip microscopy and shared cost. The costs incurred for Ov16 RDT and skin-snip microscopy test and activities were US$2,412 (7.5%) and US$22,670 (70.5%), respectively. The shared cost incurred accounted for 22% of the total cost of the study (US$7,074). Within the total cost, 94.8% (US$30,484) corresponded to fieldwork cost, comprised of transport and lodging (72.0%), supplies, devices and instruments (21.4%), and labour cost (1.4%). Training and reporting costs accounted for 5.1% (US$1,640) and 0.1% (US$32) of the total cost, respectively. The total cost per participant of the Ov16 RDT was US$ 23.72 (US$6.03 for test-specific costs and US$17.69 for shared costs); the cost of skin-snip microscopy was US$74.36 (US$56.67 for test-specific costs and US$17.69 for shared costs). The cost per participant of deploying Ov16 RDT and skin-snip microscopy testing based on 400 participants and split by category is presented in Fig 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Cost analysis per Ov16 RDT and skin-snip microscopy for 400 individuals tested for each diagnostic.

https://doi.org/10.1371/journal.pntd.0012191.g002

Proportionally reducing costs according to the numbers tested in our study (254 for Ov16 RDT and 94 for skin-snip microscopy) and adjusting for fixed inputs, the total cost of the activity was estimated at US$24,708 for the 6 communities surveyed, with a net cost per participant of US$27 for Ov16 RDT and US$207 for skin-snip microscopy, highlighting the impact of the volume effect [10].

Discussion

This study aimed to determine the usability, acceptability, and cost of the Ov16 RDT for onchocerciasis surveillance in endemic communities in the middle belt of Ghana, particularly in the Tain District and Wenchi Municipality where it has been reported that onchocerciasis persists after 27 years of ivermectin MDA [21,22]. The positivity of anti-Ov16 IgG4 antibodies among study participants was 24%, the microfilarial positivity was 12% and the nodule prevalence was 6%. Female sex, age over 30 years, being a farmer, residence near streams and frequent stream exposure were all associated with higher odds of anti-Ov16 seropositivity (Table 2). Among the anti-Ov16 seropositives, there was a significantly higher proportion of females, those aged 30 years or older, living near streams, visiting such streams, and engaged in farming. However, the proportions of those presenting skin microfilariae or palpable nodules were significantly lower.

Since the study communities have received ivermectin MDA for nearly three decades, those aged ≥30 years at the time of our study would have been exposed as children to baseline levels of transmission. We have previously reported that the study area was initially mesoendemic, with a baseline microfilarial positivity of 41% to 48% [21]. This is broadly in agreement with the 33% seropositivity we found among those 30 years or older. Of the 126 females sampled, more than half (61%) were aged ≥30 years, whereas of the 116 males sampled, less than half (44%) belonged to this age group; nearly 100% of those 30 years of age or older were farmers (76/77 females and 50/51 males).

As the blackfly vectors breed in fast-flowing rivers and streams [1], it is perhaps not surprising that those living near or visiting such water bodies were more frequently represented among seropositive individuals, but contrary to expectation the proportion of those visiting streams daily was lower than that of those visiting less often. Farming can be another activity that increases exposure to blackfly bites, and historically, agricultural communities are known to have been highly endemic for onchocerciasis as farms are often situated near rivers that provide water for crops [32]. As blackflies are outdoor and diurnal biters, farmers working outdoors for several hours of the day may have an increased risk of exposure. In the Tombel Health District, Southwest region of Cameroon, and after 15 years of CDTI, Nyagang et al. [33] reported higher onchocerciasis prevalence in those aged more than 60 years, who were mostly farmers. The greatest prevalence of infection (2.1%) was among farmers [33]. A study in Southwest Ethiopia, also after 15 years of CDTI, reported that individuals aged ≥35 years had a microfilarial prevalence of 15.7% compared to 1.4% in those aged 15–24-years (no children under 15 were examined), and that approximately 60% of the participants were farmers, of whom 9% were microfilaria-positive [34]. The authors stated that adults were engaged in outdoor activities, increasing their exposure to blackfly bites [34].

Although anti-Ov16 IgG4 seroprevalence levels, particularly in the young, have been considered to indicate exposure to infection, knowledge of the parasite stage(s) that elicit IgG4 antibody seroconversion to the Ov16 antigen, as well as of the dynamics of seroconversion and potential seroreversion or antibody decay under transmission interruption remain poorly understood. A recent study by Cama et al. [35], estimated the antibody response half-life to be 3.3 (2.7–4.1) years, indicating that the majority of seropositive people will serorevert over time following interruption of transmission. Willen et al. [36,37] reported, using data from Ghana, on the development of an IgG-based immunoassay to quantify antibodies to Simulium damnosum sensu lato saliva that, if combined with assays to detect past or current exposure to O. volvulus could help to investigate exposure to blackfly bites and to the parasite.

In our study, anti-Ov16, skin-snip and nodule positivity were generally higher among those aged 30 years or older (except for microfilarial prevalence, which was lowest in those ≥40 years of age). Since only 94/254 (37%) of the participants agreed to be skin snipped, with complete data on age and sex for 91 of these, our results need to be interpreted with caution. Notwithstanding our small sample sizes, 11%, 23% and 4% of the 5–9-year olds examined were positive for IgG4 antibodies against Ov16, microfilaridermia and onchocercomas, respectively (S1 Table). As the nodule-positive child was aged 5 and the pre-patent period (from infection to detectable microfilaridermia) of O. volvulus has been estimated to range between 1 and 3 years [38], we can infer that this child became infected when he was 3–4 years of age, underscoring the occurrence of ongoing transmission in the study villages. Although anti-Ov16 seroprevalence by age was generally higher for females compared to males (except for the 5–9-year age group), sample sizes were small, leading to wide 95%CIs (Fig 3), precluding us from drawing firm conclusions about sex- and age-specific exposure patterns.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Anti-Ov16 seropositivity by age group and sex among 242 participants with recorded age and tested by Ov16 RDT.

Hatched bars: Females, black bars: males, error bars: 95% confidence intervals.

https://doi.org/10.1371/journal.pntd.0012191.g003

There were differences in positivity of anti-Ov16 and skin-snip positivity. This is likely explained by the fact that treatment leads to a reduction in skin microfilariae but antibodies to Ov16 may remain for much longer [12], with the levels of anti-Ov16 and skin-snip positivity becoming increasingly disassociated as treatment progresses. The finding of low anti-Ov16 seropositivity by RDT in skin-snip positive individuals is not totally unexpected, as it has been documented elsewhere [39]. Two factors may play a role; firstly, a proportion of infected individuals may not be able to mount an IgG4 antibody response to the Ov16 antigen (~20% has been reported in some studies, but between 5–10% in others [39–41]). Cross-reaction of Ov16 antibodies with other filarial worms is unlikely to influence the findings as only cross-reaction with Mansonella ozzardi, which is not present in Africa, has been reported [42]. Secondly, microfilariae of Mansonella spp. (rather than of O. volvulus) could potentially be present in some skin snips [43]. However, microfilariae of M. perstans are blood-dwelling and hence unlikely to have contaminated the bloodless skin snips we endeavoured to take [43]. Microfilariae of M. streptocerca are skin-dwelling, but the distribution of this parasite is confined to tropical rainforest areas [43], whereas our study site is mostly savannah (see Fig 1). Therefore, it is unlikely that the presence of Mansonella spp. microfilariae explains the discrepancy between skin-snip and anti-Ov16 positivity by RDT. Our results agree with reports of low Ov16 RDT sensitivity in areas with long-term treatment [39].

An interesting observation from our study was that the proportion of nodule positivity was about half that of skin snip positivity which was also about half the proportion of Ov16 positivity. This pattern warrants further investigation, as it may have practical implications for control programmes. If consistent it could enable the estimation of one diagnostic outcome (e.g., Ov16 seropositivity, skin snip positivity, or nodule prevalence) based on knowledge of another, thereby informing surveillance and decision-making in onchocerciasis-endemic areas.

We acknowledge that the WHO guidelines for stopping MDA and verifying elimination of transmission [29] do not advocate the use of anti-Ov16 serology to detect active infection or its use in all ages, but rather in those under 10 years of age using ELISA, with a seropositivity <0.1% at the upper 95% confidence upper bound in a sample of 2,000 children being taken as an indication of transmission interruption [29]. Modelling studies have investigated the applicability, to predict EOT outcomes, of this (and less stringent) threshold(s) and age group(s) under different assumptions of exposure and regulation of parasite abundance [44], and ongoing studies (some in Ghana) are being conducted to test such thresholds in the field.

We found that conducting Ov16 RDT testing in all ages provided useful information interpretable in the context of the duration of treatment in the area, as done by others in Mali and Tanzania [45,46]. Our findings also indicate that the study communities are not yet at the point of commencing Stop-MDA surveys despite nearly three decades of ivermectin MDA. We have documented the continued presence of O. volvulus infection and its associated clinical manifestations in these communities [21,22], and investigated factors influencing treatment adherence which may partly account for the persistence of transmission [23]. Entomological studies should also be undertaken in the study area to assess infection levels in simuliid population samples.

Studies done in other African settings with long-term CDTI (using the Ov16 RDT with whole blood) have reported, in the Ulanga district of Morogoro, Tanzania (an initially hyperendemic area), seroprevalence levels of 14% (27/191) in those aged 6–10 years and 33% (26/79) in those aged 11–12 after more than 20 years of CDTI [47]. In a formerly highly endemic area in the Mbam and Sanaga river valleys of Cameroon, Siewe Fodjo et al. [48] found an Ov16 seropositivity of 47% (68/145) in Bilomo (Centre Region) and 52% (13/25) in Kelleng (Littoral Region) among children aged 7–10 years after >13 years of CDTI. A factor likely involved in the persistence of onchocerciasis in this area is the very high (and perennial) biting rates of the blackfly vectors, as documented in other localities of the Mbam and Sanaga river systems, both historically and recently [49,50]. As our study area was originally mesoendemic, and annual biting rates in the region were of the order of 3,000–4,000 per person [21], it would be advisable to update entomological studies in the Tain District and Wenchi Municipality of Ghana to better understand the potential contribution of blackfly abundance, species composition and transmission intensity to persistent infection. A study in the Nkoranza North District (see Fig 1) documented highly seasonal transmission, with monthly biting rates of approximately 800 in the peak transmission season and 0–100 in other months of the year [51]. More recently, entomological studies have been conducted to better understand the role of Simulium vectors in sustaining onchocerciasis transmission close to our study areas [26]. Nditanchou and colleagues working in the Kwanware-Ottou transmission focus (within 5-10 km of our study communities) in Ghana found that the most productive blackfly breeding was within 5 km of Kwanware-Ottou. They also reported that blackfly daily biting rates were highest in Kwanware and Ottou, with 199 and 160 bites per day, respectively. Infection in blackflies was found only in Kwanware and Ottou, with infectivity rates of 5.9% (per 1000) and 6.7%, respectively [26]. These findings underscore the need for targeted vector control interventions in our study area and other transmission hotspots, using environmentally friendly approaches such as the “slash and clear” method [52].

In terms of diagnostic test acceptability, most participants preferred the Ov16 RDT to skin-snip microscopy and nodule palpation on account of its being less invasive, less painful, and more rapidly generating results that the participants could see in situ. These attributes of the RDT could be capitalised upon to increase the willingness of the population to participate in surveillance activities. Our findings are consistent with those of Dieye et al. [12] in south-eastern Senegal, who reported a greater willingness of participants to take part in onchocerciasis surveillance activities involving Ov16 RDT. Interestingly, the participation rates in Ov16 RDT (99.7%) and skin-snip microscopy (32.7%) in the Senegal study [12] are very similar to ours in Ghana (100% and 37%, respectively), suggesting that about a third of the population might be willing to participate in skin-snipping surveys. As endemic countries transit the path towards onchocerciasis EOT, it is increasingly important that the surveillance tools used in control and elimination programmes are both effective and acceptable. Residents of meso- and hyperendemic communities have typically had a long history of MDA treatment and are, therefore, increasingly apathetic towards control and surveillance activities, especially because the disease is no longer perceived as a major public health problem in some areas.

As mentioned, about a third of the individuals taking part in the ‘exit’ interviews remarked that although skin-snipping is painful, they would be willing to participate in future parasitological surveys using skin-snip microscopy. This was partly due to the research team members explaining to the participants the need for the test and the potential benefits to their health outcomes. These findings underscore the significant influence that surveillance staff can have on participation rates. They also highlight the importance of incorporating targeted education in NTD programmes to ensure that community members understand the rationale and value of the diagnostic tools used in surveillance, thereby fostering greater acceptance and participation.

The Ov16 RDT was also generally preferred among the technicians involved owing to its being relatively simple and easy to use, quick turnaround and potential for increasing community participation in future surveys, in agreement with the study in Senegal [12]. However, the technicians in our study were aware of the Ov16 RDT limitations concerning its inability to distinguish between past and present exposure/infection.

For a total of 400 individuals, the total cost per participant was estimated at US$24 for Ov16 RDT and US$74 for skin-snip microscopy, the former being approximately a third of the cost of the latter. The difference in cost between the two tests per individual (US$50) was mainly driven by differences in transport and lodging, supplies, devices and instruments, labour and training. Regarding training requirements, Ov16 RDT required a relatively short time (approximated 2 hours) to train nine technicians. By contrast, skin-snip microscopy training (sample-taking technique and microscopy) took 14 days. Also, supplies, devices and instruments needed for skin-snip microscopy include expensive instruments such as corneoscleral punches, microscopes, microtitre plates, pipettes etc., whereas those for Ov16 RDT principally include the kits from the manufacturer. However, the capital investment of a microscope and Holth corneoscleral punches may not need to be made for each subsequent survey, whereas consumables and Ov16 RDT cassettes would have to be. Although the cost of transport and lodging was considered jointly for the deployment of the two tests and allocated according to the proportion of costs for each specific test and their shared costs, it is important to note that if Ov16 RDT alone were used in typical surveillance, fewer days would likely be necessary, making the Ov16 RDT even less costly. Nevertheless, challenges with Ov16 RDT procurement must be considered when planning surveillance activities.

This study employed the Ov16 RDT using whole blood to obtain results in the field. According to the product insert, the Ov16 RDT demonstrates a sensitivity of 81.1% (95% CI: 70.7%–88.4%) and a specificity of 99.0% (95% CI: 94.8%–99.8%) when compared to skin-snip microscopy as the reference standard [10]. However, collecting dried blood spot (DBS) samples that are eluted in the lab and used in the RDT may reach a sensitivity similar to that of the Ov16 ELISA recommended by the WHO [48,49]. Obtaining high-quality DBS samples, storing and transporting them appropriately to the lab, obtaining eluates and testing these in the laboratory with the Ov16 RDT will certainly increase time and cost [53,54], but this might be an acceptable trade-off between improving the test sensitivity, mitigating the risk of missing infected individuals, and obtaining a reasonable acceptability amongst communities and technicians that should be further investigated. Conducting this work would also follow the recommendation made in [39] that further comparisons of the field performance of the Ov16 RDT be made with the laboratory performance of RDT using DBS, particularly in low-transmission settings (either because of these being hypoendemic—and treatment-naïve—or due to long-term treatment).

Conclusions

This study has demonstrated the usability of the Ov16 RDT to identify ongoing transmission using seroprevalence in children aged 5–9 years in an area with nearly three decades of CDTI, and more generally, exposure patterns in the general population. The test was also more acceptable and feasible than skin-snip microscopy, less costly to implement, and may help overcome the increasing reluctance of endemic communities to participate in onchocerciasis surveillance activities. Regardless of these advantageous features, further R&D efforts are necessary to increase the diagnostic performance of serological tests and improve their ability to distinguish between past and patent infection [56].

Supporting information

Acknowledgments

We thank all the staff and students of the Department of Biological Science and the Centre for Research in Applied Biology who provided support in one way or another to ensure the successful implementation of this research. We also appreciate the efforts of the directors and staff of the Wenchi Municipal Health Directorate, Nsawkaw District and the Subinso Health Centre for their enormous contributions towards the study. Our appreciation also goes to the chiefs and people of the study communities who partnered with us in this study. The map data used in this document is attributable to OpenStreetMap contributors and is licensed under the Open Database License (ODbL). More information can be found at https://www.openstreetmap.org/copyright.