Rift Valley fever seroprevalence and associated factors in non-vaccinated livestock (cattle and small ruminants) in Arua, Lira and Apac districts, Uganda

Abstract

Introduction

Rift Valley fever (RVF) is a zoonotic viral disease of major veterinary public health and economic importance, transmitted primarily by mosquitoes and associated with significant livestock losses and human infections. Despite its epidemic potential and its demonstrated endemicity in western Uganda, data on its seroprevalence and associated factors from northern Uganda remain limited. This study determined the seroprevalence of RVF and its associated factors in unvaccinated livestock in Apac, Arua, and Lira districts.

Methods and materials

A cross-sectional study was conducted in selected sub-counties of Apac, Lira, and Arua districts in northern Uganda. Blood samples were collected from 704 randomly selected unvaccinated livestock (cattle, goats, and sheep). Sera were tested for RVF IgG antibodies using a competitive ELISA (IDvet) kit. Data on animal and herd characteristics were collected through interviews of livestock owners using a structured questionnaire. A robust modified Poisson regression analysis was done to estimate prevalence ratios for the factors associated with RVF IgG seropositivity.

Results

RVF IgG seroprevalence was 28% (95% CI: 24.8–31.4). Seropositivity was highest in Arua (54.5%) compared to Apac (28.6%) and Lira (18.1%). Seropositivity was higher in cattle (31.3%) when compared with goats (23%). In multivariable analysis, cattle were more likely to be RVF seropositive than goats (aPR = 1.37, 95% CI: 1.06–1.76, p < 0.05), while indigenous breeds had significantly lower likelihood of seropositivity than cross breeds (aPR = 0.40, 95% CI: 0.29–0.56, p < 0.05). Animals with recent (in the past 8 weeks) illness had significantly higher likelihood for RVF seropositivity than animals without recent illness history (aPR = 3.66, 95% CI: 2.93–4.57, p < 0.05).

Conclusion

There is substantial RVF seropositivity among livestock in northern Uganda, suggesting possible RVFV circulation. Ecological factors, livestock species, breed types, and recent illness history are associated with RVF seropositivity, underscoring the need for strengthened surveillance and targeted control strategies.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12917-026-05853-6.

Introduction

Rift Valley fever (RVF) is a globally important zoonotic disease, posing substantial threats to livestock production, livelihoods and public health [39, 47]. The disease is caused by the Rift Valley fever virus (RVFV), a Bunyavirus, which is primarily transmitted through the bite of infected Aedes and Culex mosquitoes [15]. In addition to vector-borne transmission, humans can become infected through direct contact with infected animal tissues, including blood, meat, and particularly aborted fetal materials and placentas [15, 20, 26]. RVF is associated with severe economic consequences in settings where livestock accounts for a significant proportion of commerce, this is because it causes high rates of abortions and neonatal mortality [26]. Beyond the direct livestock processing losses, outbreaks often trigger restrictions on animal movement and trade, which further exacerbates economic hardship for livestock-dependent communities and disrupts the global livestock industry [11].

Since its first major outbreak in Kenya in 1931, multiple RVF outbreaks have been reported across Africa involving livestock species such as cattle, sheep, and goats, as well as humans [4, 7, 12, 14]. Sub-Saharan Africa has the highest number of reported RVF outbreaks globally, largely due to ecological conditions that favor mosquito vector proliferation such as heavy and seasonal rainfall, warm temperatures, high humidity, abundant water bodies, and dense vegetation cover [1, 22, 37]. Coupled with extensive and often poorly managed livestock production systems, these environmental conditions create an ideal setting for sustained RVFV transmission [8]. As a result, many countries in the region have classified RVF as a priority veterinary public health disease (WOAH 24. However, effective sero-surveillance is rarely conducted and in limited geographic areas; consequently, timely recognition of outbreaks is a major challenge for prevention and control efforts of RVF in the region [47].

In Uganda, limited diagnostic capacity at district and lower health system levels hampers early detection of RVF animal and human cases. Furthermore, most previous sero-epidemiological studies in Uganda have been concentrated in the western cattle corridor, where outbreaks have historically been reported [9, 19, 32]. The studies that included northern Uganda were country-wide sero-surveys without specific focus on the Districts of Apac, Arua, and Lira [2, 33, 44]. This geographical bias has resulted in knowledge gaps which has led to sub-optimal understanding of the national epidemiology of RVF, and hampered identification of silent inter-epidemic circulation of RVFV. Consequently evidence-based policies for prevention may not be fully informed. As an example, despite the World Organization for Animal Health’s (WOAH) recommendations for livestock vaccination as a primary option to prevent RVF outbreaks, routine vaccination is not comprehensively practiced in northern Uganda, potentially leaving livestock vulnerable, if, indeed, active transmission is ongoing [24]. However, even in areas where high seroprevalence and recurrent outbreaks have consistently been documented, routine RVF vaccination programs have not been widely implemented, suggesting that additional efforts are needed to translate serosurveillance findings into effective prevention policies [11].

Notably, northern Uganda, which serves as a geographical link between the western and eastern cattle corridors, remains understudied despite possessing ecological conditions conducive for RVF transmission [16]. The region is characterized by seasonal flooding, warm temperatures, relatively flat terrain, and socio-economic activities that support mosquito breeding and livestock-vector contact [36]. In addition, increasing pressure on grazing land in the western and central cattle corridors has contributed to livestock movement into northern Uganda, potentially increasing animal density and facilitating the spread of RVFV across regions [16]. Given ecological, epidemiological, and, ultimately, policy considerations, this study aimed to determine the seroprevalence of Rift Valley fever and identify factors associated with RVF exposure among unvaccinated livestock in Arua, Lira, and Apac districts, Uganda.

Materials and methods

Study design

A cross-sectional community-based study was conducted in non-vaccinated livestock (cattle and small ruminants) in 3 districts Apac, Lira and Arua of Northern Uganda in January 2025 for Apac and Lira and in January 2026 for Arua.

Study area

The key agricultural and ecological zones for this study were Apac and Lira districts located in the Lango Sub-region of Northern Uganda (Fig. 1). Apac is bordered by Kole, Oyam, Lira, Dokolo, Amolatar, Nakasongola, and Kiryandongo districts, with Apac Town situated about 62 km southwest of Lira City and 230 km north of Uganda’s capital (Kampala City). The district features flat to gently undulating terrain with extensive wetlands linked to Lake Kwania, part of the Lake Kyoga basins, which provides mosquito breeding habitats relevant to RVF transmission. Lira District, bordered by Pader, Agago, Otuke, Alebtong, Dokolo, Apac, and Kole, comprises savannah grasslands and seasonal wetlands under a tropical climate with distinct wet and dry seasons. In both districts, over 80% of the population practices mixed crop-livestock farming, producing cassava, maize, simsim, beans, and groundnuts, while rearing cattle, goats, and sheep.

Fig. 1.

Showing location of Apac, Lira and Arua districts, Northern Uganda

Arua District is located in the West Nile Sub-region of Northern Uganda (Fig. 1). It is a major administrative and commercial hub bordering Maracha, Terego, Yumbe, Koboko, and the Democratic Republic of Congo. Arua City lies approximately 480 km northwest of Kampala City and serves as a key cross-border trade center linking Uganda to Central Africa. The district is characterised by a mixture of gently rolling hills, savannah woodland, and seasonal wetlands, with drainage systems feeding into the River Nile basin. Although all three Districts are located in the Northern region, Arua is geographically located in a different sub-region than the other two districts, and is slightly different with relatively rough terrain ending the western East-African Rift Valley. We included Arua because the three districts share comparable agro-ecological and livestock production systems that favor RVF maintenance and transmission. The districts also experience extensive livestock movement, communal grazing, and active inter-district and transboundary animal trade with South Sudan and the Democratic Republic of Congo, factors known to increase the risk of disease introduction, circulation, and persistence. The population is predominantly rural, with over 75% engaged in subsistence and small-scale commercial agriculture. Common crops include maize, beans, cassava, groundnuts, and simsim, while livestock production focuses on cattle, goats, and poultry [17].

Map of the study area

Sample size determination

Kish Leslie formular [21] with designed effect was used. The average seroprevalence estimate in livestock used was from four studies in western Uganda [30, 32, 33, 44].

where:

• n = required sample size

• Z = standard normal deviate at 95% confidence level (1.96)

• p = estimated prevalence. (17.15 for cattle; 4.8 for goats and 5.55 for Sheep)

• D.E = Design effect (1.5)

• d = margin of error (0.05)

The calculated sample sizes (including a 10% non-response allowance) were 360 cattle, 116 goats, and 132 sheep; however, 320 cattle, 291 goats, and 93 sheep were ultimately sampled. The observed differences in species and district-specific sample sizes (Apac: 160 cattle, 126 goats, 29 sheep; Arua: 23 cattle, 48 goats, 30 sheep; and Lira: 137 cattle, 117 goats, 34 sheep) this was due to the livestock composition of the selected households and the availability of eligible animals at the time of sampling, with fewer sheep sampled because sheep ownership was relatively uncommon in the study area. Detailed district and species level analysis is presented in supplementary Table 1 and 2.

Data collection

Two data collection teams were set up and given specific locations (villages) to sample. A team was led by an epidemiologist from College of Veterinary Medicine, Animal Resources and Biosecurity (COVAB)/Makerere University and it included—laboratory scientists, enumerators and animal handlers.

Multistage sampling was used during the selection of the study participants First, a complete list of all villages in the study districts was obtained from the respective district administrative offices. Each village was assigned a unique identification number, and the numbers were written on identical pieces of paper, folded to conceal their identities, thoroughly mixed in a container, and randomly drawn one at a time without replacement. Within selected villages, target households were identified in collaboration with local veterinary officers, community leaders, and village chairpersons. Household lists of livestock keepers were compiled, and eligible households rearing non-vaccinated cattle, goats, or sheep were identified. Systematic random sampling was subsequently used to select households for inclusion in the study. Once a target household was identified, a unique household identification number was assigned. Data on the household and 1 sheep, 1 goat and 2 or 3 cattle to sample were collected using a structured questionnaire in KoboCollect (version 2025.3.3, Computer software Harvard Humanitarian Initiative. https://www.kobotoolbox.org/). The form also captured an animal identification number, age, sex, and species, vaccination status, among others. The questionnaire used to collect data in the study has been provided as supplementary file1. The study animals were privately owned by individual livestock farmers residing in the selected study communities. Prior to enrolment and sample collection, the purpose of the study was explained to the livestock owners, and written informed consent was obtained from each participating owner for the inclusion of their animals in the study. For owners who could not read and write, consenting was done in the presence of a witness.

Non-vaccinated status was established primarily through farmer self-report during questionnaire administration and corroborated through consultations with local veterinary officers regarding the absence of documented Rift Valley Fever (RVF) vaccination campaigns in the study areas during the study period. Livestock owners were specifically asked whether sampled animals had ever received RVF vaccination. In addition, no routine RVF vaccination program had been implemented in the selected districts at the time of study. Translators worked with the local leaders to capture data from respondents who were illiterate after appropriate consenting.

Sample collection

Blood samples were collected from each selected livestock for laboratory analyses. During the process, animals were restrained within their enclosure for ease of handling. Injection sites were then disinfected with 70% isopropyl alcohol before collecting ⁓ 6 ml of venous blood through jugular venipuncture or through the coccygeal vein of the cow. The samples were collected using non-heparinised vacutainer tubes to allow for blood clotting and collection of serum which was later centrifuged at 3000xg for 10 min to harvest serum. Collected blood sera were kept and transported at 4–8˚C to Uganda Virus Research Institute (UVRI) where they were barcoded before being stored in the sample storage freezer at −20˚C for further analyses.

Laboratory screening of serum samples

At the Uganda Virus Research Institute (UVRI), all stored serum samples were initially screened for Rift Valley Fever virus (RVFV) total antibodies using the ID Screen® Rift Valley Fever Competition Multispecies ELISA kit (https://www.innovative-diagnostics.com/), following the manufacturer’s instructions. The assay has demonstrated high diagnostic performance, with reported sensitivity ranging from 91–100% and specificity of approximately 100% based on manufacturer validation and independent ring trial evaluations [6]. Subsequently, all samples that tested positive for RVFV total antibodies were further screened using the ID Screen® RVF IgG Capture ELISA kit for animals to differentiate recent or active infections from past exposure.

Data analysis

Raw data collected via KoboCollect were downloaded into Microsoft Excel (version 2024). Data consistency checks were conducted to ensure logical responses (e.g., only households reporting animal keeping were required to specify the number of animals owned). Cleaned data were subsequently exported to STATA (Version 18, StataCorp LLC) for statistical analysis. The outcome variable, RVF IgG seropositivity, was generated from laboratory IgG test results and categorised as binary, where negative results were coded as 0 and positive results as 1. Explanatory variables included district, animal species, breed type, sex, animal age category, illness history in the previous two months, and treatment status. String variables were converted into numeric categorical variables using encoding procedures to facilitate statistical analysis.

Descriptive analysis was conducted to summarise the characteristics of the study population. Frequencies and percentages were computed for categorical variables, and 95% confidence intervals (95% CI) for proportions were estimated. Bivariate analysis was performed to assess associations between independent variables and RVF IgG seropositivity. Cross-tabulations were conducted using Pearson’s chi-square (χ²) test with row percentages to compare seropositivity across categories of explanatory variables. Because RVF IgG seropositivity represented a common binary outcome, prevalence ratios rather than odds ratios were estimated to avoid overestimation of associations. Unadjusted prevalence ratios (uPR) and their corresponding 95% confidence intervals were estimated for each explanatory variable using Poisson regression models with robust variance estimation. Variables that were associated with RVF IgG seropositivity at bivariate analysis (p < 0.2) were included in the multivariable model. Adjusted prevalence ratios (aPR) with 95% confidence intervals were estimated and an aPR greater than 1 indicated increased risk of RVF IgG seropositivity, an aPR less than 1 indicated decreased risk, while an aPR equal to 1 denoted no association. Statistical significance was considered at p < 0.05 throughout the analysis.

Results

Descriptive characteristics of the study population

A total of 704 livestock were included in the study (Table 1). The vast majority of animals were indigenous breeds (92.2%, 649/704); cross-breeds accounted for 7.8% (55/704). Females constituted 64.2% (452/704) of the study population. By species distribution, cattle formed the largest proportion (320, 45.5%), followed by goats (291, 41.3%), while sheep accounted for the smallest proportion (93, 13.2%). Most livestock were adults (73.4%, 517). By district, 315 (44.8%) animals were from Apac, 101 (14.6%) from Arua, and 288 (40.9%) from Lira. A history of illness (mostly nasal discharges) in the last two months was reported in 75 animals (19.3%), while 167 (23.7%) had ever received treatment during their lifetime, mainly tick acaricides and antibiotics for suspected infections or routine livestock management.

Table 1.

Descriptive characteristics of the study population (n = 704)

Variable	Category	Frequency (n)	Percentage (%)	95% CI
District	Apac	315	44.7	41.1–48.4
	Arua	101	14.4	11.9–17.1
	Lira	288	40.9	37.3–44.6
Species	Cattle	320	45.5	41.8–49.2
	Goat	291	41.3	37.7–45.0
	Sheep	93	13.2	11.0–15.9
Breed type	Indigenous	649	92.2	89.9–93.9
	Cross-breed	55	7.8	6.1–10.1
Sex	Female	452	64.2	60.6–67.7
	Male	252	35.8	32.3–39.4
Animal age	Young	48	6.8	5.2–8.9
	Juvenile	139	19.7	17.0–22.9
	Adult	517	73.4	70.0–76.6
Any illness in the last 2 months	No	568	80.7	77.6–83.4
	Yes	136	19.3	16.6–22.4
Animal ever treated for any illness	No	537	76.3	73.0–79.3
	Yes	167	23.7	20.6–27.1

Prevalence of RVF IgG seropositivity

The overall RVF IgG seropositivity was 28% (95% CI: 24.8 −31.4) Fig. 2.

Fig. 2.

The RVF IgG seropositivity among livestock in Northern Uganda

RVF IgG seroprevalence varied by district, species, and breed type. By district, Arua had the highest seroprevalence (54.5%) (Table 2). While seroprevalence was similar across species, sheep marginally had the highest seroprevalence (32.3%), closely followed by cattle (31.3%), whereas goats had the lowest prevalence (23.0%) (Table 2). These differences were also statistically significant.

Table 2.

Overall Seroprevalence stratified by district and species

Variable	Category	Negative n (%)	Positive n (%)
District	Apac	225 (71.4)	90 (28.6)
	Arua	46 (45.5)	55 (54.5)
	Lira	236 (81.9)	52 (18.1)
Species	Cattle	220 (68.8)	100 (31.3)
	Goat	224 (77.0)	67 (23.0)
	Sheep	63 (67.7)	30 (32.3)

Factors associated with RVF IgG seropositivity among overall livestock in study sites

Bivariate analysis

In the bivariate analysis, several host and management-related factors showed significant associations with RVF IgG seropositivity among livestock (Table 3). At the district level, livestock in Arua had higher odds of RVF IgG seropositivity compared with Apac (uPR = 1.24, 95% CI: 0.90–1.72), although this association was not statistically significant (p = 0.188). In contrast, livestock in Lira were significantly less likely to be seropositive compared with those in Apac (uPR = 0.63, 95% CI: 0.47–0.85; p = 0.003). At species level, cattle had higher odds of RVF IgG seropositivity compared with goats (uPR = 1.36, 95% CI: 1.04–1.77; p = 0.024), while sheep also showed increased seropositivity, although this was marginal (uPR = 1.40, 95% CI: 0.98–2.01; p = 0.068). Regarding breed type, indigenous animals were significantly less likely to be seropositive compared with cross-breeds (uPR = 0.31, 95% CI: 0.26–0.38; p < 0.001).

Table 3.

Bivariate analysis of factors associated with RVF IgG seropositivity among the overall livestock (n = 704)

Variable	Category	Neg n (%)	Pos n (%)	χ2	p-value	uPR (95% CI)	p-value
District	Apac (Ref)	225 (71.4)	90 (28.6)	49.26	< 0.001	1.00
	Arua	46 (45.5)	55 (54.5)	49.26	< 0.001	1.91 (1.48–2.45)	< 0.001*
	Lira	236 (81.9)	52 (18.1)	49.26	< 0.001	0.63 (0.47–0.85)	0.003*
Animal species	Goat (Ref)	220 (68.8)	100 (31.3)	6.09	0.048	1.00
	Cattle	224 (77.0)	67 (23.0)	6.09	0.048	1.36 (1.04–1.77)	0.024*
	Sheep	63 (67.7)	30 (32.3)	6.09	0.048	1.40 (0.98–2.01)	0.068
Breed type	Cross-breed (Ref)	13 (23.6)	42 (76.4)	69.30	< 0.001	1.00
	Indigenous	494 (76.1)	155 (23.9)	69.30	< 0.001	0.31 (0.26–0.38)	< 0.001*
Sex of animal	Female (Ref)	326 (72.1)	126 (27.9)	0.01	0.933	1.00
	Male	181 (71.8)	71 (28.2)	0.01	0.933	1.01 (0.79–1.29)	0.933
Animal age	Adult (Ref)	385 (74.5)	132 (25.5)	20.60	< 0.001	1.00
	Young	21 (43.8)	27 (56.25)	20.60	< 0.001	2.20 (1.68–2.94)	< 0.001*
	Juvenile	101 (72.7)	38 (27.34)	20.60	< 0.001	1.07 (0.79–1.461)	0.664
Any illness in last 2 months	No (Ref)	468 (82.39)	100 (17.61)	157.12	< 0.001	1.00
	Yes	39 (28.68)	97 (71.32)	157.12	< 0.001	4.05 (3.29–4.99)	< 0.001*
Animal treated for illness	No (Ref)	406 (75.61)	131 (24.39)	14.46	< 0.001	1.00
	Yes	101 (60.48)	66 (39.52)	14.46	< 0.001	1.62 (1.27–2.06)	< 0.001*

^* Significant at 95% C.I. (P < 0.05), uPR Unadjusted prevalence ratios, Neg Negative, Pos positive

Age was also associated with seropositivity, where young animals had higher odds compared with adults (uPR = 2.20, 95% CI: 1.68–2.94; p < 0.001), while juvenile animals showed no significant difference (uPR = 1.07, 95% CI: 0.79–1.46; p = 0.664). Sex was not significantly associated with RVF IgG seropositivity (uPR = 1.01, 95% CI: 0.79–1.29; p = 0.933). Livestock with a reported history of illness in the previous two months had markedly higher odds of seropositivity compared with those without illness (uPR = 4.05, 95% CI: 3.29–4.99; p < 0.001). Similarly, animals that had received treatment for illness also showed higher seropositivity (uPR = 1.62, 95% CI: 1.27–2.06; p < 0.001).

Multivariate analysis

After adjusting for potential confounding factors, several variables remained independently associated with RVF IgG seropositivity (Table 4). Livestock from Lira district was 0.69 times less likely to be RVF IgG seropositive compared to livestock in Apac and this association was statistically significant (aPR = 0.69, 95% CI: 0.52–0.91, p = 0.009). Meanwhile livestock from Arua district were 1.24 times more likely to be RVF IgG seropositive compared to livestock in Apac (aPR = 1.24, 95% CI: 0.90–1.72, p = 0.188).

Table 4.

Multivariate analysis of factors associated with RVF IgG seropositivity among the overall livestock (n = 704)

Variable	Category	aPR (95% CI)	p-value
District	Apac (Ref)	1.00
	Arua	1.24 (0.90–1.72)	0.188
	Lira	0.69 (0.52–0.91)	0.009*
Animal species	Goat (Ref)	1.00
	Cattle	1.37 (1.06–1.76)	0.014*
	Sheep	1.37 (0.96–1.96)	0.082
Breed type	Cross-breed (Ref)	1.00
	Indigenous	0.40 (0.29–0.56)	< 0.001*
Animal age	Adult (Ref)	1.00
	Young	1.25 (0.95–1.65)	0.116
	Juvenile	0.82 (0.57–1.18)	0.281
Any illness in last 2 months	No (Ref)	1.00
	Yes	3.66 (2.93–4.57)	< 0.001*
Animal treated for illness	No (Ref)	1.00
	Yes	1.22 (0.97–1.53)	0.091

^*Significant at 95% C.I. (P < 0.05), aPR Adjusted prevalence ratios

Cattle were 1.37 times more likely to be RVF IgG seropositive when compared with goats (aPR = 1.37, 95% CI: 1.06–1.76, p = 0.014). Indigenous animal breeds were 0.40 times less likely to be RVF IgG seropositive compared with cross-breeds (aPR = 0.40, 95% CI: 0.29–0.56, p < 0.001). Animals that had experienced illness within the previous two months were 3.66 times more likely to be RVF IgG seropositive compared with animals without reported illness (aPR = 3.66, 95% CI: 2.93–4.57, p < 0.001). Animal age and obtaining treatment for an illness in the 2 months preceding data collection were not significantly associated with RVF IgG seropositivity after adjustment. The association analysis was further stratified by district and species (supplementary Table 1 and 2).

Discussion

This study investigated the seroprevalence of RVF and its associated factors in livestock in the districts of Apac, Arua, and Lira in northern Uganda. The findings revealed that the overall RVF seroprevalence was 28% suggesting substantial RVF exposure of animals in this region, raising concerns about human risks, as well. The observed level of seropositivity is higher than the national livestock seroprevalence of 15.3% reported in Uganda [44] and also higher than the 7% reported in Kisoro District, southwestern Uganda [5]. However, it is slightly lower than the 33.8% reported in southwestern Uganda [34] and similar to the 33.3% reported in Garissa County, Kenya [28]. While these differences suggest spatial heterogeneity in RVFV exposure influenced by ecological and management conditions, it confirms that RVF is animal health consideration in this region where this information was previously not well characterized. The observed differences in seroprevalence between this study and previous studies conducted within Uganda and elsewhere may be attributed to variations in ecological, climatic, behavioral and husbandry factors across study locations [23, 29, 43, 45]. Northern Uganda experiences seasonal flooding, extensive wetlands, and favorable environmental conditions for mosquito breeding, which promote RVFV transmission among livestock populations [25, 27, 31]. In addition, differences in livestock management practices such as communal grazing, unrestricted animal movement, and shared watering points could further facilitate exposure to infected mosquito vectors [44].

The elevated seroprevalence observed in Arua relative to Lira and Apac may be attributed to localized ecological and demographic factors that optimize vector habitats and arboviral transmission. Geostatistical modeling identifies northwestern Uganda as a primary hotspot for high RVFV seroprevalence, with peak risks stretching directly from Lake Albert up to the South Sudan border, an area that includes Arua District [42] This regional vulnerability is ecologically driven by high variability in vegetation greenness and an elevated topographic wetness, which reflects the vein-like network of rivers and wetlands prone to water pooling and seasonal flooding [40, 42]. In Arua, these conditions are worsened by proximity to the Nile riverbanks and favorable rainfall patterns, creating ideal breeding sites for floodwater Aedes and Culex mosquitoes [31, 46]. Furthermore, this ecologically receptive landscape is coupled with high human population growth and unmonitored cross-border livestock movement from neighboring territories, significantly increasing the probability of RVFV introduction and exposure compared to the more interior districts of Lira and Apac [42].

The observed lower seroprevalence observed in Lira than in Apac could be due to the livestock production systems in Lira district, which are increasingly characterized by smaller herd sizes and more spatially separated and indoor animal keeping practices associated with urban expansion and changing land use patterns. Such conditions may reduce contact between susceptible livestock and mosquito vectors, thereby limiting opportunities for RVF virus transmission and circulation among animals. Furthermore, there has been continuous influx of cattle especially from the southwestern and central cattle corridors of Uganda into Apac district more than in Lira [16]. This movement of animals may have not only increased the opportunities for vector-host contact and for sustained RVF virus circulation within livestock populations [29, 48], but also introduced infected or previously exposed animals into the district, potentially contributing to the higher RVF seroprevalence observed in Apac.

At species level, seroprevalence patterns in this study differed from findings reported in the Uganda national livestock survey Luke Nyakarahuka et al. [33] which recorded an overall seropositivity of 6.9%, with strong species-level variation; cattle (10.7%) had higher seroprevalence than goats (2.6%) and sheep (2.0%). In contrast, the present study found substantially higher species-specific seroprevalence among all three species. This discrepancy may reflect differences in ecological and epidemiological settings, as we specifically focused on a high-risk area of northern Uganda characterized by seasonal flooding and favorable mosquito breeding conditions. In addition, differences in sampling structure and distribution across species may also contribute to the observed variation, where smaller or uneven subgroup sample sizes can amplify apparent prevalence differences between species compared to large-scale national surveys.

This pattern suggests that all three species are actively involved in RVFV transmission cycles in the study area, suggesting active and/or previous circulation of the virus in a region where RVF transmission has not previously been characterized. The finding underscores the need for strengthened sero-surveillance systems and the implementation of One Health approaches, integrating animal, human, and environmental health sectors to mitigate the risk of future RVF outbreaks. However, these comparisons should be interpreted cautiously because the national study used a much larger sample and different spatial coverage, whereas the current study focused on a high-risk ecological hotspot in northern Uganda.

This study also found that indigenous livestock breeds were significantly less likely to be RVF seropositive compared to crossbred animals. It is possible that indigenous African livestock breeds possess unique adaptive traits and genetic composition through prolonged exposure to endemic environmental and infectious pressures, which may contribute to enhanced resilience or altered susceptibility to vector-borne infections compared to exotic or crossbred animals [28, 35]. However, further studies are required to determine whether the observed differences in RVF seroprevalence between the different livestock breeds are attributable to genetic resistance, immunological adaptation, or ecological and management-related factors.

This study also revealed that animals reported to have suffered illness within the previous two months were significantly more likely to be RVF seropositive. This finding may indicate that some of the recent illness episodes were associated with RVF infection. IgG antibodies are detectable within 8 days of RVF infection. Since the assay used in this study detected RVF-specific IgG antibodies, seropositivity cannot differentiate between active or previous infection, as IgG antibodies are known to persist for extended periods following infection [3, 38]. Animals experiencing recent illness may have undergone recent RVF virus exposure sufficient to stimulate detectable antibody production [18]. However, interpretation of this finding should be approached cautiously because of the high concentration of seropositive animals that reported to have been ill recently. This clustering may have amplified the magnitude of the observed association and suggest further studies among recently ill animals are needed [10]. Nonetheless, the observed association may suggests that recent clinical illness among livestock within endemic settings may serve as a useful indicator for targeted RVF surveillance and warrants further investigation using molecular diagnostics capable of detecting active viral infection, such as reverse transcription polymerase chain reaction (RT-PCR) assays [13, 41].

Study limitations

The cross-sectional study design limited the ability to establish temporal or causal relationships between identified risk factors and RVF seropositivity. Since we only tested for IgG antibodies, it was not possible to determine the timing of infection or ongoing viral circulation. To mitigate this limitation, future studies should adopt longitudinal cohort designs combined with molecular diagnostic methods such as RT-PCR to detect active RVF infections and monitor transmission trends over time. Given that our data collection for Arua district was done January 2026 while the other 2 sites were done in January 2025, the variations in weather (eg. Rainfall distribution and temperature) could have affected the virus circulation. However, to minimize variations, careful attention was paid to the month of data collection, that is January with an assumption that the climatic conditions of Arua in January 2025 could be similar to that of January 2026. Furthermore, some animal and household information, including illness history, treatment history, and management practices, relied on self-reported responses from livestock owners, which may have introduced recall bias and misclassification. The study minimized this by using trained research assistants and standardized KoboCollect tools to improve consistency and accuracy of data collection. Finally, the study focused on selected sub-counties within three districts and may not fully represent the RVF epidemiological situation across northern Uganda. Expanding future surveillance to additional districts and ecological zones would improve the generalizability of findings.

Conclusion

This study demonstrated a high seroprevalence of RVF in non-vaccinated livestock in Apac, Arua, and Lira districts of northern Uganda, indicating substantial natural exposure and possible endemic circulation of Rift Valley fever virus (RVFV) in the region. Geographic location, livestock species, livestock breed, and recent illness history were significantly associated with RVF seropositivity. Specifically, livestock from Apac, cattle species, Cross breeds and having recent illness were associated with higher likelihood of RVF seropositivity. Higher seropositivity in Apac district highlights the importance of ecological suitability and livestock density in sustaining viral circulation, while the lower seropositivity among indigenous breeds may indicate adaptive resilience or differences in exposure patterns and management patterns feeding programs. The association between recent illness and RVF seropositivity further suggests the need for enhanced surveillance of clinically ill animals in endemic settings. The study underscores the continuing public health and economic threat posed by RVF in northern Uganda and emphasises the importance of targeted surveillance and prevention strategies.

Recommendations

Based on the study findings, recommendations include strengthening district-specific RVF surveillance systems, particularly in ecologically high-risk districts such as Apac, through routine serological and molecular monitoring of livestock. A One Health approach integrating veterinary, human health, and environmental sectors should be enhanced to improve early detection and coordinated outbreak response. Vector control interventions targeting mosquito breeding sites around wetlands and grazing areas should be prioritised. Livestock movement, monitoring and community sensitization on RVF prevention should also be strengthened. Further longitudinal, molecular, and immunological studies are recommended to determine active viral circulation and assess the role of indigenous breed resilience. Given the limited or near absence of routine RVF vaccination in livestock within the study area, this region may also provide a valuable population for assessing the effectiveness and modeled economic utility of immunization strategies using existing RVF vaccine formulations, as well as a suitable site for evaluating candidate vaccines under development.

Abbreviations

aPR

Adjusted Prevalence Ratio

CI

Confidence Interval

COVAB

College of Veterinary Medicine, Animal Resources and Biosecurity

ELISA

Enzyme-Linked Immunosorbent Assay

FAO

Food and Agriculture Organization

GPS

Global Positioning System

IgG

Immunoglobulin G

RT-PCR

Reverse Transcription Polymerase Chain Reaction

RVF

Rift Valley Fever

RVFV

Rift Valley Fever Virus

STATA

Statistical Software for Data Science and Analysis

uPR

Unadjusted Prevalence Ratio

UVRI

Uganda Virus Research Institute

WOAH

World Organization for Animal Health

Authors’ contributions

JO conceived and designed the study; coordinated field activities; participated in data collection, data analysis, and interpretation of findings; and drafted the manuscript. BA, CK, LN, and RFB contributed to study conceptualization, supervision, interpretation of findings, and manuscript review. IOO, SO, IO, RO, BN, CN, AT, WO, FOL, WA, LO, OS, PO, DNO, and OA participated in field coordination, sample collection, laboratory analysis, data management, and manuscript review. MM contributed to the interpretation of findings and critical revision of the manuscript. CK and RFB contributed equally to this work and are co-senior authors. All authors read and approved the final manuscript.

Funding

This study was supported through institutional and collaborative research support from Makerere University. The funders had no role in the design of the study, data collection, analysis, interpretation of findings, manuscript preparation, or the decision to publish.

Data availability

The datasets generated and/or analysed during the current study are not publicly available due to confidentiality agreements and protection of participant information but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study obtained approval from the Uganda National Council for Science and Technology (UNCST) (HS6654ES). Ethical approval was also granted by the Makerere University School of Health Sciences Research Ethics Committee (MAKSHSREC 2024–635). Animal sampling activities were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the College of Veterinary Medicine, Animal Resources and Biosecurity, Makerere University (SVAR-IACUC/117/2024). All procedures involving human participants were conducted in accordance with the ethical principles of the Declaration of Helsinki and relevant national guidelines for conducting health research in Uganda. Participation was entirely voluntary, and only livestock owners willing to participate were enrolled in the study. Written informed consent was obtained from all livestock owners before the enrolment of their animals in the study and before any sample collection procedures were undertaken. For livestock owners who were unable to read or write, the consent process was conducted in a local language they understood with the assistance of trained research assistant personnel. All animal handling and sampling procedures were carried out in accordance with approved institutional animal welfare guidelines and IACUC protocols to minimize animal stress, pain, and discomfort. Research assistants were appropriately trained in biosafety and biosecurity procedures, and personal protective equipment was used during sample collection and laboratory analyses. Confidentiality and anonymity of all participants and study data were maintained throughout the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.