The Epidemiology of Bovine Viral Diarrhea Virus in Low- and Middle-Income Countries: A Systematic Review and Meta-Analysis

Introduction: Bovine viral diarrhea virus (BVDV) causes reproductive inefficiencies and negatively impacts the economy of low- and middle-income countries (LMICs). It is characterized by a combination of syndromes that result in poor production performance and calf morbidity and mortality. BVDV control is possible by introduction of biosecurity measures, test-and-cull, and vaccination programs as accomplished in high-income countries. Knowledge of BVDV epidemiology is limited in many LMICs, which hinders implementation of effective control programs. We carried out a systematic review and meta-analysis to estimate the burden of BVDV, identify risk factors related to its occurrence, and health and economic impacts on production systems.

Materials and Methods: Relevant BVD articles were collated from library databases; 690 abstracts and full texts were found in an initial search followed by filtering of 59 manuscripts. We accounted for quality and risk of bias in the meta-analysis. Prevalence, exposure, and current infection at regional, production, and farming system levels were estimated using logistic random-effects meta-regression models. Finally, we calculated the proportion of studies that addressed risk factors and health and economic impacts across different production systems to inform future preventative strategies in LMICs.

Results: Seroprevalence was high and varied between regions. Mean weighted prevalence was 39.5% (95% CI 25–56.1), 45.2% (95% CI 35.9–54.8), 49.9% (95% CI 25.5–74.3), and 21.6% (95% CI 0.5–56) for sub-Saharan Africa, South America, Middle East, and Asia, respectively. Seroprevalence varied across farming systems, with smallholder farming showing the highest values. Herdsize was the most frequently reported risk factor, and the percentage of articles that reported herdsize as a risk factor were 20.6%, 33.3%, and 38.4% for dairy, beef and mixed systems respectively. Abortion (13.7% of articles) was the main reported health impact in dairy systems. Some articles reported milk drop (4.6% of articles), but no article investigated the economic cost of BVDV in farming systems.

Conclusion: Animal-level seroprevalence varied across all regions. Most of the studies focused on BVDV seroprevalence. There were some articles that investigated risk factors and health impacts, and there were even less that investigated economic impacts. Future studies should focus on identifying risk factors and quantifying health and economic impacts across systems. Understanding these aspects is crucial to develop management strategies to apply across diverse production systems in LMICs.

Introduction

Bovine viral diarrhea virus (BVDV) causes a pathogenic infection in livestock, mainly cattle and wild ruminants, with a global distribution (1, 2) resulting in abortion, calf mortality, and poor reproductive performance. Livestock production and keeping play a significant role in poverty and hunger alleviation in many countries around the world (3). Livestock in LMICs provide an important food source, drought power, and manure source, as well as regular monetary income. Current meat and milk demand exceeds livestock production for growing populations in LMICs (4–6). BVDV impacts negatively on animal production and population livelihood (6). Unlike other major infectious diseases, such as foot-and-mouth disease or brucellosis, there have been few efforts to understand BVD in LMIC settings and develop pathways for control or eradication (7).

BVDV is a member of the Flaviviridae family in the genus Pestiviruses and has 2 main types reported, BVDV-1 and BVDV-2. Recently, there has been a BVDV-3 or HoBi-like virus found in several regions (1, 8); however, this has not yet been reported in sub-Saharan African (SSA) countries (1, 9). BVD-1 and BVD-2 have two subtypes referred to as the cytopathic or cell-killing form and the non-cytopathic form which infects cells but does not cause cell death (10). Both of the subtypes however, still cause disease in animals leading to reproductive and economic losses. Specifically, losses, such as abortion, poor reproductive performance, low milk yield, and high calf morbidity and mortality have been quantified in high-income countries (11) but poorly monitored in LMICs. BVD is widespread, and prevalence varies across regions, i.e., in sub-Saharan Africa, antibody prevalence is between 51 and 77% (12–14), and antigen prevalence is up to 19%, while in Europe, antibody prevalence is about 46% and antigen prevalence is about 0.2% (15). The variations in prevalence are influenced by a variety of factors, including farming systems and options for control in different areas.

Transmission of the disease is mainly driven by presence of persistently infected (PI) animals in a herd or contact with them. PI animals are infected in utero during gestation, usually between days 18 and 125 (16), such that they do not develop an immune response to the virus but become persistently viremic and then shed the virus after birth (1, 10). Animals can also be transiently infected (TI) for a short period (usually between 2 and 3 weeks) and can shed the virus for a short time before they mount an immune response and then clear the infection (10, 17). Presence of infection is usually maintained in a herd because of ongoing production of PI animals and then shedding large amounts of the virus infecting naïve animals. Clinical outcomes of BVD depend on the host, stage of pregnancy, and strain of the infecting BVDV. Infection usually results in signs, such as transient viremia associated with leukopenia, thrombocytopenia, cell death in the thymus, pyrexia, and diarrhea leading to immunosuppression, which allows for co-infections such as respiratory pathogens plus fetal loss, including abortion, and fetal abnormalities in pregnant animals (8, 17, 18). Lack of pathognomonic clinical signs and the more chronic nature of the disease make it harder to diagnose in farms. Serum antibody diagnostic tests such as enzyme-linked immunosorbent assays (ELISAs) can be conducted to detect animals exposed after birth, but PI animals exposed in utero (between days 18 and 125 of gestation) do not produce antibodies and can only be detected by identifying the virus using either antigen ELISAs or molecular tools such as polymerase chain reactions (PCRs) (84, 85). Many high-income countries such as the United Kingdom have eradication programs using combinations of ELISAs and PCR to detect and remove PI animals from infected herds, vaccination (to prevent infection during pregnancy), test-and-cull, and implementation of biosecurity measures (9–11, 19). The diagnostic tests available for testing BVD-specific antibody and antigen are generally very reliable (20) and reported to have excellent sensitivity and specificity (10, 21–23). The accuracy of diagnostic tests helps give a good understanding of the epidemiology of BVD. Diagnostic tests are also essential components of an eradication program, and their accuracy has made BVD control programs economically achievable (19).

The potential impact of BVDV infection is well-recognized in high-income countries where its epidemiology is well-understood and its economic impacts are well-described and quantified. For example, in North American systems, it has been estimated that infection with BVDV can cost up to US$88/animal in beef herds due to calf mortality which leads to considerable financial losses for individual farmers (2, 24). Acute infections have been shown to cause up to a 23% milk drop in the 2 weeks following infection in United Kingdom dairy herds (25). The global distribution of BVDV (26) and its reported associated reproductive losses highlight why the virus has now been listed as a class B disease by the Office International des Epizooties (OIE) (27), and a reportable disease in cattle only and not in multiple species (1). There is, however, a gap in our understanding of the epidemiology of BVDV and its health and economic impacts on LMICs that need to be addressed urgently to reduce inefficiencies in livestock production that lead to antibiotic misuse and contribute to climate change, and to improve animal welfare. There are a few studies that have been conducted to determine the seroprevalence of BVDV antibodies, and it was our goal to bring together these studies in a systematic review in an attempt to understand the epidemiology and quantify the importance of risk factors and economic impacts of BVDV on cattle-rearing communities in LMICs.

Systematic reviews are an important tools for collating data from multiple published studies to improve parameter estimates such as prevalence and to identify research gaps (28, 29) and may be an important resource to policymakers when considering the design of disease control programs. Meta-analysis helps to give a precise estimate of the overall or combined effects of studies (30). This systematic review and meta-analysis of BVDV epidemiology in LMICs aims to (1) summarize the available data on the prevalence of BVDV, (2) describe the common risk factors and BVDV's health and economic impacts on LMICs, and (3) identify and collate gaps in knowledge for future research.

Materials and Methods

This systematic review followed the guidelines and checklist of the PRISMA Group (PRISMA Transparent Reporting of Systematic Reviews and Meta-Analysis) (31). Searches were conducted by the first author in December 2020 and updated in January 2022. The databases used were CAB Abstracts, Embase, Medline, PubMed, Scopus, Web of Science, Global Health, and the gray literature indexed in ProQuest. Search terms were adapted for the different databases.

Search Strategy

The search strategy for CAB Abstracts, Embase, Medline, and Global health consisted of search terms and subject headings related to BVD in LMICs according to the 2020 World Bank listing of countries that are LMICs (32). We used the following search terms: “bovine viral diarrhea,” “bovine viral diarrhea virus,” “bovine viral diarrhea,” “bovine viral diarrhea virus,” “BVD,” “deveoping,” “less developed,” “under developed,” “middle income,” “low income,” “LMIC,” “LAMI,” “transitional countries,” “risk factors,” and “economic,” and these were combined with Boolean operators “OR” and “AND” (33).

Inclusion Criteria

Inclusion criteria were defined based on the population, intervention, comparison, outcomes of an article, and study design framework (known as the PICOS framework), but in case of absence of intervention and comparison groups for PICOS, we also followed the PEO (population, exposure, outcome) framework (34). Studies analyzing exposure to BVDV in cattle from LMICs were included if they addressed questions related to risk factors for BVDV infection or if they reported health or economic outcomes for this condition. We followed the guidelines provided by the PRISMA group. We included all study types, all cattle production systems, dairy, beef, and mixed, all study designs, all years of study or publication, all studies reporting the prevalence of BVDV and its risk factors or its health and economic impacts, BVD positivity in all sample types, including blood, semen, bulk tank milk, ear-notch, skin and other tissues, LMICs using World Bank 2020 classification of countries, and studies in English or Spanish. We included articles in Spanish because of high numbers of studies identified in South America with relevant information for this review (35–39).

Exclusion Criteria

We excluded species that were not cattle (studies specific on sheep, goats, pigs, or wild ruminants), because BVD is predominantly a disease of cattle and they are generally considered to be natural hosts (40). We excluded review articles (as they did not include any type of study, investigations or comparison groups, reportable data, and results to analyze), articles focused on cattle populations in high-income countries, and articles where the language was not English or Spanish.

Quality Assessment of Articles

We evaluated the quality of the articles we used by risk of bias (ROB) assessment (41). Assessment of ROB helps informs readers of any potential bias in individual studies and helps facilitate their interpretation and gives an understanding of how bias might impact estimates of reliability. ROB assessment was conducted on individual studies and was based on the critical appraisal skill program (CASPS) checklist tool (42, 43). The checklist was slightly modified to ensure relevance to the review topic. All questions were imported into the Covidence version 2.0 (Veritas Health Innovation, Melbourne, Australia) software tool for systematic reviews (33). The ROB assessment consisted of 11 questions checking and assessing article quality. These included questions about the aims and design of the study, whether the target populations were well-defined, and if the results were correctly analyzed, (Supplementary Table S1). Risks were ranked based on high, low, or some concerns by at least two independent reviewers (BZS, LGG, and LHC). Low ROB meant that in the reviewer's opinion the ROB question was clearly addressed by the article. High ROB meant that the ROB question was not clearly addressed by the article or no information was provided at all, which could lead to bias. “Some concerns” meant that the there was no clear information provided to assess the ROB question in the article as assessed by the reviewers, making it difficult to judge. All articles with high risk were still eligible for data extraction and analysis (44). Although not recommended by CASPS to score articles, we needed a grading system to give an overall picture of the quality of our study articles. We graded our articles as low risk of bias (score = 2), some concerns (score = 1), and high risk of bias (score = 0), applying the criteria outlined by Yan et al. (44). All the scores were then added to give a total score for each article. The total scores ranged between 9 and 21, and a cut-off of 10 or less was agreed upon as high risk of bias article, a score between 11 and 18 was agreed upon as having some concerns, and 18 and greater was agreed upon as a low risk of bias article. Low-risk articles were then given a weighting of 2, high scoring articles were given a weighting of 0, and articles with some concerns were given a weighting of 1. All the scores and weights were then added to give each article an overall judgment using a robvis traffic light system as outlined by McGuinness et al. (45).

Statistical Analysis

Statistical analyses, plotting, and mapping were conducted using the R statistic software (46, 47) in R Studio version 4.0.4 (48) using functions from the following key packages: tidyverse (49), meta (30), and ggplot2, which is a part of tidyverse. The articles imported into Covidence were extracted as a csv file and imported in R Studio for cleaning and analyses. Choropleth maps were plotted using geographic information system (GIS) software and spatial libraries sp, sf, ggmap, tmap, and rgdal (50, 51). A meta-analysis was conducted using the meta packages dmeta, metaphor, and meta (30, 52). A weighted meta-analysis of animal-level prevalence of BVD reported in the reviewed articles was conducted using a logistic random-effects model (53). The random procedure incorporates an extra variance component to account for variability between the studies (heterogeneity) in addition to the within-study (sampling) variance as a result of sampling error. The weightings were calculated using the inverse variance method. The model assumes that r_ij, which was the number of positive animals in study i, was a randomly distributed binomial variable where the number of animals tested is in the study was n_ij and the unobserved true prevalence was p_ij as follows: r_ij ~ binom(n_ij, p_ij).

The log transformation of the odds (logit) can be expressed as a linear model as follows:

$\begin{array}{l}log\left(\frac{p_{ij}}{1-p_{ij}}\right)={\beta }_0+U_{0j}+C_k\end{array}$

where $U_{0j}~N(0,{\sigma }_{u0}^2)$ and where β₀ is the intercept, U_0j is the random effect for the intercept, and C_k is an addition fixed effect to account for potential heterogeneity by continent. The results of the overall meta-analysis were presented as a forest plot.

A choropleth map was plotted to show the seroprevalence of antibodies to BVDV by country. Where more than one study was available a mean of the study prevalences was used. The risk of bias descriptive results were plotted using a robvis traffic light plot (45) and a weighted barplot to graphically represent the ROB in each study. A sensitivity analysis was conducted by repeating the meta-analysis but removing studies with high ROB.

A descriptive analysis of reported risk factors and health impacts of BVD was conducted, and a dot matrix plot was used to display the proportions of articles reporting a given factor or an impact. Proportions of risk factor articles were calculated with confirmed risk factors/total studies.

Results

There were 690 articles identified in the initial search (Figure 1). CAB Abstracts yielded 137 articles (from 1973 to 2020), Embase 28 (From 2000 to 2020), Medline 42 (from 1946 to 2020), and Global Health 26 (from 1910 to 2020). ProQuest consists of journals, theses, and dissertations from 1952 to 2020 and yielded 6,000 results. The articles were screened based on relevance to our inclusion criteria, and 335 were excluded. Out of the remaining 355, 50 duplicates were removed. The remaining 305 were further screened for title and abstract relevance and a further 101 were removed. The remaining 204 articles were screened for relevant information regarding risk factors and health or economic outcomes. A total of 121 articles were imported into Covidence. All of them passed the abstract screening stage into the full text review stage. The full text of six eligible studies was not available online, nor was it available in any holding library, and therefore these studies were eliminated. A further 56 articles were eliminated based on language other than English or Spanish, wrong outcomes, wrong settings, and for being review articles rather than primary studies. A total of 59 articles were accepted for full review, data extraction, and analysis. Data were extracted from each articles by at least two independent reviewers (BZS, LGG, and LHC) and input directly into the Covidence software tool. Consensus data were extracted for statistical analysis and plotting. Five articles out of the final 59 had various investigations, and we added the 5 separate investigations (54–58) to our 59 articles, giving us a total of 64 investigations. Therefore, it is important to note that because the five articles have included more than one prevalence type, the final sum of the number of studies is not identical to the total number of records. The process of study retrieval and selection is indicated in Figure 1.

FIGURE 1

Figure 1. Flow diagram indicating the process of study retrieval and inclusion as described in the PRISMA statement.

Characteristics of Included Studies

The final set of articles included studies from four regions, Asia, SSA, South America, and the Middle East, and covered 23 individual countries. Brazil had the highest number of studies (n = 12), while most countries had very few studies (Supplementary Table S2). The studies were classed as cross-sectional (n = 55; 84.3%), case-control (n = 4; 6.2%), or cohort (n = 4, 6.2%); in one (1.5%) case, the design was not clear from the description. The majority of the studies reported conducting random (n = 25), purposeful (n = 8), or convenience (n = 8) sampling, with only a few studies designing a multi-stage sampling process (n = 6). Some of the studies did not mention their sampling or selection process. The samples used were mainly serum, bulk tank milk, and animal tissue (ear pinnae tissue “earnotch”). BVDV antibodies or antigens were analyzed by ELISA (n = 50), fluorescent antibody tests (FATs) (n = 3), polymerase chain reactions (PCRs) (n = 7), serum neutralization (SN) or virus neutralization (VN) (n = 8) tests, or virus isolation (VI) (n = 1). Majority of the articles had a combination of various diagnostic tests. Vaccination was not a common practice in most of the study sites reported, and only two studies indicated that routine BVD vaccination was performed.

Out of the final set of studies, 41 (64.0%) reported results of analyses of risk factors, 7 (10.9%) reported results of analyses of health impact, and 3 (4.6%) studies reported analyses of economic impacts of BVD. Only 1 study (59) reported analyses of all 3 outcomes of interest (Figure 2).

FIGURE 2

Figure 2. Heatmap showing breakdown of study ID and the outcomes they analyzed. The plot shows the study authors' contributions to investigation of the outcomes of interest for risk factors, health and economic impacts of BVD.

Quality Assessment

Our results show the number of studies that looked at BVD and answered the 11 questions set to address ROB (refer to list of domain questions in Supplementary Table S1). The result shows that across all studies, 50% of the studies in this review were not free of selective outcome reporting (i.e., the studies should have obtained data on all domains). More than 50% of the studies had problems with their sampling methods, study design, sampling of target population, and ethical approval. They were either not reporting correctly or not giving the authors enough information to judge.

Prevalence: Seroprevalence

All the articles gave information about prevalence. The animal-level seroprevalence was as high as 66.3% (95% CI: 61.7–70.6%) with a wide variation between regions. The country-level weighted mean seroprevalence range is between 8.2 and 66.4% and is mapped in Figure 3, while Figure 4 shows the unadjusted individual study prevalence estimates and 95% CI (based on raw numbers of positives and sample size and not accounting for any design effects or imperfect tests). Regional prevalence across Asia, South America, SSA, and the Middle East was estimated to be 21.6% (CI 6.0; 56), 45.2% (CI 35.9; 54.8), 39.4% (CI 25.5; 56.17), and 49.9% (CI 25.5; 74.3), respectively. Figure 5 presents the weighted adjusted estimates at both the continent level and overall (estimates in gray).

FIGURE 3

Figure 3. Plot showing risk of bias weights by authors judgement. The weighted plots show proportion of information with each judgement within each domain. Supplementary Table S1 explains the legend on the y-axis.

FIGURE 4

Figure 4. Choropleth map showing weighted mean reported seroprevalence by country. The regions included in this review were Africa, Asia, Middle East, and South/Central America.

FIGURE 5

Figure 5. A forest plot of the weighted meta-analysis of reported BVD antibody seroprevalence by country. The plot shows the study ID, the country where the study was conducted, the sample size, the number of positive test results for antibodies to BVDV, the risk of bias rank overall, the point estimate with 95% confidence interval with the box proportional to the weighting of the fixed effect, the numerical value of the point estimate of BVD prevalence (the effect size) and the estimated 95% confidence interval (unadjusted in black, weighted adjusted in green).

Weighted means are estimated by production system (Table 2) and farm type (Table 3). The majority of studies came from dairy farms with a mean of 44.5%. The mean seroprevalence estimate for other production systems was between 30 and 50% with large confidence intervals, and there was little evidence of major difference between systems (Table 2). Similarly, there appear to be little differences across farm types (Table 3). Most of the studies were conducted in small holder farms with mean seroprevalence of 41.3%.

TABLE 1

Table 1. Final list of studies included in the systematic review of BVDV in LMICS ordered by year of publication.

TABLE 2

Table 2. Summary table of weighted mean seroprevalence of antibodies to BVDV in low- and middle-income countries stratified by production system.

TABLE 3

Table 3. Summary table of weighted mean seroprevalence of antibodies to BVDV in low-and middle income countries stratified by farming system.

Prevalence: Antigen Prevalence

Only nine articles reported prevalence of antigen-positive animals (i.e., infection), identifying PI or TI animals. The prevalence of antigen was 7.0% (CI 0.0; 100.0) in Asia, 10.0% (CI 0; 90.0) in SSA, 6.0% (CI 5.0; 8.0) in South America, and 9% (CI 0.0; 99.8) in the Middle East (Figure 5).

Risk Factors

Forty-one studies (64%) reported results of risk factor analyses and the summary is presented in Figure 7. Age, breed, herd size, introducing animals, breeding practices, farm workers, and cleanliness were the risk factors studied across various production systems. Herd size was studied across all systems with proportions of 20.6%, 33.3%, and 38.4% in dairy, beef, and mixed systems, respectively. In dairy systems, age (24.1%), introducing animals (3.4%), and herd size (20.6%) were mostly studied and reported. Supplementary Table S4 presents in more detail all the risk factor groupings investigated in this review.

Health Impact

Only seven studies (10.9%) reported on health impacts of BVDV (39, 55, 59–62). The common health impacts reported across regions and systems were abortions, and fetal complications in dairy systems. The proportion of studies that investigated abortion was 13.7%, and the proportion of studies that reported on fetal complication was 13.7% (Figure 7).

Economic Impact

Only 3 (4.6%) studies (59, 63, 64) reported on economic losses in terms of milk yield but none quantified the losses as a cost at the herd, local, or national economic levels.

Discussion

The objective of this review was to identify the literature reporting the prevalence of BVDV in LMICs and to summarize the risk factors associated with it, and finally its health and economic impacts. It is generally advised to limit a systematic review to a specific study design to avoid heterogeneity of study types (65). Our inclusion of a variety of study types could mean that studies are not completely comparable in the understanding of prevalence and other risk factors. Although some of our studies had small populations and sample sizes that could be a limitation in this review, they all passed the ROB analysis as low risk of bias (see Figure 3; Supplementary Figure S1 for scoring). Consequently, we have included all study types and study designs to include as much relevant information as possible, which may lead to better understanding of BVDV epidemiology in LMICs.

From the final 59 articles from which relevant data could be extracted, there were 64 studies (Table 1). Between individual studies across regions, there was an extremely wide variation in estimates of seroprevalence (Figure 5). The variation in mean seroprevalence in different farming systems could be a result of true differences in seroprevalence or due to bias estimates from poorly designed studies. As systematic random sampling was rarely conducted to take into account the existing cattle structure in different cattle populations, this could introduce some bias and give seroprevalence that is not representative of the region of study. Also, sample sizes in the different studies were not equal, as some studies had bigger populations and bigger sample sizes thereby getting a true reflection of prevalence in the region, while smaller sample size may be a limitation in obtaining the true prevalence of the disease. The findings highlight the benefit of conducting a systematic review to inform the design of future studies to obtain representative prevalence estimates from the cattle population of interest.

The variations in prevalence at the regional level highlights that there is a significant gap in information across vast parts of the global south where livestock are an important part of many household livelihoods. Based on the weighted meta-analysis using a random-effects model and given this large range in individual study seroprevalence, the regional mean seroprevalence did not vary much across South America, SSA, and the Middle East but appeared to be slightly lower in Asia (21.6%). At the country level, Peru had the highest weighted seroprevalence but with only three studies. The variation may reflect to some extent the importance and scale of cattle keeping in these areas compared to Asia and possibly the lower seroprevalence may be down to a lower cattle density in some parts of Asia. These estimates are similar to the others reported in the literature (12, 13, 66). Despite the range of seroprevalence across geographies, the overall high seroprevalence reported across LMICs suggests that there are likely inefficiencies in livestock production across these regions (4, 67). As referred to previously, many of the studies are quite limited both in numbers sampled and in terms of geographical spread, which contributes to high levels of uncertainty in all the regional weighted means. It is therefore important for future studies to fully describe their study design and population structure for estimates to be correctly interpreted.

In addition to serological surveys, antigen studies also showed the presence of the virus in different regions, suggesting that there is ongoing transmission (Figure 6). Due to the transient nature of most BVDV infections and limited life span of PI animals, antigen-test based surveys are not necessarily useful for estimating the magnitude of transmission but can be useful for describing the molecular epidemiology of virus strains circulating (54). Few studies in this review utilized genomic tools to describe virus types present and future investigations should incorporate such testing to understand transmission networks within and between cattle populations.

FIGURE 6

Figure 6. A forest plot of the weighted meta-analysis of reported BVDV antigen prevalence by country. The plot shows the study ID, the country where the study was conducted, the sample size, the number of positive test results for BVDV, the risk of bias rank overall, the point estimate with 95% confidence interval with the box proportional to the weighting of the fixed effect, the numerical value of the point estimate of BVDV prevalence (the effect size) and the estimated 95% confidence interval (unadjusted in black, weighted adjusted in green).

In this review, we summarized risk factors based on 41 articles that specifically investigated risk factors. From Figure 7, we see that across production systems, herd size, introducing new animals into a farm, breeding practices, including AI, farm workers, age, and breed, were the most frequently investigated. Herd size was as a risk factor consistent in all production systems in our review, and this finding also consistent with studies conducted in high-income regions (68, 69). Most of the studies indicated that larger density stocking was a risk factor to BVD and having a smaller herd size was a protective factor. Majority of the studies that indicated herd size as a risk factor were assessed to have used a sample size that was representative in their study and a study design appropriate for their investigation. Authors discussed that herd size as a risk factor could be influenced by poor management practices, cattle density, and presence of PI animals (11, 60, 66). Larger herd size increases the risk of BVDV seropositivity (68) compared to smaller herds possibly because there is higher probability of transmission to animals in the correct stage of pregnancy ensuring generation of new PI calves that can support persistent transmission. Another reason may be as indicated by Lindberg et al. (70) that there is a low likelihood of self-clearance of virus due to continuous maintenance of the virus in herds by PIs. In contrast, smaller herds tend to have better opportunity for self-clearance and eliminate further opportunity for BVDV exposure in naïve animals.

FIGURE 7

Figure 7. A dot matrix plot showing risk factors and health impacts reported by production system. The plot shows a meta-analysis of the pooled studies. The size of the points represents proportion of studies that reported the risk factor or health impact of BVDV for a given production system. The proportions were calculated by the formula: number of confirmed risk factor papers/total number of papers in each production system. Detailed explanations of the risk factors and how they have been grouped can be found in Supplementary Table S3.

Age was a frequently studied risk factor in beef and dairy systems. Our results show that there was a variation in directionality of age as a risk factor. Most of the studies indicated that being above 6 months increased the risk of exposure except for one study that reported that there was no association with age (71). Our review is different from a study by Houe and Meyling (72) who determined that in Danish dairy herds, incidence risk of BVDV infection was approximately similar in all age groups. The dissimilarity between our review findings and those of Houe et al. could be due to different systems of production. For example, the studies included in this review dealt mainly with smallholder and small farm sizes that do not practice vaccination or culling; therefore, animals stay in the farm until they are ready to be slaughtered. Other studies similar to ours reported that age was a significant risk factor for BVDV seropositivity with older animals being more positive to the virus (11, 66, 73). The pathogenesis of BVD shows that animals that are born as PI animals may die early because of mucosal disease or be slaughtered because of retardation usually by 2 years of age (74). However, if PI animals remain within a cattle population, they continue to maintain BVDV in the herd, exposing animals in the herd over time (73), possibly explaining the reported increasing likelihood of seropositivity with age. Because it's not clear if the risk factor is increasing or decreasing with age, further studies are needed to determine external factors which might be contributing to these varying results.

Breed was investigated as a risk factor in proportionally more studies conducted on beef systems than those conducted on dairy and mixed systems (Figure 7). It is difficult to compare between studies because there are diverse breeds in different areas. There is conflicting evidence in the literature about the importance of breed as a risk factor for example, Demil et al. (66) reported that there was no significant association between breed and BVDV seropositivity, whereas other studies have found a link between breed and BVD seropositivity. However, breed types were not consistent in the studies (62, 75). Breeds that are most susceptible have not been categorized either by farming or production systems. From our review, it was not possible to ascertain any clear pattern, so it is not possible to draw any conclusions about any specific breeds that are of higher or lower susceptibility to BVDV due to the heterogeneity of breeds in different regions. Future studies could conduct genotyping (76) to detect genomic regions associated with BVD to determine if particular breeds are more resilient or susceptible to BVDV.

Factors associated with biosecurity and hygiene were risk factors in beef and dairy systems. Farm workers, including veterinarians, milkers and other employees, and being a farm that provides milk to industries were associated with BVDV seropositivity. The high seroprevalence in dairy systems (Table 2) may also be due to the involvement of milkers, staff, and equipment during milking (73, 75, 77), which could be potential sources of contamination. Our review is similar to (75, 78, 79) and suggests that farm workers can be a potential source introducing the virus into farms. As suggested by Almeida et al. (11), farm workers can move between farms, visiting many farms in a day, and using the same clothing and instruments, and indirectly transmit the virus through plastic gloves. An important cleanliness measure to control infection from farm workers can include use of protective clothing and not allowing transportation staff to enter cow houses (69, 73). Good to medium hygiene and burying dead animals were reported as risk factors. Burning of dead animals rather than burying or disposal was reported as a protective factor, as this process successfully eliminates BVDV (80). Although not common across all production systems, some studies highlighted that breeding practices [artificial insemination (AI) and natural breeding programs] were risk factors to BVDV in dairy farms. Two of the studies reported that for AI as a risk factor, natural mating using own bulls was a protective factor (11, 81). This may indicate that infected bulls are being introduced into dairy herds for insemination, or infected semen is being distributed to AI technicians (59, 62). Introducing new animals into a herd was frequently studied and reported as a risk factor in dairy systems but not in beef systems. BVDV can be introduced into a herd by introducing new animals either through purchase or as gifts. Our result is supported by previous authors who also found that purchase of animals can introduce a PI animal into the new herds (75, 79, 82) from infected farms, thereby maintaining BVDV infection at the population level.

Exploring the motivations for herd biosecurity, cattle movements, and breeding practices would inform the design and feasibility of future control programs in LMIC settings. For example, when a study in Ireland by Graham et al. (69) found that risk of BVDV was mainly contracted from neighboring herds, movement of animals, and through farm workers a control measure was put in place to restrict cattle movement and to notify neighboring farms of their proximity to PI animals. Such measures gave farmers the opportunity to enhance their biosecurity and help in national control programs. A study conducted by Van Schaik et al. (83) suggests that understanding the risk factors and epidemiologic spread of a disease can contribute to economic benefits, which may include a more closed farming systems, rearing own young stock, and providing the cost of farm clothing.

There was little information found in studies on health implications of BVD in LMICs. The health impacts of the disease are the main cause of negative economic impacts (66). Our review found that health investigations were mainly conducted on dairy farms, with abortion and fetal complications being the more commonly reported health complications (Figure 7). However, overall few studies reported abortions or fetal complications in dairy farms, and even fewer studies reported abortions and fetal complications in beef and mixed systems. Majority of the studies that reported on abortions had their investigations on coinfections of BVDV with other diseases, so it is unclear whether BVDV was directly a cause of abortion. Previous studies reported that there are no links between BVDV and abortion (14) and have attributed the cause of abortion to be other causative agents or that cattle exposed to BVDV have cleared the infection before breeding (66). Similarly, in our review, the presence of other diseases, such as neosporosis, brucellosis, leptospirosis, and mastitis (Supplementary Table S3), was studied alongside BVDV. One of the studies by Ortega et al. (80) indicated that the presence of other diseases causing abortion were risk factors to BVDV. They reported that giving animals ivermectin, concentrated feed supplement, and organophosphates was a protective factor in farms. Although these medications would not directly treat BVD, they are possibly associated with farmers who are engaged in improved biosecurity practices. As other infections can also lead to abortion and other reproductive problems in BVD farms, it is important to consider these factors and establish a health program that includes providing prophylaxis for these infections (75). Further work will be important to investigate whether there is a relationship between abortion and other reproductive disorders and BVD in different systems as there's still not enough evidence, and these studies should aim to benchmark BVD and abortion from other causes of abortion such as return to oestrus and placental retention.

The review identified that the economic impact of BVD was minimally described and the cost implications were not quantified in production systems in LMICs. The review also highlighted the impact on milk yield in three articles, with all the studies indicating that decrease in milk yield was significantly associated with BVDV infection. The cost implications of BVD have been investigated in many high-income settings, but minimal work has been done in LMICS. Considering the importance of livestock production in livelihood incomes in LMICs, exploring whether BVD is associated with low milk production and poor production outcomes needs to be further investigated.

Study Limitation

One of our study limitations was the inclusion of studies published only in English and Spanish. This was due to availability of speakers of the English and Spanish languages for data extraction. We excluded articles in languages not English or Spanish, as we believe these may lead to bias to studies conducted by researchers speaking other languages. This could also lead to loss of data from relevant articles that were not conducted in English or Spanish. Another limitation might be the inclusion of all study designs. It is generally advised to limit a systematic review to a specific study design to avoid heterogeneity of study types (65). We included all study types and study designs to incorporate all relevant information that may lead to better understanding of BVD epidemiology in LMICs. However, our inclusion of a variety of study types could mean that studies are not completely comparable in the understanding of prevalence and other risk factors. Some of the articles in this review had small populations and sample sizes, which could be a limitation. We believe that our ROB analysis and meta-analysis for adjusted weighing has dealt with issues of heterogeneity and comparable sample sizes.

Conclusion

It can be concluded that BVDV is present and circulating in the regions of SSA, Asia, Middle East, and South America, and across various farming and production systems found in these regions; however, the prevalence estimates vary across these regions. The variety of risk factors of epidemiological importance is likely to be linked to various production systems within each locality. Standardized methods of estimating prevalence and the varying prevalence in diverse farming and production systems will need to be considered in trying to offer solutions for disease control strategies. The health and economic impacts associated with BVD are mostly related to reproductive performance but are poorly quantified. Future investigations should focus on quantifying the negative effect of BVD on cattle production systems in LMICs to prioritize and inform future approaches to control.

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

Author Contributions

BZ-S, RK, and BB: writing. BZ-S, LGG, and LH-C: data extraction. LH-C, BB, and BZ-S: statistical analysis. RK, BB, EC, and BZ-S: review and editing. RK, BB, and EC: supervisors. All authors contributed to the article and approved the submitted version.

Funding

BZ-S was supported by an EastBio PhD studentship funded by BBSRC. BB, LH-C, and EC were supported by the Bill & Melinda Gates Foundation and by a UK aid from the UK Foreign, Commonwealth, and Development Office (Grant Agreement OPP1127286) under the auspices of the Centre for Tropical Livestock Genetics and Health established jointly by the University of Edinburgh, Scotland's Rural College, and the ILRI. LGG was supported by the University of Edinburgh through the Principal's Career Development and the Edinburgh Global Scholarships.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fvets.2022.947515/full#supplementary-material

References

1. Ridpath JF. Bovine viral diarrhea virus: global status. Vet Clin North Am Food Anim Pract. (2010) 26:105–21. doi: 10.1016/j.cvfa.2009.10.007

PubMed Abstract | CrossRef Full Text | Google Scholar

2. Bauermann FV, Flores EF, Ridpath JF. Antigenic relationships between Bovine viral diarrhea virus 1 and 2 and HoBi virus: possible impacts on diagnosis and control. J Vet Diagnostic Investig. (2012) 24:253–61. doi: 10.1177/1040638711435144

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Lysholm S, Ramabu SS, Berg M, Wensman JJ. First-time detection of bovine viral diarrhoea virus, BVDV-1, in cattle in Botswana. Onderstepoort J Vet Res. (2019) 86:1–7. doi: 10.4102/ojvr.v86i1.1764

PubMed Abstract | CrossRef Full Text | Google Scholar

4. Baur I, Tabin L, Banda ML, Chiumia D, Lips M. Improving dairy production in Malawi: a literature review. Trop Anim Health Prod. (2017) 49:251–8. doi: 10.1007/s11250-016-1184-5

PubMed Abstract | CrossRef Full Text | Google Scholar

5. de Clare Bronsvoort BM, Thumbi SM, Poole EJ, Kiara H, Auguet OT, Handel IG, et al. Design and descriptive epidemiology of the infectious diseases of east african livestock (IDEAL) project, a longitudinal calf cohort study in western Kenya. BMC Vet Res. (2013) 9:1–22. doi: 10.1186/1746-6148-9-171

PubMed Abstract | CrossRef Full Text | Google Scholar

6. Perry B, Grace D. The impacts of livestock diseases and their control on growth and development processes that are pro-poor. Philos Transact R Soc B Biol Sci. (2009) 364:2643–55. doi: 10.1098/rstb.2009.0097

PubMed Abstract | CrossRef Full Text | Google Scholar

7. Sumption K, Domenech J, Ferrari G. Progressive control of FMD on a global scale. Vet Rec. (2012) 170:637–9. doi: 10.1136/vr.e4180

PubMed Abstract | CrossRef Full Text | Google Scholar

8. Neill JD. Molecular biology of bovine viral diarrhea virus. Biologicals. (2013) 41:2–7. doi: 10.1016/j.biologicals.2012.07.002

PubMed Abstract | CrossRef Full Text | Google Scholar

9. Bauermann F V, Ridpath JF, Weiblen R, Flores EF. HoBi-like viruses: an emerging group of pestiviruses. J Vet Diagnost Investig. (2013) 25:6–15. doi: 10.1177/1040638712473103

PubMed Abstract | CrossRef Full Text | Google Scholar

10. Lanyon SR, Hill FI, Reichel MP, Brownlie J. Bovine viral diarrhoea: pathogenesis and diagnosis. Vet J. (2014) 199:201–9. doi: 10.1016/j.tvjl.2013.07.024

PubMed Abstract | CrossRef Full Text | Google Scholar

11. Almeida LL, Miranda ICSS, Hein HE, Neto WS, Costa EF, Marks FS, et al. Herd-level risk factors for bovine viral diarrhea virus infection in dairy herds from Southern Brazil. Res Vet Sci. (2013) 95:901–7. doi: 10.1016/j.rvsc.2013.08.009

PubMed Abstract | CrossRef Full Text | Google Scholar

12. Ularamu HG, Sibeko KP, Bosman AB, Venter EH, van Vuuren M. Genetic characterization of bovine viral diarrhoea (BVD) viruses: confirmation of the presence of BVD genotype 2 in Africa. Arch Virol. (2013) 158:155–63. doi: 10.1007/s00705-012-1478-5

PubMed Abstract | CrossRef Full Text | Google Scholar

13. Hyera JMK, Liess B, Frey HR. Bovine viral diarrhoea virus infection in cattle, sheep and goats in northern Tanzania. Trop Anim Health Prod. (1991) 23:83–94. doi: 10.1007/BF02361187

PubMed Abstract | CrossRef Full Text | Google Scholar

14. Van Vuuren M. Bovine viral diarrhoea virus infection in livestock in southern Africa. CAB Rev Perspect Agric Vet Sci Nutr Nat Resour. (2006) 1–15. doi: 10.1079/PAVSNNR20061009

CrossRef Full Text | Google Scholar

15. Richter V, Kattwinkel E, Firth CL, Marschik T, Dangelmaier M, Trauffler M, et al. Mapping the global prevalence of bovine viral diarrhoea virus infection and its associated mitigation programmes. Vet Rec. (2019) 184:711. doi: 10.1136/vr.105354

PubMed Abstract | CrossRef Full Text | Google Scholar

16. Grooms DL. Reproductive consequences of infection with bovine viral diarrhea virus. Vet Clin North Am – Food Anim Pract. (2004) 20:5–19. doi: 10.1016/j.cvfa.2003.11.006

PubMed Abstract | CrossRef Full Text | Google Scholar

17. Hons GR, Rocchi BVM. The Moredun Foundation. Control of Bovine Viral Diarrhoea (BVD) (2020).

Google Scholar

18. Kalaycioglu AT. Bovine viral diarrhoea virus (bvdv) diversity and vaccination. review. Vet Q. (2007) 29:60–7. doi: 10.1080/01652176.2007.9695228

PubMed Abstract | CrossRef Full Text | Google Scholar

19. Dubovi EJ. Laboratory diagnosis of bovine viral diarrhea virus. Biologicals. (2013) 41:8–13. doi: 10.1016/j.biologicals.2012.06.004

PubMed Abstract | CrossRef Full Text | Google Scholar

20. Saliki JT, Dubovi EJ. Laboratory diagnosis of bovine viral diarrhea virus infections. Vet Clin North Am Food Anim Pract. (2004) 20:69–83. doi: 10.1016/j.cvfa.2003.11.005

PubMed Abstract | CrossRef Full Text | Google Scholar

21. Hanon JB, De Baere M., De la Ferté C, Roelandt S, Van der Stede Y, Cay B. Evaluation of 16 commercial antibody ELISAs for the detection of bovine viral diarrhea virus–specific antibodies in serum and milk using well-characterized sample panels. J Vet Diagnostic Investig. (2017) 29:833–43. doi: 10.1177/1040638717724839

PubMed Abstract | CrossRef Full Text | Google Scholar

22. Kampa J, Ståhl K, Renström LHM, Alenius S. Evaluation of a commercial erns-capture ELISA for detection of BVDV in routine diagnostic cattle serum samples. Acta Vet Scand. (2007) 49:1–7. doi: 10.1186/1751-0147-49-7

PubMed Abstract | CrossRef Full Text | Google Scholar

23. Xia H, Liu L, Nordengrahn A, Kiss I, Merza M, Eriksson R, et al. A microsphere-based immunoassay for rapid and sensitive detection of bovine viral diarrhoea virus antibodies. J Virol Methods. (2010) 168:18–21. doi: 10.1016/j.jviromet.2010.04.009

PubMed Abstract | CrossRef Full Text | Google Scholar

24. Gunn GJ, Stott AW, Humphry RW. Modelling and costing BVD outbreaks in beef herds. Vet J. (2004) 167:143–9. doi: 10.1016/S1090-0233(03)00112-6

PubMed Abstract | CrossRef Full Text | Google Scholar

25. Tiwari A, VanLeeuwen JA, Dohoo IR, Keefe GP, Haddad JP, Tremblay R, et al. Production effects of pathogens causing bovine leukosis, bovine viral diarrhea, paratuberculosis, and neosporosis. J Dairy Sci. (2007) 90:659–69. doi: 10.3168/jds.S0022-0302(07)71548-5

PubMed Abstract | CrossRef Full Text | Google Scholar

26. Evans CA, Pinior B, Larska M, Graham D, Schweizer M, Guidarini C, et al. Global knowledge gaps in the prevention and control of bovine viral diarrhoea (BVD) virus. Transbound Emerg Dis. (2019) 66:640–52. doi: 10.1111/tbed.13068

PubMed Abstract | CrossRef Full Text | Google Scholar

27. Valle PS, Martin SW, Skjerve E A. bayesian approach to estimating the performance of a bovine virus diarrhoea virus (bvdv) antibody elisa bulk-tank milk test. Prev Vet Med. (2001) 50:71–87. doi: 10.1016/S0167-5877(01)00221-5

PubMed Abstract | CrossRef Full Text | Google Scholar

28. Ahn E, Kang H. Introduction to systematic review and meta-analysis: a health care perspective. Korean J Anesthesiol. (2018) 71:1–38. doi: 10.4097/kjae.2018.71.2.103

PubMed Abstract | CrossRef Full Text | Google Scholar

29. Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ. (2011) 343:d5928. doi: 10.1136/bmj.d5928

PubMed Abstract | CrossRef Full Text | Google Scholar

30. Borenstein M, Hedges L, Rothstein H. Meta-Analysis Fixed Effect vs. Random effects Test. (2007). 162 p. Available from: www.meta-analysis.com (accessed October 28, 2021).

Google Scholar

31. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred Reporting Items for Systematic Reviews Meta-Analyses: The PRISMA Statement. (2010). Available from: www.prisma-statement.org/ (accessed August 25, 2021).

Google Scholar

32. Cochrane (EPOC 2020),. World Bank List of Economies. Classification Based on EPOC LMIC Filters 2019 (v.4). (2020). Available online at: https://epoc.cochrane.org/lmic-filters (accessed January 1, 2021).

33. Babineau J. Product review: covidence (systematic review software). J Can Heal Libr Assoc. (2014) 35:68. doi: 10.5596/c14-016

PubMed Abstract | CrossRef Full Text | Google Scholar

34. Munn Z, Stern C, Aromataris E, Lockwood C, Jordan Z. What kind of systematic review should i conduct? A proposed typology and guidance for systematic reviewers in the medical and health sciences. BMC Med Res Methodol. (2018) 18:1–9. doi: 10.1186/s12874-017-0468-4

PubMed Abstract | CrossRef Full Text | Google Scholar

35. Buitrago Horta ER, Jimenez Escobar C, Zambrano Varon JL. Identification of Factors Associated With Exposure to Bovine Viral Diarrhea Virus (BVDV) in Dairy-Herd Calves in the Bogota savanna. Revista de Medicina Veterinaria. (2018). p. 63–73.

Google Scholar

36. Carrillo MV, Esp AC, Figueredo MV, Medrano KG, Contreras JAM. Determining the presence of antibodies to Neospora caninum and bovine viral diarrhea virus and its relationship to the reproductive performance of bovine females of Oicatá (Boyacá). Med Vet Zootec. (2014) 9:238–47.

Google Scholar

37. Villamil V V, Ramírez GC, Vera VJ, Jaime JA. First evidence of Bovine Viral Diarrhea Virus (BVDV) genotype 2 in Colombia. Rev Fac Med Vet Zootec. (2018) 65:11–26. doi: 10.15446/rfmvz.v65n1.72020

CrossRef Full Text | Google Scholar

38. Fernández JVR, Utrera ÁR, Martínez JPZ, Jenkins SO, Zurita LG, Islas AF, et al. Prevalence of antibodies against bovine viral diarrhea in non-vaccinated cows from the States of Puebla, Tabasco and Veracruz, Mexico Rev Mex Ciencias Pecu. (2018) 9:555–66. doi: 10.22319/rmcp.v9i3.4599

CrossRef Full Text | Google Scholar

39. Villar FA, Salazar EL. Seroprevalence of bovine viral diarrhea in dairy herds of the Mantaro Valley, Junin Region, Peru. Revista de Investigaciones Veterinarias del Perú. (2018) 29:1515–26. doi: 10.15381/rivep.v29i4.15347

CrossRef Full Text | Google Scholar

40. OIE Terrestrial Mannual,. Vol. Terrestria, Manual of Diagnostic Tests and Vaccines for Terrestrial Animals 2021. (2018). Available from: https://www.oie.int/fileadmin/Home/eng/Health_standards/tahm/3.04.07_BVD.pdf (accessed February 27, 2021).

Google Scholar

41. O'connor AM, Hu D, Totton SC, Scott N, Winder CB, Wang B, et al. A systematic review and network meta-analysis of bacterial and viral vaccines, administered at or near arrival at the feedlot, for control of bovine respiratory disease in beef cattle. Anim Heal Res Rev. (2019) 20:143–62. doi: 10.1017/S1466252319000288

PubMed Abstract | CrossRef Full Text | Google Scholar

42. Ma LL, Wang YY, Yang ZH, Huang D, Weng H, Zeng XT. Methodological quality (risk of bias) assessment tools for primary and secondary medical studies: what are they and which is better? Mil Med Res. (2020) 7:1–11. doi: 10.1186/s40779-020-00238-8

PubMed Abstract | CrossRef Full Text | Google Scholar

43. Critical Appraisal Skills Programme,. CASP (Systematic Review)Checklist. (2019). Available online at: https://casp-uk.net/referencing/ (accessed January 27, 2021).

44. Yan L, Mir M, Sanchez P, Beg M, Peters J, Enriquez O, et al. Standard quality assessment criteria for evaluating primary research papers from a variety of fields. Arch Pathol Lab Med. (2020) 144:1041–7. doi: 10.5858/arpa.2020-0217-SA

PubMed Abstract | CrossRef Full Text

45. Mcguinness LA, Higgins JPT,. Risk-of-Bias VISualization (robvis): An R package Shiny Web App for Visualizing Risk-of-Bias Assessments. (2020). Available from: https://cran.r-project.org/ (accessed March 3, 2021).

PubMed Abstract | Google Scholar

46. Bunn MKA. An Introduction to dplR. Seattle, WA: Washington University (2019) 10:1–8.

Google Scholar

47. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing. Available online at: https://www.R-project.org/ (accessed December 27, 2021).

48. RStudio Team,. RStudio: Integrated Development Environment for R. Boston, MA: RStudio, PBC (2021). Available online at: http://www.rstudio.com/ (Accessed December 27, 2021).

Google Scholar

49. Wickham H, Averick M, Bryan J, Chang W, D'Agostino L, McGowan, et al. Welcome to tidyverse. J Open Source Softw. (2019) 4:1686. doi: 10.21105/joss.01686

CrossRef Full Text | Google Scholar

50. Rlovelaceleedsacuk RL, Cheshire J, Oldroyd R. Introduction to Visualising Spatial Data in R Part I : Introduction. EloGeo (2017). p. 1–20.

51. Becker R.A, Wilks AR. Package “maps”. CRAN Repos. (2021). Available from: https://cran.r-project.org/web/packages/maps/maps.pdf (accessed February 28, 2021).

52. Harrer M, Cuijpers P, Furukawa TA, Ebert DD. Doing Meta-Analysis With R: A Hands-On Guide. 1st ed. Boca Raton, FL; London: Chapman & Hall/CRC Press (2021).

Google Scholar

53. Schwarzer G,. meta: An R Package for Meta-Analysis. Vol. 7. R News (2007). p. 26–33. Available from: http://cran.r-project.org/doc/Rnews/ (accessed October 28, 2021).

PubMed Abstract | Google Scholar

54. Vanleeuwen J, Muraya J, Gitau GK, Makau DN, Crane MB, Mckenna SLB, et al. Seroprevalence and risk factors of neospora caninum and bovine viral diarrhoea virus in smallholder dairy cattle in Kenya. East Afr. Sci. Technol. Innov. (2021) 3:1–21. doi: 10.37425/eajsti.v3i1.391

CrossRef Full Text | Google Scholar

55. Stahl K, Bjorkman C, Emanuelson U, Rivera H, Zelada A, Moreno-Lopez J, et al. prospective study of the effect of Neospora caninum and BVDV infections on bovine abortions in a dairy herd in Arequipa, Peru. Prev Vet Med. (2006) 75:177–88. doi: 10.1016/j.prevetmed.2006.02.006

PubMed Abstract | CrossRef Full Text | Google Scholar

56. Hasan SD, Alsaad KM. Bovine viral diarrhea and persistently infection of cattle at Nineveh province, Iraq. Basrah J Vet Res. (2018) 17:14–32. doi: 10.33762/bvetr.2018.160309

CrossRef Full Text | Google Scholar

57. Konnai S, Mingala CN, Sato M, Abes NS, Venturina FA, Gutierrez CA, et al. A survey of abortifacient infectious agents in livestock in Luzon, the Philippines, with emphasis on the situation in a cattle herd with abortion problems. Acta Trop. (2008) 105:269–73. doi: 10.1016/j.actatropica.2007.12.004

PubMed Abstract | CrossRef Full Text | Google Scholar

58. Handel IG, Willoughby K, Land F, Koterwas B, Morgan KL, Tanya VN. Seroepidemiology of bovine viral diarrhoea virus (BVDV) in the Adamawa Region of Cameroon and use of the spot test to identify herds with PI calves. PLoS ONE. (2011) 6:e21620. doi: 10.1371/journal.pone.0021620

PubMed Abstract | CrossRef Full Text | Google Scholar

59. Marques ALA, de Assis AC, Simoes SVD, de Tolentino MLD, de Azevedo SS. Risk factors associated with Bovine Viral Diarrhea Virus (BVDV) infection in the semiarid of the state of Paraiba, in the northeast region of Brazil. Semin Ciencias Agrar. (2016) 37:3095–105. doi: 10.5433/1679-0359.2016v37n5p3095

CrossRef Full Text | Google Scholar

60. Uddin MA, Ahasan ASML, Islam K, Islam MZ, Mahmood A, Islam A, et al. Seroprevalence of bovine viral diarrhea virus in crossbred dairy cattle in Bangladesh. Vet World. (2017) 10:906–13. doi: 10.14202/vetworld.2017.906-913

PubMed Abstract | CrossRef Full Text | Google Scholar

61. Okumu TA, John NM, Wabacha JK, Tsuma V, VanLeeuwen J. Seroprevalence of antibodies for bovine viral diarrhoea virus, Brucella abortus and Neospora caninum, and their roles in the incidence of abortion/foetal loss in dairy cattle herds in Nakuru District, Kenya. BMC Vet Res. (2019) 15:1–6. doi: 10.1186/s12917-019-1842-8

PubMed Abstract | CrossRef Full Text | Google Scholar

62. Aragaw K, Sibhat B, Ayelet G, Skjerve E, Gebremedhin EZ, Asmare K. Seroprevalence and factors associated with bovine viral diarrhea virus (BVDV) infection in dairy cattle in three milksheds in Ethiopia. Trop Anim Health Prod. (2018) 50:1821–7. doi: 10.1007/s11250-018-1624-5

PubMed Abstract | CrossRef Full Text | Google Scholar

63. Noaman V, Nabinejad AR. Seroprevalence and risk factors assessment of the three main infectious agents associated with abortion in dairy cattle in Isfahan province, Iran. Trop Anim Health Prod. (2020) 52:2001–9. doi: 10.1007/s11250-020-02207-8

PubMed Abstract | CrossRef Full Text | Google Scholar

64. Fernandes LG, de Campos Nogueira AH, De Stefano E, Pituco EM, Ribeiro CP, Alves CJ, et al. Herd-level prevalence and risk factors for bovine viral diarrhea virus infection in cattle in the State of Paraíba, Northeastern Brazil. Trop Anim Health Prod. (2016) 48:157–65. doi: 10.1007/s11250-015-0937-x

PubMed Abstract | CrossRef Full Text | Google Scholar

65. Thrusfield M, Christtley R. Veterniary Epidemiology. 4th ed. John Wiley & Sons (2018). p. 1–864.

Google Scholar

66. Demil E, Fentie T, Vidal G, Jackson W, Lane J, Alemu S, et al. Prevalence of bovine viral diarrhea virus antibodies and risk factors in dairy cattle in Gondar city Northwest Ethiopia. Prev Vet Med. (2021) 191:105363. doi: 10.1016/j.prevetmed.2021.105363

PubMed Abstract | CrossRef Full Text | Google Scholar

67. Campbell Z, Coleman P, Guest A, Kushwaha P, Ramuthivheli T, Osebe T, et al. Prioritizing smallholder animal health needs in East Africa, West Africa, and South Asia using three approaches: literature review, expert workshops, and practitioner surveys. Prev Vet Med. (2021) 189:105279. doi: 10.1016/j.prevetmed.2021.105279

PubMed Abstract | CrossRef Full Text | Google Scholar

68. Sarrazin S, Veldhuis A, Méroc E, Vangeel I, Laureyns J, Dewulf J, et al. Serological and virological BVDV prevalence and risk factor analysis for herds to be BVDV seropositive in Belgian cattle herds. Prev Vet Med. (2013) 108:28–37. doi: 10.1016/j.prevetmed.2012.07.005

PubMed Abstract | CrossRef Full Text | Google Scholar

69. Graham DA, Clegg TA, Thulke H-H, O'Sullivan P, McGrath G, More SJ. Quantifying the risk of spread of bovine viral diarrhoea virus (BVDV) between contiguous herds in Ireland. Prev Vet Med. (2016) 126:30–8. doi: 10.1016/j.prevetmed.2016.01.017

PubMed Abstract | CrossRef Full Text | Google Scholar

70. Lindberg A, Houe H. Characteristics in the epidemiology of bovine viral diarrhea virus (BVDV) of relevance to control. Prev Vet Med. (2005) 72:55–73. doi: 10.1016/j.prevetmed.2005.07.018

PubMed Abstract | CrossRef Full Text | Google Scholar

71. Selim AM, Elhaig MM, Moawed SA, El-Nahas E. Modeling the potential risk factors of bovine viral diarrhea prevalence in Egypt using univariable and multivariable logistic regression analyses. Vet world. (2018) 11:259–67. doi: 10.14202/vetworld.2018.259-267

PubMed Abstract | CrossRef Full Text | Google Scholar

72. Houe H, Meyling A. Prevalence of bovine virus diarrhoea (BVD) in 19 Danish dairy herds and estimation of incidence of infection in early pregnancy. Prev Vet Med. (1991) 11:9–16. doi: 10.1016/S0167-5877(05)80040-6

CrossRef Full Text | Google Scholar

73. Talafha AQ, Hirche SM, Ababneh MM, Al-Majali AM, Ababneh MM. Prevalence and risk factors associated with bovine viral diarrhea virus infection in dairy herds in Jordan. Trop Anim Health Prod. (2009) 41:499–506. doi: 10.1007/s11250-008-9214-6

PubMed Abstract | CrossRef Full Text | Google Scholar

74. Houe H. Epidemiology of bovine viral diarrhea virus. Vet Clin North Am Food Anim Pract. (1995) 11:521–47. doi: 10.1016/S0749-0720(15)30465-5

PubMed Abstract | CrossRef Full Text | Google Scholar

75. Anderson B, Sott T, Da Costa Freitas FL, Nascif IA, Rozolen JM, Fonseca-Alves CE, et al. Serological prevalence of bovine viral diarrhoea (BVD-1) in non-vaccinated dairy cattle in the municipality of Realeza-Paraná. Semin Agrar. (2020) 41:907–14. doi: 10.5433/1679-0359.2020v41n3p907

CrossRef Full Text | Google Scholar

76. Casas E, Hessman BE, Keele JW, Ridpath JF A. genome-wide association study for the incidence of persistent bovine viral diarrhea virus infection in cattle. Anim Genet. (2015) 46:8–15. doi: 10.1111/age.12239

PubMed Abstract | CrossRef Full Text | Google Scholar

77. Raaperi K, Nurmoja I, Orro T, Viltrop A. Seroepidemiology of bovine herpesvirus 1 (BHV1) infection among Estonian dairy herds and risk factors for the spread within herds. Prev Vet Med. (2010) 96:74–81. doi: 10.1016/j.prevetmed.2010.06.001

PubMed Abstract | CrossRef Full Text | Google Scholar

78. Gates MC, Humphry RW, Gunn GJ, Woolhouse MEJ. Not all cows are epidemiologically equal: Quantifying the risks of bovine viral diarrhoea virus (BVDV) transmission through cattle movements. Vet Res. (2014) 45:1–5. doi: 10.1186/s13567-014-0110-y

PubMed Abstract | CrossRef Full Text | Google Scholar

79. van Roon AM, Mercat M, van Schaik G, Nielen M, Graham DA, More SJ, et al. Quantification of risk factors for bovine viral diarrhea virus in cattle herds: A systematic search and meta-analysis of observational studies. J Dairy Sci. (2020) 103:9446–63. doi: 10.3168/jds.2020-18193

PubMed Abstract | CrossRef Full Text | Google Scholar

80. Ortega DO, Sarmiento RAM, Torreglosa JCT, Rocha JF. Prevalence and risk factors of bovine viral diarrhea in Colombian cattle. Vet world. (2020) 13:1487–94. doi: 10.14202/vetworld.2020.1487-1494

PubMed Abstract | CrossRef Full Text | Google Scholar

81. Saa LR, Perea A, García-Bocanegra I, Arenas AJ, Jara DV, Ramos R, et al. Seroprevalence and risk factors associated with bovine viral diarrhea virus (BVDV) infection in non-vaccinated dairy and dual purpose cattle herds in Ecuador. Trop Anim Health Prod. (2012) 44:645–9. doi: 10.1007/s11250-011-9948-4

PubMed Abstract | CrossRef Full Text | Google Scholar

82. Rypuła K, Płoneczka-Janeczko K, Czopowicz M, Klimowicz-Bodys MD, Shabunin S, Siegwalt G. Occurrence of BVDV infection and the presence of potential risk factors in dairy cattle herds in Poland. Animals. (2020) 10:1–11. doi: 10.3390/ani10020230

PubMed Abstract | CrossRef Full Text | Google Scholar

83. Van Schaik G, Dijkhuizen AA, Huirne RBM, Schukken YH, Nielen M, Hage HJ. Risk factors for existence of Bovine Herpes Virus 1 antibodies on nonvaccinating Dutch dairy farms. Prev Vet Med. (1998) 34:125–36. doi: 10.1016/S0167-5877(97)00085-8

PubMed Abstract | CrossRef Full Text | Google Scholar

84. Covidence Systematic Review Software, Veritas Health Innovation. Melbourne, VIC. Available online at: www.covidence.org (accessed January 1, 2021).

85. Booth R, Brownlie J. Do Type 2 viruses pose a risk to bovine viral diarrhoea virus control in the UK? Livestock. (2015) 20:66–70. doi: 10.12968/live.2015.20.2.66

CrossRef Full Text | Google Scholar