Protocol for developing a Database of Zoonotic disease Research in India (DoZooRI).

INTRODUCTION: Zoonotic and emerging infectious diseases (EIDs) represent a public health threat that has been acknowledged only recently although they have been on the rise for the past several decades. On an average, every year since the Second World War, one pathogen has emerged or re-emerged on a global scale. Low/middle-income countries such as India bear a significant burden of zoonotic and EIDs. We propose that the creation of a database of published, peer-reviewed research will open up avenues for evidence-based policymaking for targeted prevention and control of zoonoses. METHODS AND ANALYSIS: A large-scale systematic mapping of the published peer-reviewed research conducted in India will be undertaken. All published research will be included in the database, without any prejudice for quality screening, to broaden the scope of included studies. Structured search strategies will be developed for priority zoonotic diseases (leptospirosis, rabies, anthrax, brucellosis, cysticercosis, salmonellosis, bovine tuberculosis, Japanese encephalitis and rickettsial infections), and multiple databases will be searched for studies conducted in India. The database will be managed and hosted on a cloud-based platform called Rayyan. Individual studies will be tagged based on key preidentified parameters (disease, study design, study type, location, randomisation status and interventions, host involvement and others, as applicable). ETHICS AND DISSEMINATION: The database will incorporate already published studies, obviating the need for additional ethical clearances. The database will be made available online, and in collaboration with multisectoral teams, domains of enquiries will be identified and subsequent research questions will be raised. The database will be queried for these and resulting evidence will be analysed and published in peer-reviewed journals. © Article author(s) (or their employer(s) unless otherwise stated in the text of the article) 2017. All rights reserved. No commercial use is permitted unless otherwise expressly granted.

The strength of this approach is that in the absence of rigorous surveillance programme and methods to track zoonotic diseases, this approach provides an alternate method to develop a continuously updated database to query zoonotic diseases and their determinants.

The main limitation of the approach is that it is largely dependent on published data, which may be of heterogenous quality, and could also result in duplicate entries. However, we envisage that the issue of publication bias might be resolved to a large extent in future iterations of the database as collaborators would contribute by adding more citations from the literature to make it a more comprehensive repository of zoonotic disease research in India.

Introduction

Zoonoses, diseases and infections that are naturally transmitted between humans and other vertebrate animals, have been recognised since the beginning of history, but their public health importance has gained increased visibility in the last few decades.1 Global analyses of emerging infectious diseases (EIDs) between 1940 and 2004 identified that 60.3% of them were zoonotic and were increasing rapidly over time.2 A substantial portion of the global burden of endemic zoonoses is from India which is one of the several ‘EID hotspots’.3 4 India is experiencing rapid urbanisation, industrialisation, increasing incomes and changing food preferences raising demands for animal source proteins, leading to a larger section of the workforce being in close contact with livestock, including poultry, making a wider section of the population extremely susceptible to zoonotic diseases. Owing to the public health importance of zoonotic diseases in India, a National Standing Committee on Zoonoses was formed in 2007.5 6 Lately, a Centre for Zoonoses Research has been proposed to be set up to harness intersectoral collaborative research using the One Health approach.7 However, not much is known about the status of zoonotic disease research and several key aspects related to it, including several aspects of clinical presentation, diagnoses, management and epidemiology in the Indian context.8 There is also a need to synthesise evidence about zoonoses in India to guide further research, set research and public health priorities and streamline disease control efforts in an evidence-informed manner.9 10 Further, given the resource restrictions and multiple competing priorities that need to be addressed by policymakers and funders alike, the synthesised evidence could help in making evidence-informed decisions to guide resource allocation. This paper reports the protocol for development of a Database of Zoonotic Disease Research in India (DoZooRI), which intends to address these gaps.

Rationale for development of DoZooRI

We envision DoZooRI as a multidisciplinary research database which would embed all zoonoses-related research in India. DoZooRI will be made accessible to clinicians, veterinarians, researchers, public health professionals, programme managers and policymakers on request. Citations within the research database would be tagged on several carefully chosen key parameters that would enable rapid retrieval of relevant research from a wide variety of sources. This would provide critical support for clinical or public health practice, including, but not limited to the fields of human, veterinary and environmental health, by facilitating evidence syntheses or planning of future research in line with research gaps. The multidisciplinary nature of the database, in fact, provides an impetus to this purpose. This is essential in the context of India, where access to electronic databases as well as trained health information specialists with the capability to develop search strategies is severely limited.11 DoZooRI will also address concerns about wasteful research, owing to new research not being based on pre-existing evidence and without their need being vindicated by existing systematic reviews demonstrating critical knowledge gaps.12 13 Mapping zoonotic disease research and conduct of further evidence syntheses through the DoZooRI will facilitate better understanding of knowledge gaps, which is critical for disease control efforts, policymaking, public health planning and informing public health and research priorities. Further, the database will enable users to analyse research capacity across institutions in India.

DoZooRI is being developed as a strategic tool to support various activities of the India Research Initiative on Peri-Urban Human-Animal-Environment Interface, which is housed at the Public Health Foundation of India, and aims to conduct and support multidisciplinary research, build workforce capacity across sectors and conduct evidence syntheses and policy advocacy.

Implications and possible collaborations

The creation of the DoZooRI presents an innovative approach to systematic evidence synthesis. It should encourage researchers to collaborate, help to identify and dissolve the compartments which have restricted research efforts to sectoral silos and encourage truly transdisciplinary collaborations to flourish.14–16 Using the database, researchers can map zoonotic diseases research; describe the characteristics of interventional research on zoonotic diseases in India; describe the epidemiology of zoonotic diseases, including differences by location, geographic terrain or other determinants; identify the research–policy disconnect and inform on realignment of research priorities in zoonotic disease prevention and control in India; understand institutional capacity to conduct zoonotic research in India and address any other issues, utilising the extended tagging system. The database could potentially be developed into a dynamic repository of the body of work being done in India with respect to zoonotic diseases. Further, if collaborations with techno-medical entrepreneurs can be leveraged, this can be developed into a platform which is updated in real time and allows auto-tagging of published research for the purpose of evidence syntheses. Additionally, this protocol for the development of the database should be of interest to professionals involved with human, animal and environmental health alike, since it outlines the development of a database that proposes curating the first of its kind of evidence base for zoonotic diseases in India. This is likely to be of interest to Indian and international scholars alike, given the global health security threat posed by zoonotic and emerging diseases.

The database is going to provide a step towards closing the evidence–policy gap in zoonotic disease prevention and control programmes by providing the policymakers, programme managers or any other stakeholders access to a searchable, tagged database of existing evidence. This opens avenues for data mining and conducting multiple rapid reviews or systematic reviews to assess the state of the published evidence. The process will bring together the tenets of open science and evidence-based policymaking to strengthen the clinical, veterinarian and public health response to zoonotic disease prevention and control in India.

Objective

To develop a database of publications resulting from research conducted on a set of priority zoonotic diseases in India and tag them in a manner so as to facilitate further evidence syntheses and landscaping of zoonotic disease research in India.

Methods and analyses

A large-scale systematic mapping of important zoonotic disease research in India will be conducted. Systematic mapping covers a broad cross-section of literature without a narrowly defined research question, as is done in a systematic review and is ideal for the research objective.17–21 No quality assessment is done unlike a systematic review but all research on zoonotic diseases is described based on preset characteristics as well as other characteristics identified iteratively as the mapping progresses. The database can, however, serve as a tool for conducting systematic reviews and rapid reviews on focussed research questions on zoonotic disease in India.

Criteria for inclusion of studies in the database

Definition of zoonotic diseases for the database

We will include research on the following nine priority zoonotic diseases from India, as defined by the study authors22:

Leptospirosis

Rabies

Anthrax

Brucellosis

Cysticercosis

Salmonellosis

Bovine tuberculosis

Japanese encephalitis

Rickettsia infections

We will not include research which has been conducted on participants of Indian origin living abroad or on those who have travelled from India to foreign countries.

Multicentric studies containing Indian participants or samples and systematic reviews including at least one study which meets the population criteria, will be included in the database. For studies conducted before 1947 and where it is not possible to differentiate it with results from post-1947 boundaries by reading the full text of the articles, the boundaries of the undivided British India will be considered.

Study design

We will include research using any kind of study design except commentaries, editorials, news item, historical articles, meeting notes and narrative reviews. This includes, but is not limited to, animal studies, observational studies, experimental studies, case reports or case series, outbreak investigation, systematic reviews, practice guidelines and qualitative research.

Time limits

We will include research articles irrespective of their year of publication.

Language restriction

We will include only articles published in English in the database. Exclusion of non-English studies introduces a very small risk of publication bias since English is the dominant language for health research in India.

Publication status

We will include all articles irrespective of publication status.

Search strategy for electronic databases

Since the database intends to identify all zoonotic disease research in India, a sensitive search strategy was designed in PubMed using relevant free-text keywords and Medical Subject Heading (MeSH) heading (online supplementary appendix 1). The search strategy was designed iteratively with the intention to achieve maximal sensitivity. This implies that we added additional free-text keywords and MeSH headings when the search did not yield enough results. Additional free-text keywords were developed by adding synonyms or alternative terms. We then adapted the search strategy for use in other electronic databases, namely CINAHL (EBSCOHost) (online supplementary appendix 2) and Global Health (EBSCOHost) (online supplementary appendix 3). Future versions of DoZooRI might include studies for more electronic databases and additional sources for grey literature. We envision that DoZooRI collaborators who conduct evidence syntheses by using the database will add grey literature citations within the database through expert contact, reference checking of articles from the database, searching clinical trial registries and registers of observational studies and systematic reviews.

Screening of articles

We will screen all citations retrieved independently by at least two researchers based on titles and abstracts (if available) and mark them as ‘retrieve’ (eligible or unclear eligibility) or ‘do not retrieve’ (not eligible). For all citations marked ‘retrieve’, full texts will be acquired and final decision on eligibility will be taken independently by two researchers. In case of any discrepancies that are not resolved through discussion between the two reviewers (SB and PC), consensus will be achieved by consultation with a third expert researcher (MK). Reasons for exclusion at full-text phase will be documented.

Data management

We will manage data and host the database in a cloud-based platform called Rayyan (http://rayyan.qcri.org/). The platform allows simultaneous screening as well as tagging of citations in a blinded fashion by multiple researchers. Rayyan also allows multiple tagging of each citation. We chose Rayyan over other options because of its two-fold advantages—it is free to use and it has an inbuilt artificial platform where it provides prediction for includes and excludes based on the choices marked by researchers manually through a colour-coded five-star system by text mining the records. An independent evaluation of Rayyan found that after screening of three-quarters of included citations 98% of articles included were correctly predicted.23 This would be an essential tool to rapidly allow the DoZooRI administrators to screen and mark includes and excludes when new studies are added as and when the database is updated from time to time. Rayyan also detects duplicates automatically.

Database tagging

We will develop a tagging system iteratively for several key parameters, which would enable users to rapidly identify and retrieve articles to answer research, policy or clinical questions of their interest. Tagging will be done independently by at least two researchers (SB and PC) and discrepancies will be solved through consensus decision within the research team. We have predetermined tags for the following key parameters:

Name(s) of zoonotic disease—from among the nine priority zoonotic diseases.

Study design—classification as per standard terminologies (eg, case report or case series, case—control, cohort, cross-sectional, retrospective study, ecological study, clinical trial, systematic review, qualitative research and in vitro studies).

Special study type tags—for example, GENE for genetic or genomic studies, EPIDEM for a study on any aspect of disease epidemiology, OUTBREAK for a study on any aspect of disease outbreak, DIAG for a study on diagnostic modality for a zoonotic disease and ECONOM on health economics-related studies.

For epidemiological studies—name of the state(s) of India where the study was conducted.

For clinical trials—randomisation status and name of intervention.

Based on host involved: human, cattle, wild animals, multihost, invertebrate vectors and so on.

Ethics and dissemination

We prepared DoZooRI from already conducted research publications and did not involve any animal or human subjects. As such, there are no ethical issues involved in the creation of the database or its subsequent use for evidence syntheses.

The findings from any research done by using DoZooRI will be published in peer-reviewed journals. The results of any such research on the database will also be shared with key stakeholders, including health professionals, researchers and policymakers in the form of policy briefs.