Background. Cystic echinococcosis (CE), caused by the tapeworm species, Echinococcus granulosus sensu stricto (G1), is one of many primary neglected zoonoses worldwide. Within endemic developing countries, CE has multiple effects on animal and human health and well-being. To address such effects, veterinary and human medical sector collaboration on prevention program delivery is essential. To begin preliminary evaluations of county specific prevention programs, a critically appraised topic (CAT) was conducted. It sought to answer: What impact do CE prevention programs have on human and animal disease prevalence, in populations living in endemic developing countries within Africa, Central Asia, and South America? Methodology. The aim was to assess the ability of prevention and control program outputs to produce measurable differences in health, social, and economic outcomes (e.g., improved access to medical services, positive behavioral change, or reduced treatment costs, respectively). Included articles were obtained using predefined inclusion/exclusion criteria from the four databases (CAB Abstracts and Global Health; the National Library of Medicine (PubMed); ScienceDirect; and WHO Institutional Repository of Information Sharing (IRIS)). The articles were appraised using three checklists: the Royal College of Veterinary Surgeons (RCVS), the Critical Appraisals Skills Programme (CASP), and the Joanna Briggs Institute checklists. Results. Ten articles were selected. Geographically, 20% of studies were conducted in South America, 30% in Africa, and 50% in Central Asia. For definitive hosts, dogs, CoproELISA antigen testing, before and after Praziquantel (PZQ) de-worming, was a primary focus. For humans, who are intermediate hosts (IH), disease surveillance methods, namely ultrasound (US), were commonly assessed. Whilst for sheep, also acting as IH, disease prevention methods, such as the EG95 livestock vaccine and de-worming farm dogs, were evaluated. Common to all studies were issues of program sustainability, in terms of regular human US screening, dog de-worming, and annual sheep vaccination. This was attributed to transient and remote human or animal populations; limited access to adequate roads or hospitals; few skilled health workers or veterinarians; an over-reliance on communities to administer preventatives; and limited resources. Conclusion. Despite variations in result validity and collection periods, useful comparisons of CE endemic countries produced key research and program recommendations. Future research recommendations included testing the significance of multiple program outcomes in relation to prevalence (e.g., the social outcome: behavioral change), further research on the impact of livestock vaccinations, and the CE transmission role of waterways and sanitation. Program recommendations included calculating and distinguishing between stray versus owned dog populations; formal representation of internal and external stakeholder interests through institutional organization; establishing sustainable guidelines around the frequency of PZQ and vaccination administration; improved veterinary-human medical training and resource sharing; and combined prevention methods and multiple canine disease management.

1. Introduction

Globally each year, cystic echinococcosis (CE) causes an estimated 19, 300 deaths, and a loss of US$3 billion to treatment costs, especially within the livestock industry [1]. Within the multi complex species (Echinococcus granulosus sensu lato), E. granulosus sensu stricto (G1) causes the highest prevalence and widest global distribution of human cases ([2], p. 2). Dogs are the definitive hosts, infected by ingesting prey or raw meat containing metacestode cysts (larvae) [3]. Although dogs remain mostly asymptomatic, they excrete proglottids or infective eggs that are ingested by multiple mammalian intermediate hosts (IH), such as sheep, cattle, wildlife, and humans [4]. IH can remain asymptomatic for years, until growing cysts rupture or cause complications in adjacent organs ([5], p.12).

For many living in low socioeconomic and/or remote areas within endemic countries, recommended medical diagnostic methods and preventatives, such as Praziquantel (PZQ) (5 mg/kg) for dogs [6] or the EG95 livestock vaccine [2, 7], are not always readily accessible. Access to social services, such as healthcare, can act as a development capability [8, 9]. That is, it presents an opportunity to enhance one’s well-being or acts as a constraint, which contributes to vicious cycles of poverty [1012]. Cost-effective preventatives, such as public health education programs, and veterinary and human medical collaboration [1317] have been suggested to overcome resource constraints. Public health education campaigns and multi-stakeholder collaboration are essential to producing social outcomes, such as behavioural change, that can reduce disease incidence. Examples of positive behaviours include: not feeding dogs raw meat, hand washing after handling dogs, and preventing canine access to common livestock, wildlife, and human environments [1820].

CE control and prevention programs have been broadly identified in endemic regions of Africa, South America, and Central Asia [16]. Since 2019, the progress is evident in countries, like Mongolia, where a multi-stakeholder program provided PZQ de-worming for dogs and human ultrasound (US) screening [21]. However, few studies have analyzed the effectiveness of prevention or control programs, in terms of specific outcomes. Most focus on human medical and surgical treatments or conducting cohort or case control studies to identify common risk factors. Thus, the aim of this critical appraisal was to conduct a program impact assessment to provide evidence-based recommendations for current public and animal health programs.

2. Methodology

Four online databases were searched (Table 1). These included the following: CAB Abstracts and Global Health (1973-present), the National Library of Medicine (PubMed), ScienceDirect, and the WHO Institutional Repository of Information Sharing (IRIS) [22] and “Echinococcosis” webpage [11]. PubMed and CAB abstracts produce comprehensive veterinary and human medical research [2325], while the ScienceDirect database produces full text, peer reviewed literature on healthcare [26]. Additionally, the WHO resources were searched, as it is one of the leading international organizations conducting CE programs in endemic regions.

2.1. Screening Process

Within the ScienceDirect database, the journals were selected from two “Subject Areas”: “Veterinary Science and Veterinary Medicine” and “Medicine and Dentistry.” Further refinement was achieved by applying inclusion/exclusion based upon relevance to the topics of “parasitology,” “public health,” “zoonoses,” and “developing countries” [26]. Additionally, within CAB abstracts, the word “veterinary” and phrases “one health” and “animal health” originally resulted in the exclusion of several relevant articles. This became evident when conducting repeat searches. Thus, these terms were excluded in the final search (Table 1) to improve sensitivity, as lower specificity could be easily corrected. For example, predefined inclusion/exclusion criteria (Table 2), synonyms, Boolean operators, parentheses, and truncation (CAB Abstracts only) were used to increase result specificity.

Furthermore, articles were refined using an ordered selection process (Table 3). The fifth exclusion criteria “non-English articles” (Table 2) presents a methodological limitation, as sensitivity is reduced. However, the aim was to avoid timely and possible inaccurate article translations. Additionally, across all databases, the words “diagnostic” and “tests” were included in search terms; as many prevention programs encompassed diagnostic tests to monitor program impact. However, studies [27] that conducted generalized evaluations of individual treatment or diagnostic tools were excluded [28].

Finally, articles were sorted based upon relevance. The article title and abstract were reviewed for key words e.g., “echinococcus,” “prevention,” and/or “control.” If relevance could not be determined, the full texts, namely, the discussion and methodology, were reviewed. Integrated prevention program assessments of multiple diseases [3133] were excluded. Individual disease analysis is important, as each disease has its own complex transmission pathway and diagnostics. However, there was one exception [34], because CE program outputs and impacts were readily distinguished from other diseases.

Despite the described replicable criteria, methodological limitations may arise from inaccurate subjective application of inclusion/exclusion criteria or study design categorization. For example, controlled clinical trials were excluded because of a high number of false positive search results. Nevertheless, PubMed still produced a non-randomized controlled trial that fit the inclusion criteria. Differences in categorization of this study existed between this CAT, article authors, and the database.

2.2. Geographical Selection

Africa, South America, and Central Asia are listed as CE endemic regions [16], consisting of low to middle income countries [3537]. The included studies within these regions were mostly in countries that ranked below 50/188 on the United Nations Development Programme (UNDP) multidimensional poverty index [38]. For example, Kyrgyzstan (142) and Kenya (143) are two of the lowest ranked [38], but countries like China were included, as it represents 40% of the world’s CE caused disability-adjusted life years (DALYs) ([39], p.138).

2.3. Critical Appraisal Method

Applying a triangulation method, which is not reliant upon one set of appraisal criteria, reduced the probability of information bias. Checklists included templates from the Royal College of Veterinary Surgeons [24], the Critical Appraisals Skills Programme [40] for health professionals, and the Joanna Briggs Institute [41] for health and medical sciences.

3. Results

3.1. Summary of Evidence

A total of ten studies (Tables 413) were selected using predefined selection criteria (Table 2) and the screening process outlined (Table 3). Most studies (70%) were non-randomized, non-blinded studies. This included four observational studies (cross-sectional studies: two randomized and two non-randomized), one experimental (non-randomized controlled trial), three quasi-experimental before/after studies (two non-randomized and one randomized), and two qualitative studies. Cross-sectional and before/after studies predominated (7/10 studies), as they produced direct evaluations of program outcomes and impact. There were variations in study designs, sample size, and collection periods, which complicated direct comparisons. Nevertheless, programs were all situated in endemic low to middle income countries and produced useful comparative findings. Geographically, two studies (20%) were conducted in Argentina (South America), one (10%) in Kenya, and two (20%) in Morocco (Africa). Five (50%) were conducted in Central Asia, specifically in China, Mongolia, and Kyrgyzstan.

3.2. Study Designs

This CAT focused on evaluating program outcomes and impacts, versus disease causality and risk factors, commonly tested by several excluded cohort and case control studies. Compared to cross-sectional studies and case control studies, cohort studies are relatively expensive, which is often not suitable in low socioeconomic contexts. Nevertheless, cohort studies are superior in terms of monitoring disease incidence rates and controlling against confounding demographic or signalment variations, by consistent respondent follow-up. Whilst case control studies are usually more cost effective, within a low socioeconomic context, access to consistent human and animal health data may be limited. Respondant follow-up was a challenge across most included studies, as sample human or animal populations were often transient [42, 43, 51] and resided in remote areas [15, 34, 43], with limited access to adequate roads, hospitals, or resources [2, 6]. These challenges make selection of a control group practically difficult and may account for the few experimental studies encountered.

In addition, ethical issues arise if an animal or human, with little access to services, is denied preventative treatment to be assigned to a control group. Indeed, Yang et al. [43] concluded that it would be ethically negligent to select a control site, after the government identified that CE was endemic in Northwest China ([43], p.357). Thus, in China [15, 43], Mongolia [15], Argentina [2, 6], and Kenya [42], many included studies utilized non-randomized convenience sampling. The majority used cross-sectional study designs, as they are relatively cost effective and practical at sampling populations at different points in time.

Furthermore, quasi-experimental studies were suitable in comparing a pre- and post-program impact. The sample population is ideally exposed to the same program output to reduce confounding bias and increase validity. Ensuring sample group similarity was an evident limitation of some studies [2, 34, 45]. For example, in Van Kesteren et al.’s [34] study, pre-intervention prevalence rates for dogs were only calculated in 4/10 communities. For the remaining six communities, post-intervention data was not comparable. Finally, a limitation common to qualitative, or mixed methods studies, was an absence of detail about questionnaires, survey or interview question type, or delivery mode [6, 15, 45].

3.3. Sample Size

Four studies did not calculate sample size and/or include confidence intervals (CI) [39, 42, 43, 46], and some provided limited information about sample collection methods [15, 39]. This may be attributed to issues of accessing remote and/or transient communities. In Yu et al.’s [39] study, all data was centrally controlled by the National Ministry of Health, China, and no information about sample collection was provided. Consideration of how political or organizational agendas align with program outcomes is important, due to the centralized control of statistical data on CE prevalence.

Nevertheless, four studies did calculate sample size [2, 6, 34, 45]. For one before/after study [2], each expected proportion (prevalence rate) for humans, dogs, and sheep was treated independently. However, calculation based upon paired data (discordant pairs) would have been suitable, as one group of animals or humans was paired to two different prevalence values, at the start and end of the program. In addition, studies with limited population size [42, 43] could have utilized a finite population sample calculation [52].

Notably, a major issue with sample size calculation was captured in the 8/10 studies that did not specify an estimated stray dog population or sample size. This meant that only owned dogs were treated [2, 6, 39, 42] or generalized terms, such as “free roaming” dogs [34] or “dog management” [15], were used. Only three studies [43, 45, 46] clearly distinguished and treated stray dogs with PZQ. However, administration methods were not standardized, and population size was also not calculated by Yang et al. [43].

Additionally, one study [45] utilized convenience sampling of dogs caught by the local dog catcher. Although ethics approval was obtained, the number () of dogs euthanised for necropsy was not justified using a pre-defined sample size calculation. This is essential to minimise and validate the number of stray dogs necessary to test for significant differences in prevalence. Convenience sampling also reduced external validity, as selected stray dogs may have only been representative of a small area.

Furthermore, one before/after study [34] revealed the benefit of Lot Quality Assurance Sampling (LQAS) to evaluate the quality of health care programs [53]. Compared to other sampling methods, LQAS allows randomized analysis of a small community sample size, but not individual village level analysis. For countries with small, but remote or widely distributed communities, utilizing a sample size, such as nineteen, is ideal and has been proven to minimize type a and b errors [34, 54]. The LQAS enables identification of community areas that fall below average in achieving a specific program target. Indeed, Van Kesteren et al. [34] traced a transition from poor PZQ dosing coverage in 8/10 villages to improvements in reaching dosing targets (p.3). Nevertheless, LQAS may be logistically costly if researchers are required to travel to multiple program areas.

3.4. Program Outcomes and Impact

It is helpful to distinguish between program outputs, outcomes, and impacts when evaluating programs. Outputs can be defined as “the goods or services produced by programs…while outcomes are defined as the impact on social, economic, or other indicators arising from the delivery of outputs” [55]. An adapted definition of impact is when a program outcome “helps solve the problem that inspired actors to create” ([56], p.460) it.

Studies focused on health outputs and outcomes (e.g., access to medical or veterinary services) and their impact, in terms of disease prevalence. One study assessed economic outcomes (e.g., accumulated financial costs), and a few studies [34, 39] evaluated social outcomes (e.g., positive behavioral change in response to public health campaigns). While a core impact of prevention programs is to resolve the issue of rising disease prevalence, most studies failed to test the significance of prevalence changes in correlation to multiple social, health, and economic outputs and associated outcomes [2, 6, 34, 42, 43].

More specifically, studies identified health education campaigns as program outputs [2, 6, 34, 39, 42, 43, 50], but only three [2, 34, 42] measured campaign outcomes (e.g., social outcome: behavioral change). For example, Arezo et al. [6] identified sanitary education (e.g., adequate disposal of infected offal) as a program output, but it was not analyzed with respect to specific outcomes. Van Kesteren et al. [34] went further to measure behavior as a social outcome, in terms of dog owner PZQ administration. While semi-structured questionnaires measured CE disease knowledge, it was not linked to a specific program output (e.g., public health education) or tested for significant correlations to specific program impacts (e.g., decreased CE prevalence).

Additionally, a study in China [45] linked communities’ level of knowledge to achieving high PZQ dosing rates in the previous year (p.5). However, significant correlations of de-worming behavior to a specific program output (e.g., public health campaign) or prevalence, were not tested. Thus, the final impact could not be concluded. Similarly, Yu et al. [39] measured behavioral change, by measuring pre and post-program de-worming coverage, but did not test for significant correlations to a program output or impact.

Solomon et al.’s [42] study went further, in terms of evaluating a specific health education campaign, which was targeted to women in Kenya, who spent most of the time at home with dogs. Prevalence reduction was attributed to health education producing positive behavioral change, such as appropriate offal disposal. Given that education programs targeted women and statistically significant changes in prevalence were linked to gender (Pearson χ2, ; ordinal Somers’ tests, ) ([42], p.591), this correlation seems valid. However, confounding variables, such as literacy rates, previous reductions in dog population size, and increased PZQ treatment, introduced confounding bias.

3.5. Confounding Variables

Numerous studies did not measure behavior as an outcome of public health education outputs, which essentially introduced confounding bias. Consistent positive behavioral change is essential to minimize CE transmission. The behavior can potentially enhance or constrain the effects of animal or human health outcomes and, ultimately, the impact of disease prevalence. For example, feeding dogs infected offal is an established transmission pathway [3] that can constrain the effects of de-worming dogs. Indeed, Van Kesteren et al. [45] concluded that health education had the potential to decrease CoproELISA prevalence, by inciting positive behavioral change, such as increasing PZQ administration. Whilst Larrieu et al. [2] concluded that programs with a combined education component enhanced the positive effects of canine anthelmintic treatment and sheep vaccination coverage ([2], p.5).

In addition to behavioral change, only one study [46] accounted for seasonal climatic variations. It is essential, as at 4 °C, E. granulosus eggs have a lifespan of ≥300 days compared to 2–14 days at 37-39°C ([3], p.438). Indeed, Amarir et al. [46] reported that calendar time and location had significant effects () on CE prevalence in stray and owned dogs in Morocco (p.440).

Finally, there were evident disparities in skill levels when conducting diagnostic tests or administering preventive treatments, such as PZQ [6, 34, 42, 43, 46]. Van Kesteren et al. [34] acknowledged the reality that leaving dog owners to self-monitor de-worming does not guarantee that recommended guidelines are followed. Across all studies, there was no detail of ongoing support or training of assigned community members to program tasks, which may have limited program impact.

3.6. Diagnostic Tests
3.6.1. Canine Definitive Host: Diagnostics, Coproantigen ELISA Prevails

Despite potential cross reactions with Taenia hydatigena, Arezo et al. [6] concluded that Coproantigen ELISA tests were superior when calculating CE prevalence rates. Justifications were based upon the higher sensitivity (78%−100%) and specificity (85%) of CoproELISA, compared to arecoline purgation tests, and CoproELISA showing similar trends to confirmation tests (PCR or Western Blot (WB) ([6], p.5). Indeed, Larrieu et al. [2] reported that the calculated prevalence rates from arecoline tests were not statistically significant , compared to CoproELISA tests (, ). Previous research has revealed that when the prevalence remains high, CoproELISA sensitivity, alone, may be used for accurate CE diagnosis [47]. Nevertheless, arecoline was utilized, as both a program output (e.g., a treatment or diagnostic method) and a research method [2, 42, 43, 46].

3.6.2. Human Intermediate Hosts: Diagnostic Ultrasound and Treatment

Assessing the impact of program surveillance, using US, was the primary focus of the four studies that measured human CE prevalence [2, 6, 39, 42]. In contrast to dogs, it was generally concluded that CoproELISA serology was less sensitive than US, although US had limited sensitivity in detecting pulmonary cysts [6, 42]. For example, 198 CE cases were US identified compared to 76 using serology ELISA ([42], p.588). Larrieu et al. [2] substituted serology (double diffusion 5, ELISA) tests for US, due to higher sensitivity.

Surveillance measures, such as US, are essential to identifying rising incidence or prevalence rates for endemic diseases, like CE. In CE endemic countries, such as Mongolia, when programs focused more on human surgical treatment than preventative dog management, they have resulted in under reporting and under diagnosis of CE cases ([15], p.64). While complete CE eradication is difficult in endemic countries, an over emphasis on post-infection control and treatment measures may inevitably lead to missed opportunities for early disease prevention. This may lead to increased demand and costs for surgical or medical treatments.

Additionally, one study acknowledged patient hesitation to undergo surgical treatment [42]. However, no study investigated the post-operative impact of surgical treatments on human quality of life. Many remote farming populations, reliant upon physical labor for their livelihood, had little ongoing access to medical care [2, 6, 7]. This would likely result in higher post-operative complication rates. Thus, initiating early prevention methods has potential benefits of reducing prevalence rates and associated treatment costs and/or medical complications.

3.6.3. Program Sustainability

A major finding across program evaluations was the unsustainability of six weekly PZQ treatments. Larrieu et al. [2] explained that programs using dog de-worming alone, often failed globally due to logistical constraints of sustaining 100% coverage; up to eight times per year, in remote areas (p.6). In Kyrgyzstan, it was concluded that six weekly PZQ intervals were not practical, due to funding and human resource constraints ([34], pp. 9, 16). Two studies recommended three to four monthly intervals to reduce canine and livestock prevalence rates to ≤1% within 10–15 years ([45], p. 6, [34]). Nearby, in northwest China, six weekly PZQ treatments were also unsustainable, as there were issues obtaining dogs’ weight, and with dosing logistics (e.g., dogs disliked taste, funding, remoteness, and skilled worker availability). Moving to Northern African, in Morocco, it was concluded that only two monthly PZQ de-worming intervals effectively controlled infective egg shedding in stray and owned dogs ([46], p. 441).

Furthermore, Yu et al. [39] attributed failed prevalence reduction in China, and the autonomous regions of Mongolia and Tibet, to the unsustainable de-worming of domestic canines, and controlling “wild canines” (p.2), such as foxes. Although studies [34, 42, 45, 50] identified the sylvatic cycle, no program outputs were discussed, apart from surveying community knowledge of wildlife transmission pathways [45].

Finally, to assess CE prevalence rates in sheep, postmortems within slaughterhouses [2, 43], serology ELISA, and WB were conducted [2]. Sustainability issues were identified in remote communities in Argentina ([2], p.5) and China ([45], p.6), in relation to EG95 sheep vaccinations. In Argentina, the issues were attributed to difficulties accessing remote areas, funding, and few skilled health or veterinary workers [14]. In China, statistically significant increases in infection (IF) rates among older sheep (>4 years, ), compared to younger sheep (<1 years, ) ([43], pp.357-8), were mostly attributed to unsustainable de-worming intervals, practices (e.g., burying dog feces), and health education delivery.

To address sustainability and a lag in vaccination flock effects, Larrieu et al. [2] and Van Kesteren et al. [45] suggested a combined program, which includes sheep vaccination, canine PQZ de-worming, and health education. Qian et al. [15] also recommended integrating CE with other neglected canine zoonotic diseases, such as rabies, to improve efficiency and reduce costs. Although excluded from this review, integrated zoonotic disease programs have potential cost-effective benefits, in terms of access to multiple health technologies ([7, 57], p.18, [58]). Indeed, it has been calculated that it would cost 30% more per dog treated separately for rabies, cystic echinococcosis, and visceral leishmaniasis, compared to an integrated program ([31], p.7).

4. Discussion and Recommendations

4.1. Measure Multiple Program Outcomes

Cross-disciplinary research that focuses on measuring multiple social, health, and economic program outputs and outcomes, in correlation to changes in disease prevalence, is essential. Most studies concentrated on measuring disease prevalence without obtaining an understanding of the specific program outputs and outcomes that caused changes. Without this understanding, future programs may fail to reproduce successful outputs or improve upon existing ones.

Failure to assess the link between social outcomes, such as positive behavioral change, with disease prevalence may be a disciplinary issue, as behavioral studies are often confined to the psychological and social sciences. Measuring behavior, pre- and post-health education campaigns is essential. However, consistently measuring sanitary practices, such as hand hygiene, may be difficult in remote communities. Additionally, identifying if access to clean water is a location specific constraint is essential. This would also highlight an area in need of program development and funding.

4.2. Regular Training: Standardized Administration and Measurement of Program Outputs (PZQ de-worming)

Stratified sampling and, if funding permitted, engaging an external statistician to analyze data may have improved the methodological reliability of studies. For example, separating dogs based upon owner versus health worker PZQ administration would control for confounding bias and increase the validity of reported changes in disease prevalence. To illustrate potential confounding effects, dog owners reported PZQ dosing in the previous four months, but over 50% of dogs tested positive for CE and 13.7% of dog owners could not recall the last de-worming ([45], p.3). In Yang et al.’s [43] study, one village resident was assigned to de-worm dogs and deliver health education, which resulted in difficulties sustaining program measures. Similarly, two other studies identified that both health workers and dog owners provided sanitary education and administered PZQ ([2], p. 6 [6]). Finally, Solomon et al. [42] specified that a health education campaign was initially delivered by organizational officers, but community members were subsequently trained (p.588).

Common to all these studies was that resident skill set level was unclear and ongoing supportive training was not regularly provided. Variations in skill set compromised program sustainability and study reliability, due to the increased likelihood of incorrect PZQ administration, irregular dosing intervals, dog owner recall bias, and potential inaccuracies in health information. Additionally, an absence of regular training may extend to professional program workers. Qian et al. [15] revealed that 75.9% of surveyed respondents from the WHO Mongolia Office, Mongolian government sectors, local hospitals, veterinary institutes, and laboratories reported not receiving CE training in the last 5 years (p.63). Thus, all included programs would benefit from ongoing training support for both local communities and organizational program workers.

4.3. Distinguish Between Stray and Owned Dog Populations

The OIE recommends distinguishing between owned and stray dogs to accurately calculate and trace population size [59]. Although the role of stray dogs in disease transmission was considered [2, 15, 42, 43], 70% of studies did not clearly make this distinction and/or administer PZQ to stray dogs. Only 20% calculated stray dog population size [45, 46]. The importance of this distinction was captured in Morocco, where photographic records tracked the number of stray dogs, who were 14 times as likely to be infected with E. granulosus compared to owned dogs (odds ratio = 14, 95% CI: 6–30; ) ([46], p.439).

While the population of owned dogs is more readily calculated, using registration records or household surveys, measuring stray dog populations is essential to minimizing CE transmission. Roaming stray dogs may access common livestock, wildlife, and human environments [1820]. A single dog can be the source of infection for 30,000 ha ([3], p. 438). Although labor-intensive, the OIE [59] recommends using a refined marking method or a wildlife biology, mark-recapture method, to calculate stray dog populations. The marking method entails accounting for daily differences in dogs’ distribution (e.g., variations in weather, access to food, shelter, and human activity). Dogs are temporarily captured and marked using a standardized method (e.g., a distinctive collar, paint smudge, or ear tag). After a few days, plotting the daily number of marked dogs against the accumulated total of marked dogs produces a representative value for population size. The second wildlife biology method entails initially marking and releasing dogs within a defined area. The number of marked and unmarked dogs is then calculated by distant re-observation of the same area. To calculate a total population estimate, the number of dogs initially marked and released is multiplied by the total number of re-observed dogs [59]. This is all divided by the number of marked dogs identified during re-observation. Notably, there is an increased risk of zoonotic disease transmission to researchers, and ethical concerns, in terms of the stress caused during the capture and marking process.

4.4. Coproantigen ELISA Tests: An Efficient All Rounder for Canine Prevalence

As identified, Coproantigen ELISA was commonly used to diagnose prevalence in both stray and domestic dogs [2, 6, 34, 45]. It has reported advantages, such as the ability to detect antigen 5-10 days post-infection and treatment, when >100 parasites are present ([3], p. 437 [60]). In the context of endemic low to middle income countries, this test is suitable in terms of achieving adequate sensitivity and affordability. In addition, CoproPCR [45, 46] was utilized to calculate canine CE prevalence, as a standalone verification test or sometimes coupled with WB [6] or arecoline tests [2]. Although more costly, and limited to research, CoproPCR tests offer a higher sensitivity and, unlike CoproELISA, can differentiate taeniid spp. from E. granulosus or E. multilocularis species ([3], p. 437).

Furthermore, arecoline purgation tests were only utilized by four studies [2, 42, 43, 46]. In addition to a comparatively lower sensitivity [6], arecoline tests can be labor-intensive, present a zoonotic risk, and cause “adverse reactions (vomiting, diarrhoea, hypersalivation)” ([3], p.437 [60]). Amarir et al. [46] fed owned and sedated stray dogs, with arecoline hydrobromide (4 mg/kg; 2 mg/kg second dose) to induce defecation and egg expulsion ([46], p.438). As signalment was not recorded, ethical issues arise, because potential adverse effects may result from contraindicated use in pregnant, young, or elderly dogs ([3], p.437). Necropsy of dogs’ small intestine is considered the most accurate diagnostic method among CoproELISA, CoproPCR, and arecoline purgation tests ([3], p. 437). However, across all studies, CoproELISA was relatively precise, economically practical, and an ethically sound option for CE diagnosis in stray and owned dogs.

4.5. Livestock Intermediate Hosts: More Studies on Vaccination Efficacy

Only two studies [2, 43] evaluated programs in reference to livestock vaccinations within endemic countries. Future longitudinal research is necessary to clearly establish the effectiveness of vaccination programs, in terms of multiple outcomes and significant differences in prevalence, not only in sheep, but other susceptible species, such as goats and cattle. However, relative cost constraints and the cited [2, 45] unsustainability of booster and annual vaccinations may account for the few livestock specific studies identified.

4.6. Improve Human and Animal Health Data Collection to Monitor Program Efficacy

As discussed, cohort studies are advantageous when it comes to consistent follow-up and the production of data on disease incidence rate. However, within rural transient populations, and low socioeconomic contexts, such studies may not be practical. Nevertheless, prevalence could be monitored over time by training community members to conduct simple data collection. This could encompass recording dates, the number of cases and clinical signs, on supplied template forms. Specifically, local human and veterinary medical clinic, and abattoir workers, could be trained to recognize key CE clinical signs (e.g., hepatic or pulmonary cysts identified during postmortems or meat inspections, respectively).

4.7. Multi-stakeholder Analysis: Interdisciplinary Research, Resource Sharing, and Formal Organization

The OIE advises that “feedback from the local community” and “relevant professionals (e.g., veterinarians, medical doctors, law enforcement agencies, educators)” [59] is essential to producing program indicators that reflect multi-stakeholder interests. Saadi et al. [50] and Qian et al. [15] conducted multi-stakeholder analyses, using qualitative field research in Morocco, China, and Mongolia. Qian et al. [15] conducted in depth interviews with representatives from the WHO Mongolia Office, Mongolian government sectors, local hospitals, veterinary institutes, and laboratories. Participants identified the benefit of bilateral China-Mongolia cooperation, in terms of joint research and training. To optimize this benefit, Qian et al. [15] suggested interdisciplinary collaboration between the fields of medicine, veterinary science, parasitology, epidemiology, and government departments. Indeed, a key WHO recommendation is adopting a “One Health” approach that entails medical and veterinary collaboration [16]. Saadi et al.’s [50] study also concluded that stakeholder relationships promote the sharing and development of knowledge, resources, and new technologies (e.g., diagnostics or treatment methods) (p.6). Thus, sustained stakeholder collaboration proves an important means to ensuring program sustainability.

More specifically, acknowledging the interest of multiple state and non-state stakeholders has been viewed as important in addressing a “democratic deficit” ([61], p.353, [62], p. 778-9) within self-regulatory governance models. This “deficit” is linked to actors governing themselves without equally representing the interests of public, private, and civil society actors. Indeed, in Morocco, Saadi et al. [50] highlighted different levels of power and interest. Internal government stakeholders wielded the most power but considered CE programs a lower priority when compared to external actors (e.g., WHO, OIE, physicians or veterinarians) who possessed less political agency. This web of stakeholders transcends animal and human health spheres and highlights the difficulty in representing multiple interests.

A reported neglect of stakeholder interests ([50], p.5) represents why measuring social outcomes is essential to program analysis. A primary social outcome indirectly identified by three studies ([39], p.146; [42, 43]) was cultural competency. Local cultural and religious beliefs may not always align with program measures. For example, Yang et al. [43] explained that within ethnic communities in North-Western China, Buddhist religion forbids the killing of any animals (p.358), which complicates dog population control measures. Indeed, Yu et al.’s [39] study, in China, Tibet, and Mongolia, also concluded that cultural acceptance of roaming stray dogs, over population control, is common ([39], p.146).

Participatory versus top-down programs, which enable open communication, are essential to fostering mutual respect between medical and veterinary practitioners with communities. Achieving this entails avoiding an over-reliance on divisional dichotomies (e.g., developed/developing, core/periphery, and modern/traditional) [63] to explain the world through one dominant western lens [64, 65]. This simply reproduces a hegemonic form of medical knowledge that may disregard local knowledge. Local cultural beliefs and practices regarding dogs’ roles in communities will continue to shape how people respond to prevention measures. Indeed, for people of Turkana, Kenya, dogs played multiple roles, which included acting as cattle rustlers, protectors against wildlife, as pets, and family members [42]. Thus, achieving mutual understanding ultimately increases the chance of positive behavioral change, which can reduce CE transmission.

To formally represent multiple stakeholders, Saadi et al. [50] recommended creating a national central office, which contains representative stakeholders who prioritize CE programs. However, most internal stakeholders who wielded relatively higher power considered CE programs a low priority. Thus, including both external and internal state actors is recommended to overcome these power dynamics. For example, a veterinarian could represent animal health interests; a human medical physician for public health interests; a government official for state interests; a representative for breeders, production animal, and slaughterhouse worker interests; and a community member for cultural interests.

5. Conclusion

To conclude, ten studies situated across three major CE endemic regions (Central Asia, Africa, and South America) were selected using pre-defined inclusion/exclusion criteria from the four databases. Common research limitations included no sample size calculation and numerous confounding variables, which limited result validity. One major confounder was the absence of standardized PZQ administration methods for stray and domestic dogs. Nevertheless, the studies produced useful comparisons, in terms of program barriers faced in remote, low to middle income countries. Generally, more research and/or programs are recommended within African and South American regions.

More broadly, future research and program development are essential, in terms of focusing more on prevention versus diagnostics and treatment. Future research recommendations included the following: measuring the effects of program outputs, in terms of multiple program outcomes, especially the social outcome of behavioral change. Going further, to test these outcomes for significant correlation to CE prevalence is essential. Additionally, research on livestock prevention measures to clearly establish the practicality and benefit of program outputs, distinguish between stray versus owned dogs to calculate population sizes and clearly target program outputs, and identify the transmission role of waterways and sanitation, is essential.

Finally, the key program recommendations include the following: regular local community training to deliver sustainable program outputs (e.g., PZQ), and conduct data collection to monitor CE prevalence; organized representation of multi-stakeholder interests; clearer or standardized guidelines around PZQ and livestock vaccination administration frequency; programs that encompass multiple prevention methods (e.g., dog de-worming, public health education, and sheep vaccination) and integrated canine disease management; and enhanced veterinary–human medical training and resource sharing to improve program sustainability.

Data Availability

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Conflicts of Interest

No conflict of interest to declare.


As this was a Critically Appraised Topic (CAT) review, no university funding was provided/necessary. However, The University of Sydney provided funding for publication costs.


The author thanks Professor David Emery (Professor of Veterinary Parasitology, The University of Sydney, NSW) for his patient and open supervision support and feedback.

Supplementary Materials

Appendix A (Table A1): exclusion/inclusion criteria applied to 34 Journals within the ScienceDirect database (2015-2021), across two relevant animal and human health subject areas: “Veterinary Science and Veterinary Medicine” and “Medicine and Dentistry.” (Supplementary Materials)