Tropical rainforests are rapidly disappearing, with serious consequences for birds, ecosystems, and the emergence of infectious diseases. Avian haemosporidians, blood parasites in birds, are often used as models to study human parasites due to their similarities in life cycles [1] - [3] . These parasites provide important insights into evolutionary and ecological processes. Protozoan blood parasites (or haemoparasites) have been reported in African rainforest birds [1] [4] [5] but there is limited information on how bird traits, such as ant-following behavior and morphology, affect parasite prevalence, particularly in forests undergoing deforestation.

Protozoan blood parasites belong primarily to the order Haemosporida. Plasmodium species are primarily transmitted by mosquitoes (family Culicidae) [6] . Haemoproteus species are vectored by biting insects including mosquitoes, biting midges (Culicoides), louse flies (Hippoboscidae), and horse flies (Tabanidae). Leucocytozoon species are transmitted by black flies (Simuliidae) [7] . Microfilariae (larval stages of filarial nematodes) are transmitted by blood-feeding arthropods—mainly Dipterans such as Simulium and Culicoides, but also other bloodsucking insects [8] [9] . At least 16 genera of filariae are known to infect birds [10] . Avian trypanosomes are also transmitted by a wide variety of arthropods (Simuliidae, Culicidae, Ceratopogonidae, Hippoboscidae, and Dermanyssidae) [11] - [14] . Compared to their mammalian counterparts, avian trypanosomes are poorly studied, especially in cases of co-infection with other avian haemoparasites [5] [15] . Despite numerous filarial species found in birds, little is known about their prevalence in ant-following birds.

Ant-following birds, common in tropical ecosystems, exhibit a unique foraging strategy: trailing army ant swarms to catch prey fleeing from the forest floor [16] . While this behavior offers feeding advantages, it may also increase exposure to parasites. These birds are among the most frequent and dominant swarm attendant [17] . Ant-following specialization is classified into three categories: occasional, regular, and professional [18] . Professional followers rely almost exclusively on a single ant species, Eciton burchellii, making them particularly vulnerable to environmental threats such as deforestation and climate change. In Western Kenya, Peters and Okalo [19] , found that the abundance of army ants (Dorylus wilverthi and D. molestus) was positively correlated with the occurrence of specialized ant-followers (Neocossyphus poensis, Alethe poliocephala, and Bleda syndactyla). Craig [20] , documented 52 regular ant-following bird species in Africa—38 of which belong to three overrepresented families: Muscicapidae (18 spp.), Pycnonotidae (13 spp.), and Turdidae (7 spp.). Because birds have shorter life spans than humans, they are ideal for studying disease dynamics across lifetimes and generations, helping us better understand host-vector-environment interactions.

Habitat disturbance and conversion to plantations reduce the diversity of birds, ants, and haemosporidian parasites. Forest fragmentation often leads to declines in army ant populations, which in turn impacts the bird species that depend on them [21] . These cascading effects influence parasite prevalence. Ant diversity generally declines with disturbance [22] [23] , likely due to changes in vegetation and microclimate. Forests converted into mixed-crop fields also show lower ant diversity [24] . Southern Cameroonian forests had significantly higher ant diversity compared to fallow or crop fields. Long-term Amazonian studies have shown that ant-following bird communities remain stable in undisturbed forests [25] . In Cameroon, forest degradation has led to sharp declines in insectivorous birds—especially ant-followers—and delayed recolonization of regenerating forests [16] [19] [26] - [29] . Ocampo-Ariza et al. [28] reported significantly lower ant-following bird richness and encounter rates in oil palm plantations than in protected or agroforestry landscapes. However, no study has examined how parasite prevalence in these birds correlates with body mass.

Research on bird haemosporidians in Africa has focused largely on Eurillas latirostris and Cyanomitra olivacea [30] - [32] , while much less is known about ant-following birds in disturbed habitats. Bonneaud et al. [30] reported higher Plasmodium prevalence in pristine forests than in disturbed ones. Similarly, Chasar et al. [31] found higher Haemoproteus and Leucocytozoon prevalence in undisturbed areas. Tchoumbou et al. [4] noted that parasite prevalence decreased with deforestation, but Haemoproteus slightly rebounded following the establishment of palm oil plantations, while Plasmodium continued to decline. There remains a clear knowledge gap regarding protozoan parasite prevalence in ant-following birds in forests undergoing active deforestation.

In Cameroon, mature forests are shrinking due to logging and slash-and-burn agriculture, while secondary forest cover is expanding [33] . The Talangaye rainforest in the South West region is undergoing rapid deforestation from palm oil development, making it emblematic of the country’s broader deforestation crisis. This loss of forest not only threatens bird biodiversity but may also alter parasite dynamics. While ant-following birds have been well studied in the Neotropics [34] - [38] , little is known about their parasite dynamics in African tropical forests undergoing active deforestation.

This study focuses on five groups of avian blood parasites, Plasmodium, Haemoproteus, Leucocytozoon, trypanosomes, and microfilariae, in ant-following birds. These parasites have important effects on host behavior, fitness, and survival. We hypothesize that:

1) Parasite prevalence will differ between pristine and fragmented forests in ant-following birds.

2) Ant-following behavior will influence the prevalence of parasites.

3) Host body mass will influence the likelihood of parasite infection.

Prevalence was calculated for each parasite across forest types. Chi-square tests assessed differences in infection prevalence between habitats. T-tests compared bird weights of infected vs. non-infected individuals. Logistic regression models assessed the influence of forest type, ant-following behavior and body weight, on infection probability. All analyses were conducted using R, with significance set at p < 0.05.

In this study, we examined the prevalence of protozoan blood parasites in the three most common ant-following bird families in Afrotropical forests. Overall, birds from pristine forests exhibited a higher prevalence of protozoan parasites, particularly Plasmodium and Leucocytozoon, except in the case of Haemoproteus, Trypanosoma, and microfilariae. The high prevalence of Plasmodium and Leucocytozoon in undisturbed forests is consistent with previous findings from Cameroon [30] [31] . Interestingly, Trypanosoma prevalence was significantly higher in fragmented forests (p = 0.0235) among ant-following birds, whereas the other parasites did not show significant differences across forest types. Bonneaud et al. [30] similarly reported that forest type was significantly associated with variation in Plasmodium prevalence in three non ant-following bird species (Andropadus latirostris, Andropadus virens, and Cyanomitra obscura). Our study identified 16 species of regular ant-following birds: 4 in the family Muscicapidae, 7 in Pycnonotidae, and 2 in Turdidae. This represents 42% of the 38 regular ant-followers identified by Craig [20] , who documented 18 Muscicapidae, 13 Pycnonotidae, and 7 Turdidae species. Notably, members of the Turdidae family have shown haemosporidian prevalence as high as 94% in temperate regions such as Germany [45] .

Ant-following birds are insectivorous species that rely on swarms of army ants, particularly driver ants of the subfamily Dorylinae in tropical Africa, to flush arthropods and small vertebrates from the forest floor [16] [46] . Among African passerines, these species are considered especially vulnerable to forest loss and fragmentation [16] [26] . Because many of the flushed arthropods are vectors of protozoan parasites, this foraging strategy likely increases parasite exposure [6] - [9] .

Five groups of protozoan blood parasites were detected in this study, with an overall infection prevalence of 47%. All five parasite types have previously been recorded in Afrotropical birds [1] [4] [5] [46] . Specifically, 47% of ant-following birds tested positive for at least one parasite, compared to only 29% in non-ant-following birds. This difference was statistically significant (p < 0.01), supporting earlier suggestions that ant-following behavior increases parasite risk [40] .

Previous studies have reported even higher parasite prevalence in Afrotropical birds (68.8% - 82.8%) [47] . However, Chaisi et al. [47] relied on qPCR and nested PCR rather than microscopy, which can detect gametocytes and distinguish active from latent infections. In contrast, our study used nested PCR only on samples that tested positive via microscopy. Among the 16 bird species sampled, 14 tested positive for protozoan parasites. Only two species, Cossypha cyanocampter and Eurillas curvirostris, showed no infections, but as only one individual was sampled per species, no conclusions can be drawn regarding their parasite status. Larger sample sizes will be necessary to determine whether these species are truly free of protozoan infections.

Among ant-following species with a sample size ≥ 15, the most parasite-rich species were Chamaetylas poliocephala, Neocossyphus poensis, Phyllastrephus xavieri, and Alethe castanea. This threshold was used to minimize bias due to differing sample sizes across host species [48] .

The high prevalence of microfilariae in Alethe castanea is consistent with previous studies [1] [40] . This ground-dwelling Muscicapid occurs in all forest types. Sehgal et al. [1] reported a microfilariae prevalence of 61.5% in A. castanea, with none detected in its close relative, Chamaetylas poliocephala (brown-chested alethe). Waltert et al. [49] also noted that A. castanea was the most frequent and consistent ant-follower at their study site. Given the high prevalence of both filariae and trypanosomes in this species, we recommend it as a candidate model for drug testing and further research on filarial and trypanosomal infections. Additional studies are needed to identify the specific filarial species involved and to better understand the origin and transmission of nematode infections in A. castanea.

High prevalence of Leucocytozoon and Haemoproteus was also found in Neocossyphus poensis, possibly due to its status as the most specialized African ant-follower in Kenya’s Kakamega Forest [19] . It is important to note that ant-following specialization varies: occasional ant-followers feed opportunistically when swarms cross their path, while obligate followers rely heavily on this strategy for survival [18] [50] .

We also observed generally low parasitemia levels. This may be due to the use of mist nets, which are more likely to capture birds in the chronic phase of infection, while acutely infected individuals may be less active and less likely to be sampled [6] . However, parasitemia is not necessarily indicative of disease severity, as even low-level infections can disrupt blood cell production and impair fitness [51] . The observed correlation between body mass and parasite load may reflect greater blood volume or immunosuppression in larger birds, or simply longer exposure due to age.

In summary, our findings provide strong evidence that ant-following behavior significantly increases the risk of protozoan infection in Afrotropical birds. These birds range widely across forest landscapes and forage near the ground, increasing their exposure to blood-feeding vectors such as mosquitoes (Culicidae) and biting midges (Ceratopogonidae) [46] [50] . Further research on these understudied species is crucial to better understand host-vector-parasite interactions in tropical forests undergoing rapid environmental change.

Forest fragmentation is associated with increased Trypanosome infection in ant-following birds, while heavier birds are more susceptible to several parasite types. Forest type was a significant predictor only for Trypanosomes (p = 0.0125). There were statistically significant differences in the prevalence of four parasites (Microfilariae, Leucocytozoon, Haemoprotues and Plasmodium) between ant-following and non ant-following birds. Higher bird weight significantly increased the odds of infection with several parasites, especially Haemoproteus (p = 0.0009) and Leucocytozoon (p = 0.0022) and Microfilariae (p = 0.0036). Ant-following birds have higher parasitemia than non-ant followers. These results highlight the need to integrate parasitological surveillance into avian conservation strategies.

The authors are immensely grateful for the international collaboration of Prof. Kevin Njabo which facilitated the completion of this work. We are grateful for authorization and field support in the forest granted by Dr. Blessed Okole and Mr. Akumsi Alfred. The authors are also immensely grateful to Mr. Forzi Francis (ornithologist) for his professional help during fieldwork as well as all the assistance of students and field guides in the PEER project 4-360.

This study received financial support from U.S. Agency for International Development through Partnerships for Enhanced Engagement in Research (project 4-360) awarded to Prof. Anong Damian Nota of the University of Buea, Cameroon and also from a Conservation Action Research Network, USA grant and Idea Wild Equipment, USA grant awarded to Dr. Malange Nanyongo.