Review question

In defining the research question, the PEO (Population, Exposures, Outcome) framework (Table 1) [56] was adopted and further refined using the FINER (Feasible, Interesting, Novel, Ethical, and Relevant) criteria for constructing a research question. Following a preliminary search, important knowledge gaps were identified, and the final research questions emerged as follows:

i)

Is there evidence that HAC (including proxies: reservoir/animal host sightings and density) increases due to land use change?

ii)

What are the anthropogenic activities associated with increased HAC?

iii)

What are the animal species associated with increased HAC?

The PEO framework will play a crucial role in organizing, focusing, and executing the review, ensuring it remains tightly aligned with the research questions by clearly defining the Population (human populations affected by land use changes), Exposure (land use changes like deforestation, urban development, habitat loss, and road building), and Outcomes (human-animal contact and related zoonotic diseases). This structured approach will guide the literature search, refine data extraction, and align findings with the research questions, enhancing the review's quality and clarity. It will also help ensure the data collected is relevant and comparable, making the review a valuable tool for understanding the complex interactions between land use change, human-animal contact, and zoonotic disease risk. Additionally, the PEO framework will support meta-analysis and narrative synthesis, providing a consistent basis for comparing studies and highlighting patterns or gaps in the literature.

Review objectives

The objective of the protocol is to establish a clear procedure for gathering information to explore the knowledge gaps and answer the research questions identified for the systematic review of the existing evidence on human animal contact, agricultural land use change and zoonotic disease risk. This includes the research questions that will be answered by the systematic review, a description of the systematic literature search process, criteria for study inclusion and exclusion, study quality ratings, data extraction, and data analysis from eligible studies.

Evidence gathering and study selection

To ensure the quality of the protocol, the review will be designed according to standard methods [57] and reported based on the Preferred Items for Systematic review and Meta-Analysis protocol (PRISMA) 2020 checklist.

Evidence gatheringSearch approach (Fig. 1)Fig. 1

Stage 1 – Building search strategy in Medline through PubMed and Embase. Initial preliminary search terms will be derived from search blocks; human animal Contact, zoonosis, disease risk, and land use change/ agricultural land use change. Here, search terms will be built to each search block from analysis of article keywords, titles, abstracts, indexed terms, and subject headings.

Stage 2 – Search in all databases will be conducted in the second stage of evidence collation. Here, eight (8) databases will be searched using search terms derived from the preliminary search and tailored to specific databases’ functionality. Databases will include Medline, Embase, Global Health, Web of Science, Scopus, AGRIS, Africa-Wide Info, and Global Index Medicus.

Stage 3 – Supplementary (snowballing) search will be conducted on references of those papers that fulfil the eligibility criteria to identify any additional relevant references. This will be subjected to the same screening and selection process as other retrieved articles.

Building search terms

key search terms will be selected to capture the broad scope of HAC, land use change, and zoonotic disease risk.

To guarantee a robust and comprehensive search strategy, various operators based on database features will be used to refine and retrieve relevant literature effectively. Boolean operators such as "AND" will be employed to combine key terms, ensuring that all specified criteria are met in the results, while "OR" will be used to include synonyms or related terms to broaden the search. To exclude irrelevant or unwanted terms, "NOT" will be applied. Proximity operators like "ADJ3" will be used to find terms within three words of each other, allowing for flexibility in how concepts are presented in the text. Additionally, wildcard operators, such as the asterisk (*) for multiple characters and the question mark (?) for single characters, will help capture variations of search terms, enhancing the breadth of the search. Truncation, using symbols like "$," will allow for the inclusion of different word endings or forms. Exact phrases will be searched using quotation marks to ensure that specific terms are found in the exact order required. Parentheses will be used to group terms and operators, controlling the order of operations within complex search queries. Lastly, field-specific operators will refine searches to specific parts of a record, such as titles, abstracts, ensuring that the most relevant literature is captured efficiently.

These terms include variations and combinations of keywords including:

Human-Animal Contact: ("human-animal interaction" OR "human-wildlife contact" OR "human-livestock interaction" OR "human-pet interaction").

Land Use Change: ("land use change" OR "deforestation" OR "urbanization" OR "agricultural expansion" OR "habitat fragmentation").

Zoonotic Disease Risk: ("zoonosis" OR "zoonotic disease" OR "spillover" OR "emerging infectious diseases").

Modifications of terms for Specific Databases (complete search strategy for all proposed databases are captured in supplementary material I).

Medline

Search Terms: Medical Subject Headings (MeSH) and free-text terms

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("infection* OR disease* adj3 (emergence OR risk OR threat OR transmission OR outbreak)")

Boolean Operators: "AND" for combining concepts, "OR" for including synonyms, "NOT" to exclude irrelevant terms

Modifiers: Filters for publication date, study design, language, and age groups

Embase

Search Terms: Emtree terms and free-text keywords

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("disease* adj3 (emergence OR spread OR exposure OR vulnerability OR transmission)")

Boolean Operators: "AND" for combining distinct concepts, "OR" for synonyms, "NOT" to filter out unrelated topics

Modifiers: Filters for geographic location, publication type, language, and study phase

Global health

Search Terms: Free-text terms and subject headings specific to global health contexts

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("emerging adj3 disease* OR communicable adj3 disease*")

Boolean Operators: "AND" for combining concepts, "OR" for including synonyms, "NEAR" to find words close together

Modifiers: Utilize regional filters, disease-specific categories, and global health indicators

Web of science

Search Terms: Topic and keyword searches

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("disease* adj3 (emergence OR threat OR transmission OR outbreak)")

Boolean Operators: "AND" to link different concepts, "OR" for alternative terms, "SAME" for search terms in the same sentence

Modifiers: Filters for study type, discipline, citation frequency, and geographical areas

Scopus

Search Terms: Article title, abstract, and keyword searches

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("emerging adj3 disease* OR disease* adj3 risk OR epidemic* OR pandemic*")

Boolean Operators: "AND" for connecting concepts, "OR" for synonyms, "WITHIN" for specifying proximity

Modifiers: Filters for subject area, document type, language, and research funding sources

AGRIS

Search Terms: Agriculture-related keywords and subject headings

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("agriculture adj5 (land use* OR change*)")

Boolean Operators: "AND" for combining different search concepts, "OR" for synonyms, "NOT" to eliminate non-agriculture topics

Modifiers: Filters for agricultural focus, language, and publication type

Africa-wide Info

Search Terms: Regional keywords and subject headings

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("disease* adj3 (emergence OR risk OR exposure OR transmission OR outbreak)")

Boolean Operators: "AND" for concept linking, "OR" for synonyms, "NEAR" for proximity searching

Modifiers: Regional focus filters, language, and study type

Global index medicus

Search Terms: Global health and zoonotic disease-related terms

Example Search:

("Human animal contact*" OR "Human animal interaction*" OR "Inter?spec* contact*" OR "Cross?spec* contact*") AND ("Agricultural land use change" OR "Land use change" OR "Land cover change") AND ("Zoonos*" OR "Zoonotic" OR "Spillover") AND ("communicable adj3 disease* OR disease* adj3 transmission OR pandemic OR risk")

Boolean Operators: "AND" for concept linking, "OR" for including alternatives, "NOT" to exclude unrelated topics

Modifiers: Global health focus, language, and publication type filters

Study selection (Fig. 2)Fig. 2

Inclusion criteria.

Studies on human animal contact.

Studies linking agriculture and transmission or emergence or spillover of zoonotic diseases.

Studies focusing on the mechanism, risk factors, and pathways of transmission of zoonotic infections from animals to humans will be considered.

Studies on biodiversity and human animal contact.

Included studies could be primary sources, case reports, observational studies, cross-sectional studies, cohort studies, case–control studies, and systemic review/meta-analyses.

No restriction on geographic area.

No restriction on the date of publication of an article.

Searches will be limited to peer-reviewed full text articles.

Exclusion criteria

Articles not focusing on human-animal contact.

Studies not in English language.

Conference papers, commentaries, opinion papers, abstracts, and editorials.

Articles focusing on molecular studies or pathophysiology of zoonotic diseases.

Articles not focusing on spillover or transmission of diseases from animals to humans.

Articles not focusing on land use and zoonotic infections.

Screening of articles

Articles will be exported to Endnote [58], where duplicates will be removed and exported to Rayyan [59], a web-based tool that aids the review screening process for articles, allowing collaborations between multiple reviewers to screen using the title and abstract of papers to ensure a critically appraised process with less bias. The screening will be conducted by two independent reviewers to ensure quality control. In the event two reviewers have a conflicting decision during the review process, a third reviewer will be consulted to resolve the discrepancies. Screening will be done in two phases, in the first phase, the reviewers will carry out “Title and abstract screening”. The references of the selected articles will be manually searched for relevant articles that will be subject to the same screening phase. Full-text articles will then be subjected to the second stage, “Full-text screening” using the same screening criteria for the previous selected articles.

Assessment of risk of bias

Articles included in the review will be subjected to quality assessment using study quality assessment methodology tools by two independent reviewers. Four (4) Risk of Bias (RoB) assessment tools will be applied in stages based on study design including: OHAT (Office for Health Assessment and Technology Risk of Bias Rating Tool for Human and Animal Studies) [60], CCS (Case Control Studies) [61], OCCSS (Observational Cohort and Cross- Sectional Studies), and CIS (Controlled interventional studies) [62]. These tools will be applied in two stages, 1.) during the assessment of individual articles and 2.) when assessing article results during extraction of data. This method follows the recommendation of the Cochrane collaboration [63]. The OHAT tool can be applied to six study designs including EA: Experimental Animal, HCT: Human Controlled Trial, Co: Cohort, CaCo: Case–Control, CrSe: Cross-sectional, and CaS: Case Series/Case report studies. It has eleven (11) risk of bias questions which are rated by selecting among 4 possible answers. These include (+ +) Definitely Low risk of bias, ( +) Probably Low risk of bias, (-) Probably High risk of bias, (–) Definitely High risk of bias. The system for answering each risk-of-bias question requires reviewers to choose between low and high risk of bias options. The questions will be applied based on the type of study article assessed. The OCCSS, CCS, and CIS tools have 14, 12 & 14 questions respectively to assess the overall quality of studies using various standard established criteria such as research questions, objectives, sample justification, exposures/outcomes validity and reliability etc. These questions will be answered with yes or no responses by reviewers and the overall quality rating will be denoted Good, Fair or Poor.

Data extraction (Table 2)

Data will be extracted by two independent reviewers using Google Forms designed to retrieve relevant information to answer the review questions.

Table 2 Data extraction formsProposed data analysisDescriptive analysis

The data analysis will be conducted to comprehensively address the research questions. Initially, the focus will be on descriptive statistics to provide a detailed characterization and summary of the data. This phase involves the systematic organization of the collected data to align with the PEO framework—Population, Exposure, Outcome. Each study's key variables, including the types and frequencies of human-animal contact (HAC), land use changes, and reported zoonotic diseases, will be meticulously cataloged. Descriptive statistics, such as measures of central tendency (mean, median) and dispersion (range, standard deviation), will be computed to offer an overview of the distribution of these variables across studies. To explore the relationships between categorical variables, such as land use changes and their association with specific zoonotic diseases, frequency distributions and cross-tabulations will be utilized. Visualization techniques will further elucidate the data, employing various graphical representations including bar charts, histograms, pie charts, and geographical maps. These visualizations will aid in depicting the spatial distribution of land use changes and HAC incidents, and in examining trends or correlations over time through line charts and scatter plots.

Meta-analysis

If data demonstrate sufficient homogeneity, a meta-analysis will be undertaken to aggregate and synthesize findings across studies. The decision to proceed with a meta-analysis will hinge on several critical factors. First, there must be a high degree of similarity in the populations, exposures, and outcomes of the included studies, which will be evaluated through qualitative comparisons and statistical tests for heterogeneity, such as Cochran’s Q test and the I2 statistic. Additionally, the feasibility of a meta-analysis depends on the availability of effect sizes or sufficient data to calculate them, reflecting the relationship between land use changes, HAC, and the risk of zoonotic disease transmission. Consistency in measurement methods across studies is also essential to ensure that results can be meaningfully synthesized. This encompasses uniformity in defining and reporting HAC, land use changes, and zoonotic diseases.

If these conditions are satisfied, the meta-analysis will involve extracting or calculating effect sizes from each study, which may include odds ratios, relative risks, or other relevant measures depending on the outcome variables. These effect sizes will be pooled using either a fixed-effect or random-effects model, based on the degree of heterogeneity observed. The selection of the model will be guided by statistical criteria and the specific characteristics of the data. An assessment of heterogeneity will be conducted through statistical tests, with subgroup analyses or meta-regression employed to explore potential sources of variability, such as geographic differences or types of land use change. Sensitivity analyses will test the robustness of the meta-analysis results, which may involve excluding studies with high risk of bias or applying alternative statistical models to evaluate result stability. Additionally, the potential for publication bias will be assessed using funnel plots and Egger’s test to ensure that the meta-analysis findings are not skewed by selective reporting. The results will be presented in both tabular and graphical formats, including forest plots to illustrate pooled effect sizes and confidence intervals.

If a meta-analysis is not feasible, a narrative synthesis will be conducted. This will be structured according to the PEO framework, focusing on the interplay between land use changes, HAC, and zoonotic diseases. The narrative synthesis will explore key patterns, identify gaps in the literature, and highlight areas for further research. This comprehensive approach, integrating descriptive and meta-analytic methods as appropriate, will ensure a thorough examination of the relationship between land use change, human-animal contact, and the risk of zoonotic diseases.

Reporting

Findings will be reported using the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA).