Written and maintained by CASRAI Editorial Board
Published · Last updated
On this page
CADIMA is a free, web-based, open-access evidence synthesis platform designed to manage, standardise, and document the complete lifecycle of systematic reviews and systematic maps. Jointly developed by Germany’s Julius Kühn-Institut and the Collaboration for Environmental Evidence, CADIMA provides academic teams with structured modules for automated reference deduplication, inter-rater consistency testing via Cohen’s kappa, dual-reviewer screening, bulk PDF management, and customisable data extraction without subscription fees or team size caps.
What CADIMA is and its core purpose
CADIMA (Central Database for Impact Assessment) was established to address a persistent vulnerability in academic research: the high operational friction and reporting variability that compromise large-scale evidence syntheses. Conducting a rigorous systematic review requires exhaustive search management, transparent article selection across multiple independent analysts, repeatable risk-of-bias judgments, and granular data extraction. When managed across disjointed spreadsheets and desktop citation libraries, review teams frequently experience version conflicts, undocumented exclusion decisions, and compromised audit trails.
CADIMA resolves these logistical barriers by providing a centralised, cloud-hosted workspace that mirrors international methodological frameworks. While originally created to standardise evidence collection for agricultural biosafety, plant breeding, and environmental risk assessments, CADIMA contains no disciplinary constraints. Research groups across conservation biology, veterinary medicine, environmental chemistry, public policy, and social sciences use the platform to execute both question-driven reviews and broad evidence mappings. By embedding workflow checkpoints directly into its database structure, CADIMA ensures that every decision—from search import through study synthesis—produces a permanent, defensible audit record.
Who maintains CADIMA and platform governance
CADIMA operates under public institutional governance rather than a commercial software-as-a-service model. The platform was created and continues to be actively maintained through an enduring institutional partnership between the Julius Kühn-Institut (JKI)—the Federal Research Centre for Cultivated Plants under Germany’s Federal Ministry of Food and Agriculture (BMEL)—and the Collaboration for Environmental Evidence (CEE). Early software conceptualisation and technical infrastructure development were funded through major European Union initiatives, notably the 7th Framework Programme projects GRACE (GMO Risk Assessment and Communication of Evidence) and G-TwYST (Genetically Modified Plants Two Year Safety Testing).
Because CADIMA is hosted directly on secure server infrastructure managed by the Julius Kühn-Institut in Germany, user data and project repositories operate under strict European General Data Protection Regulation (GDPR) compliance. This non-commercial institutional backing provides several critical structural guarantees for research teams:
- Perpetual open access: Unlike commercial platforms that gate collaborative seats, advanced extraction modules, or reference counts behind tiered subscriptions, all CADIMA capabilities remain freely accessible to academic, governmental, and independent researchers worldwide.
- Methodological alignment: CADIMA’s design committee works directly with CEE synthesis methodologists, ensuring the software adheres natively to CEE Guidelines for Systematic Reviews in Environmental Management and the Environmental Evidence journal reporting standards.
- Data integrity and exportability: Project administrators retain complete ownership of raw bibliographic records, screening allocations, reviewer scoring notes, and extracted matrices, which can be exported in open tabular formats at any stage of the investigation.
Step-by-step review workflow in CADIMA
CADIMA structures evidence synthesis projects into sequential, gate-checked stages that guide review teams through the requirements established by international synthesis standards. The platform interface guides project leads and reviewers through eight core operational steps:
- Project initialization and protocol specification: The review team establishes the review metadata within CADIMA, registering the study title, explicit research objectives, conceptual framework (such as PICO, PECO, or PCC), and the pre-registered protocol URL. Project managers configure administrative access, inviting co-investigators via email and designating permission levels.
- Search string definition and reference import: The team documents database-specific search strings, platform parameters, search dates, and indexing sources. Bibliographic export files are imported into CADIMA using standard RIS format (.txt or .ris files) exported from databases such as Web of Science, Scopus, CAB Abstracts, or PubMed.
- Creation of the merged reference list and automated deduplication: Once all reference files from disparate databases are uploaded, CADIMA merges the citations into a unified repository. The system runs an automated duplicate detection algorithm that evaluates author names, publication years, titles, and journal volume/issue metadata. Potential duplicate clusters are presented in a dedicated interface where administrators verify matched records before consolidating them.
- Pilot consistency screening and Cohen’s kappa calculation: To ensure uniform comprehension of eligibility criteria before mass screening begins, CADIMA provides a native consistency check module. A representative random sample (typically 5% to 10% of total citations) is allocated to multiple reviewers. CADIMA automatically computes Cohen’s kappa coefficient (κ) to measure inter-rater agreement. If agreement falls below the established methodological threshold (typically κ falls below 0.60), the team convenes to clarify ambiguous inclusion/exclusion definitions before re-testing.
- Title and abstract screening: The full deduplicated library enters primary study selection. Reviewers evaluate titles and abstracts independently against explicit eligibility criteria. CADIMA supports single-screening, dual-independent screening, or mixed allocations. When dual screening is active, decisions remain masked until both analysts have submitted ratings. A dedicated conflict-resolution dashboard flags discordant ratings (e.g., Include versus Exclude) for team consensus or principal investigator adjudication.
- Full-text retrieval and bulk PDF document management: For citations advancing beyond abstract screening, CADIMA facilitates full-text document acquisition. Project managers can execute bulk PDF uploads of up to 500 document files per batch. CADIMA links uploaded PDF files directly to their respective bibliographic records, allowing team members to review full manuscripts within an integrated reading pane. Full-text eligibility is assessed against detailed exclusion reasons, which are logged for standard reporting.
- Critical appraisal and risk-of-bias evaluation: Review teams configure customized critical appraisal instruments within CADIMA to assess study internal validity, reporting quality, and potential confounding biases. Reviewers rate specific methodological domains (such as selection bias, exposure assessment accuracy, outcome measurement fidelity, and statistical validity), recording qualitative justifications alongside categorical risk scores.
- Data extraction and offline spreadsheet synchronization: CADIMA provides flexible data extraction mechanisms tailored to either quantitative effect sizes or qualitative evidence topologies. Research teams can build structured extraction templates within the web interface or export blank coding matrices as spreadsheets. Reviewers can perform offline data extraction and re-import completed spreadsheets into CADIMA, where the platform validates cell entries against project schema rules.
Throughout this operational lifecycle, CADIMA continuously tracks study counts across all identification, screening, eligibility, and inclusion gates. This real-time audit ledger allows researchers to export structured flow counts directly compatible with the PRISMA flow diagram methodology and CEE ROSES standards.
Methodological strengths and practical limitations
Choosing CADIMA over commercial platforms involves balancing specific methodological protections against software usability and automation features. Understanding where CADIMA excels and where its boundaries lie allows project coordinators to match the tool to their institutional capacity and review architecture.
Key methodological strengths:
- Built-in statistical inter-rater agreement: Few evidence platforms natively calculate Cohen’s kappa within the screening interface. In most alternative tools, investigators must export screening decisions into external statistical software (such as R or SPSS) to calculate inter-rater reliability. CADIMA’s real-time calculation ensures teams verify screening consistency before conducting thousands of independent evaluations.
- Exemplary audit trail for environmental standards: CADIMA was built from the ground up to satisfy the rigorous documentation demands of the CEE and ROSES (RepOrting standards for Systematic Evidence Syntheses). Every exclusion decision at the full-text stage requires an explicit, codified rationale, preventing undocumented attrition bias.
- Zero financial barrier: Grant budgets frequently face constraints that prevent recurring per-review software expenditures. CADIMA eliminates commercial paywalls, allowing large international consortiums with dozens of student or volunteer coders to collaborate without seat licenses.
- Hybrid online/offline extraction flexibility: For fieldwork, low-bandwidth research environments, or extensive multi-analyst coding campaigns, CADIMA’s ability to download structured extraction templates and re-import validated spreadsheets offers operational resilience.
Practical limitations and boundaries:
- Absence of active machine learning prioritization: Unlike modern AI tools for systematic literature review such as Rayyan or ASReview, CADIMA does not incorporate active learning algorithms to re-rank unscreened abstracts by predicted relevance. Reviewers must process their allocated batches through traditional manual workflows.
- Functional, non-modernized user interface: CADIMA’s user interface prioritizes database stability and granular record tracking over aesthetic sleekness. Researchers accustomed to modern commercial software may find the menu navigation, table views, and configuration screens utilitarian and less intuitive.
- Batch upload ceiling: While CADIMA supports bulk PDF uploading, batches are capped at approximately 500 files per upload transaction. Reviews with thousands of retrieved full texts require multiple sequential batch uploads.
- No integrated statistical meta-analysis: CADIMA prepares, validates, and exports structured extraction matrices, but it does not execute quantitative statistical meta-analyses (such as pooled risk ratios, forest plots, or funnel plots). Quantitative pooling must be conducted in external packages such as R (metafor), RevMan, or Stata.
Systematic reviews versus systematic maps in CADIMA
A distinct advantage of CADIMA is its dual architecture supporting both traditional systematic reviews and systematic evidence maps. While many medical evidence tools assume a narrow, hypothesis-testing PICO structure culminating in meta-analysis, CADIMA natively accommodates the broader structural requirements of a mapping review.
A systematic map aims to collate, catalogue, and describe the existing state of evidence across a wide thematic area, identifying knowledge clusters, evidence gaps, and methodological trends rather than calculating an aggregate effect estimate. Within CADIMA, project leads can configure the review architecture to support mapping objectives by:
- Configuring broad multi-attribute coding schemas: Systematic maps capture extensive descriptive variables, such as geographic coordinates, study designs, targeted ecological biomes, intervention categories, and funding sources. CADIMA’s extraction builder allows complex categorical, hierarchical, and free-text fields without requiring numerical outcome measures.
- Customizing critical appraisal tiers: In mapping exercises, formal quantitative risk-of-bias tools may be superseded by study-validity filters or descriptive quality indicators. CADIMA allows administrators to adapt appraisal rubrics to capture structural study characteristics.
- Generating multi-dimensional evidence matrices: CADIMA exports rich tabular datasets formatted for downstream geographic mapping, bibliometric analysis, and interactive evidence atlas generation.
Comparing CADIMA with alternative evidence synthesis tools
Researchers evaluating systematic literature review tools must weigh trade-offs across cost, ease of use, team coordination, and methodological discipline. The comparison table below highlights key operational differences between CADIMA and four widely used evidence synthesis platforms:
| Platform | Developer / Governance | Pricing Model | Screening Automation / AI | Native Cohen’s Kappa | Primary Disciplinary Strength |
|---|---|---|---|---|---|
| CADIMA | Julius Kühn-Institut and CEE (Public) | Completely free (Open Access) | Deterministic deduplication; manual screening | Yes (automated pilot module) | Environmental sciences, agriculture, multidisciplinary mapping |
| Covidence | Veritas Health Innovation (Cochrane partner) | Commercial subscription (per review or institutional) | Automated deduplication, keyword highlighting | No (consensus and conflict flagging only) | Clinical medicine, healthcare, public health |
| Rayyan | Rayyan Systems Inc. (Commercial) | Freemium (tiered individual and team plans) | Active learning / 5-star relevance prediction | No (blind review percentage metrics) | Rapid multi-disciplinary title and abstract screening |
| EPPI-Reviewer | UCL EPPI Centre (University-based) | Subscription (monthly or annual fee) | Advanced text mining, machine learning priority screening | Yes (advanced statistical appraisal) | Complex mixed-methods reviews, social policy, living reviews |
| DistillerSR | DistillerSR Inc. (Commercial enterprise) | Commercial enterprise (high annual license) | Proprietary AI screening, automated extraction assistance | Yes (integrated inter-rater reliability metrics) | Regulatory compliance, pharmaceutical, medical device vigilance |
As summarized above, CADIMA remains the most robust zero-cost platform providing dedicated support for both full-text screening and customizable data extraction without restricting team size or project duration. For teams requiring deep machine learning prioritization during screening, pairing an initial screening pass in an active-learning tool with subsequent full-text extraction in CADIMA represents a viable, cost-effective hybrid strategy.
Collaboration workflows, user roles, and data security
Systematic reviews and evidence maps are inherently collaborative endeavors requiring coordinated effort across subject matter specialists, methodologists, information specialists, and student researchers. Managing contributor privileges effectively prevents accidental protocol deviations and maintains blind review conditions.
Within CADIMA, project access is structured through role-based permissions administered by the primary project manager:
- Project Administrator: Maintains complete administrative authority over project architecture, including importing search files, executing duplicate removal, modifying screening criteria, configuring critical appraisal and extraction templates, managing user allocations, and adjudicating disputed decisions.
- Reviewer / Screener: Granted access to assigned citation batches for title/abstract screening, full-text evaluation, or data extraction. Reviewers cannot modify project protocols, alter screening rubrics, or view blinded assessments submitted by co-reviewers until masking is officially lifted.
- Viewer / Auditor: Read-only access enabling external stakeholders, peer reviewers, or advisory board members to examine review progress, inspect the audit trail, and verify screening decisions without modifying underlying database records.
From a data governance perspective, all project data is securely archived on German federal servers at the Julius Kühn-Institut. Communications and data transfers utilize encrypted HTTPS protocols. Because the hosting institution is an official federal scientific agency, researchers working with proprietary, sensitive agricultural, or preliminary research data benefit from strict European data protection standards and freedom from commercial data mining or resale practices.
The data table below compares the technical specifications, institutional hosting environments, format compatibilities, and governance frameworks of prominent evidence synthesis platforms.
Frequently asked questions
What is CADIMA used for?
CADIMA is used to design, conduct, and document systematic reviews and systematic evidence maps across scientific disciplines. It provides dedicated tools for importing literature search results, removing duplicate records, performing pilot consistency tests, screening titles, abstracts, and full texts, executing critical appraisal, and extracting research data into structured formats.
Is CADIMA free to use?
Yes. CADIMA is completely free and open-access for all registered researchers. There are no subscription fees, tiered feature restrictions, limits on the number of projects created, or caps on the number of collaborating team members.
Who developed and funds CADIMA?
CADIMA was developed through a partnership between Germany’s Julius Kühn-Institut (Federal Research Centre for Cultivated Plants) and the Collaboration for Environmental Evidence (CEE). Initial development was funded by EU 7th Framework Programme research grants (including the GRACE project), with ongoing infrastructure maintenance funded by the German Federal Ministry of Food and Agriculture (BMEL).
What file formats does CADIMA support for importing literature?
CADIMA supports standard RIS bibliographic files (.ris or .txt) exported from major reference managers and indexing databases, including EndNote, Web of Science, Scopus, PubMed, and CAB Abstracts. Search results from different platforms can be uploaded independently and merged within CADIMA.
How does CADIMA handle duplicate references?
CADIMA includes an automated deduplication module that scans merged search files for matching authors, publication years, titles, and journal volume or issue numbers. The system groups suspected duplicates and provides a side-by-side comparison screen where project administrators can verify matches before finalizing the deduplicated list.
Does CADIMA support team screening and conflict resolution?
Yes. CADIMA allows administrators to allocate citations among multiple reviewers for single or dual-independent screening. Under dual screening, reviewer ratings remain blinded until both analysts complete their evaluation. Discordant ratings are automatically flagged in a dedicated conflict resolution module for team discussion or administrator adjudication.
How does CADIMA calculate inter-rater reliability?
CADIMA includes a native consistency check module that calculates Cohen’s kappa coefficient (κ) based on a randomly sampled batch of studies screened by two or more reviewers. The calculated kappa score helps teams verify whether their application of eligibility criteria meets accepted reliability benchmarks (typically κ ≥ 0.60) before launching full-scale screening.
Can CADIMA perform offline data extraction?
Yes. In addition to its online extraction forms, CADIMA allows researchers to download project extraction templates as spreadsheet files. Reviewers can extract data offline in standard spreadsheet software and re-upload the completed files back into CADIMA, where data entries are verified against the project schema.
Sources
- CADIMA Official Evidence Synthesis Platform, Julius Kühn-Institut and Collaboration for Environmental Evidence, https://www.cadima.info, checked 2026-11-15
- Kohl, C., McIntosh, E. J., Unger, S., Haddaway, N. R., Kecke, S., Schiemann, J., and Wilhelm, R. (2018). Online tools supporting the conduct and reporting of systematic reviews and systematic maps: a case study on CADIMA and review of existing tools. Environmental Evidence, 7(1), 12. https://doi.org/10.1186/s13750-018-0115-5, checked 2026-11-15
- Collaboration for Environmental Evidence (CEE). Guidelines and Standards for Evidence Synthesis in Environmental Management, https://www.environmentalevidence.org, checked 2026-11-15
- Julius Kühn-Institut (JKI), Federal Research Centre for Cultivated Plants, Biosafety in Plant Biotechnology Research Group, https://www.julius-kuehn.de, checked 2026-11-15
- Haddaway, N. R., Macura, B., Whaley, P., and Pullin, A. S. (2018). ROSES RepOrting standards for Systematic Evidence Syntheses: pro forma, flow-diagram and descriptive summary of the plan and completion of collaborative evidence syntheses. Environmental Evidence, 7(1), 7. https://doi.org/10.1186/s13750-018-0121-7, checked 2026-11-15
Data: Evidence Synthesis Platform Governance and Feature Matrix
Comparative evaluation of evidence synthesis and systematic review platforms, detailing developer governance, access models, inter-rater reliability automation, data extraction modes, and reporting alignments compiled from official platform documentation and published methodology literature.
Period covered: 2018/2026. Coverage: Global.
| Platform | Governing Body | Access Model | Primary Synthesis Target | Inter-Rater Reliability Automation | Offline Extraction Mode | PDF Batch Upload Ceiling (files) | Reporting Framework Alignment |
|---|---|---|---|---|---|---|---|
| CADIMA | Julius Kühn-Institut & CEE | Free Open Access | Systematic reviews & systematic maps | Automated Cohen's kappa calculation | Supported (Excel/CSV sync) | 500 | CEE Guidelines & ROSES / PRISMA |
| Covidence | Veritas Health Innovation | Paid subscription (per review or institutional) | Intervention systematic reviews | Conflict flagging only (no kappa) | Not supported (web interface only) | 0 | Cochrane Handbook & PRISMA |
| Rayyan | Rayyan Systems Inc. | Freemium tiered plans | Title and abstract screening | Percentage agreement (no kappa) | Limited (export only) | 0 | PRISMA |
| EPPI-Reviewer | UCL Social Research Institute | Paid monthly or annual license | Mixed methods & complex reviews | Advanced inter-rater statistics | Not supported (web interface only) | 0 | EPPI Centre Standards & PRISMA |
| DistillerSR | DistillerSR Inc. | Commercial enterprise subscription | Regulatory & clinical reviews | Automated Cohen kappa and agreement | Supported (audit export/import) | 0 | PRISMA & Regulatory GxP |
Sources
- Online tools supporting the conduct and reporting of systematic reviews and systematic maps: a case study on CADIMA and review of existing tools (accessed 2026-11-15)
- CADIMA Official Evidence Synthesis Platform Documentation (accessed 2026-11-15)
- Guidelines and Standards for Evidence Synthesis in Environmental Management (accessed 2026-11-15)
Compiled by CASRAI from the sources above. Free to reuse under CC BY 4.0 with attribution.







