# Resource Sharing in Global Research Collaborations: Ethical Guidelines Resource sharing ethics in global research asks who gives, who gains, and who decides when samples, data, equipment, funding, and expertise move between research partners in different countries. The single most useful distinction is between sharing that builds capacity on both sides and sharing that quietly transfers value from the lower-resource partner to the better-resourced one. A workable rule of thumb: before any resource crosses a border, all parties should agree in writing on who owns it, who benefits from it, and what happens when the collaboration ends. This guide walks through the main sharing models, the ethical criteria that separate good agreements from bad ones, worked scenarios drawn from real international collaborations, and the mistakes that most often damage partnerships. ## Why Sharing Raises Distinct Ethical Questions Sharing is not automatically ethical. A refrigerated courier box of serum samples, a shared genome dataset, a donated surgical system, or a training workshop can each produce benefit or harm depending on how the arrangement is designed. One recent argument from a workshop at the 2022 Global Forum on Bioethics in Research in Cape Town is that ethical conduct of research is the responsibility of every stakeholder in the research ecosystem, from funders and governments to institutions, ethics committees, and individual research teams, and that ethical scrutiny should be distributed rather than concentrated in a single review step [1]. That framing matters for resource sharing because the ethical load cannot be dumped on one party. A collaborator who ships specimens abroad has obligations. So does the collaborator who receives them. Responsible conduct of research, the umbrella term for honesty, accountability, fair credit, and care in professional practice, applies to sharing decisions just as it applies to authorship and [data management](/blog/guides/data-management-basics-principles-processes-and-best-practices). A materially unfair sharing agreement is a responsible conduct problem even when every experiment inside it is technically flawless. ## What Gets Shared in Global Collaborations International collaborations share six broad categories of resource, and each carries its own ethical risks. 1. **Biological materials.** Tissue, serum, DNA, isolates, cell lines, and environmental samples. Key risks are consent scope, secondary use, and commercialization of materials donated by communities that will never access the resulting products. 2. **Data.** Clinical records, genomic sequences, imaging, and registry data. Key risks are re-identification, data sovereignty, and unequal analytical capacity to use the data once shared. 3. **Equipment and consumables.** Sequencers, reagents, single-use surgical devices, and software licenses. Key risks are cost inequity across regions and maintenance burdens left behind after a project ends. 4. **Funding.** Grants, subawards, and in-kind support. Key risks are indirect-cost disputes, currency and transfer barriers, and donor-driven priority setting. 5. **Knowledge and training.** Curricula, protocols, mentoring, and statistical or writing support. Key risks are one-way training flows that drain local talent and curricula designed without local input. 6. **Labor and credit.** Sample collection, recruitment, follow-up, and data entry performed by local staff. Key risks are unfair authorship and acknowledgment practices. The ethical structure of a collaboration depends mostly on which categories are moving and in which direction. Sharing data one way while training flows the other way is a very different arrangement from one-way extraction of samples with no return. ## Resource Sharing Ethics in Global Research: The Main Options Compared | Sharing model | What is shared | Who controls terms | Pros | Cons | Best fit | |--|--|--|--|--|--| | Open commons | Data, protocols, code, preprints released openly | Global community, minimal restriction | Maximizes reuse, broadens access for low-resource groups, accelerates science | Consent limits, privacy risk, no guaranteed benefit flow back to contributors | Non-identifiable data, methods, software | | Consortium co-ownership | Samples, data, and infrastructure governed jointly | All partner institutions through a written charter | Balanced power, shared credit, durable governance | Slow to establish, requires sustained administration | Long-term multi-country programs | | Cost-recovery exchange | Equipment time, reagents, services shared at documented cost | Providing institution, within an agreed schedule | Sustainable, avoids hidden subsidy, transparent | Fees can exclude low-resource partners if not subsidized | Core facilities, device evaluation networks | | Donation and mission-based provision | Equipment, consumables, surgical or clinical visits | Donor or visiting team, ideally with host input | Rapidly fills acute gaps, visible impact | Dependency risk, maintenance burden, host priorities can be overridden | Time-limited gaps with a genuine handover plan | | Capacity-building partnership | Training, mentoring, curriculum, joint projects | Co-designed by both sides | Builds durable local capability, aligns with fair partnership norms | Slow, requires long funding horizons, results lag | Regions with limited research infrastructure | | Licensing and material transfer | Samples or tools exchanged under contracts such as material transfer agreements | Legal offices of each institution | Clear ownership, protects both parties, enables commercialization | Legal cost, delay, can entrench inequity if terms are one-sided | Transfer of valuable biological materials or proprietary tools | No single model is best in every situation. The right choice depends on the value of the resource, the capacity gap between partners, and whether the collaboration is a one-time exchange or a decade-long program. ### Open Commons Sharing In an open commons model, partners release data, code, or protocols for unrestricted reuse, often through repositories or preprint servers. The ethical appeal is obvious. A laboratory in a country with limited journal subscriptions gains the same access to methods as a well-funded group in North America or Europe. The limits are equally real. Human data cannot always be opened without violating consent terms given to participants. An open release also transfers value outward with no mechanism to return benefit to the community that generated the data. Openness is an ethical default for methods and de-identified aggregate data. It is not an ethical default for sensitive human data or community-derived samples. ### Consortium Co-Ownership A consortium treats shared resources as jointly owned assets with a written governance charter. Partners agree on decision rights, credit, data access tiers, and wind-down procedures before the first sample ships. This model is the strongest for equity because no single institution controls the resource. The cost is administrative weight. Consortia require meetings, memoranda of understanding, and someone willing to track compliance. For a two-laboratory sample exchange, this is overkill. For a ten-country cohort following patients for a decade, it is the only model that survives staff turnover. ### Cost-Recovery Exchange In a cost-recovery model, a provider shares equipment time, reagents, or services and charges the documented cost of providing them. This is standard for sequencing cores, imaging facilities, and analytical services. The ethical question is what "cost" means in a global context. The same single-use device can carry very different prices in different health systems. An international dataset compiled by the EAU Endourology Group across 40 countries found a median purchase price of €676.7 (interquartile range €579 to €975) for single-use flexible ureteroscopes, ranging from €310 in Romania to €1,638 in Cyprus, with comparable spread for flexible navigable access sheaths at a median of €180 [2]. When acquisition costs vary this widely by country, a flat cost-recovery fee can quietly price partners out. Fair cost-sharing arrangements either subsidize lower-resource partners explicitly or tier fees by documented ability to pay. ### Donation and Mission-Based Provision Donation models move equipment, supplies, or visiting clinical expertise from better-resourced to less-resourced settings. They can close real gaps. Experience from spine surgery missions in low-resource settings shows that success depends on financial planning and transparency, equipment logistics, operating room throughput, and, just as critically, sustained partnership with host institutions and local teams, with patient selection carrying particular ethical weight under resource constraints [3]. The recurring failure mode is dependency. A donated device without a service contract, spare parts, or trained operators becomes shelf art within a year. A visiting surgical team that performs cases without training local staff leaves no capacity behind. Donation is ethical when it includes a genuine handover plan and host leadership in patient selection and priorities. ### Capacity-Building Partnerships Capacity building treats knowledge as the shared resource. Instead of shipping samples out, the collaboration brings skills in. A series of five sequential workshops in clinical research for pediatric oncology, running from 2020 to 2025, combined online modules with on-site project-based learning and covered study design, data registries, statistics, [research ethics](/blog/guides/research-ethics-fundamentals-a-practical-guide-for-early-career-scientists), and scientific writing, drawing an average of 35 applicants per workshop from 17 countries across seven disciplines [4]. Environmental scans of advocacy and policy leadership training similarly emphasize that international academic partnerships need a shared, rigorous evidence base to inform collaborative curriculum co-design across contexts [5]. The ethical strength of this model is durability. Skills stay with people. The weaknesses are pace and funding. Capacity building needs multi-year horizons, and short grant cycles push funders toward measurable, one-time deliverables instead. ### Licensing and Material Transfer When biological materials or proprietary tools cross borders, institutions typically use a material transfer agreement, a contract defining ownership, permitted use, publication rights, and liability. This model protects both parties and is often legally required. The ethical caution is asymmetric bargaining power. A large institution can dictate terms that give the providing low-resource institution no share of downstream commercial value and no role in publications. Fair agreements negotiate publication rights, benefit sharing, and use restrictions before the material moves, not after. ## The Criteria That Decide Real Choices ![Decision path asking whether data can be opened, whether the program is long term, and whether materials are valuable, leading to a sharing model.](/article-images/illustrations/resource-sharing-ethics-in-global-research-sharing-model-choice.webp) When you evaluate any sharing arrangement, five criteria do most of the work. **Equity of benefit.** Does value flow in both directions? A Lancet Oncology Commission on equitable cancer genomics found that access to genomics and precision oncology remains profoundly unequal across and within countries, with genomic studies and clinical trials concentrated in high-income settings and among populations of predominantly European ancestry, and that fragmented data ecosystems restrict evidence generation and knowledge sharing [6]. A sharing agreement that adds samples or data to a high-resource consortium while returning nothing locally replicates exactly the pattern that Commission describes. **Governance and accountability.** Who decides secondary use, recontact, and withdrawal? Ethics oversight works best when it is distributed across the ecosystem rather than treated as a single checkpoint, and when researchers and ethics committees build collaborative, non-adversarial relationships recognizing mutual responsibilities [1]. **Data and sample sovereignty.** The community or country that contributed a resource retains a legitimate interest in how it is used. Sharing data does not dissolve the originating community's stake. **Sustainability.** Will the shared resource still function in three years? Equipment without maintenance plans, training without follow-up, and datasets without curation funding all fail on this criterion. **Credit and authorship.** The people who collected samples, recruited participants, and entered data deserve authorship or explicit acknowledgment proportional to their contribution. This is a core responsible conduct of research obligation, not a courtesy. ## Worked Scenarios ### Scenario 1: Sharing Genomic Data Across Borders A genomics consortium wants tumor and germline sequences from hospitals in several countries. The high-income partners have sequencing capacity and analysts. The lower-income partners have patients and tissue. The ethical trap here is documented in the precision oncology literature. Genomic studies concentrate in high-income settings, data ecosystems are fragmented, and populations of non-European ancestry are underrepresented [6]. A fair consortium therefore negotiates co-analysis rather than one-way transfer. Local partners retain the right to analyze their own contributed data, joint publications include local analysts in substantive roles, and any commercial discoveries trigger a pre-agreed benefit sharing term. If artificial intelligence tools are developed on the pooled data, the fairness problem compounds, because a systematic review of hospital AI research found that 83 percent of included studies originated from high-income countries, even though 83 percent of those studies addressed at least one World Health Organization ethical AI principle in depth, most often inclusiveness and equity [7]. Principles stated in papers are not the same as equity built into consortium charters. ### Scenario 2: Sharing Equipment Knowledge and Costs A surgical research group wants to compare device costs across member countries to inform procurement. Here, transparent data sharing itself is the resource. The endourology collaboration that pooled purchase prices for single-use flexible ureteroscopes across 40 countries produced per-procedure cost estimates ranging from €415 to €1,881, information that no single country could have assembled alone [2]. The ethical rule in this scenario is symmetry. If partner institutions contribute proprietary pricing data, they should get co-authorship and access to the pooled analysis. A group that harvests cost data from partners and publishes without them has converted a collaboration into extraction. ### Scenario 3: Sharing Training and Implementation Know-How A research group plans to introduce a new diagnostic and treatment service into primary care facilities in a lower-income country. An implementation study in Ethiopia for cutaneous leishmaniasis thermotherapy illustrates the co-design approach. The study, anchored in the Medical Research Council framework for complex interventions, follows a three-phase design of co-design and contextual adaptation, pilot testing with iterative optimization, and operational scale-up, using a multi-tiered task-sharing strategy with primary care staff and Health Extension Workers [8]. Global initiatives such as the Global Initiative for Obstructive Sleep Apnea apply the same logic, with strategic pillars covering workforce capacity building, context-sensitive care pathways, and task-sharing approaches [9]. The ethically sound version of this scenario shares the intervention blueprint, the training curriculum, and the analytic plan, not just the final publication. Participatory designs go further. A research-community partnership in Melbourne co-developed a survey with 57 participants and then ran workshops with eight community members to turn findings into resources, validating results through partner meetings [10]. Community partners who shape the research are owed more than an acknowledgment line. ### Scenario 4: Multicenter Trials Across Five Countries A cluster randomized trial of hip fracture care across India, Nepal, the Philippines, Thailand, and Vietnam embedded a process evaluation at 10 sites, combining dashboard data with roughly 100 interviews and focus group discussions to understand what works, for whom, and why [11]. Sharing an intervention protocol across five countries obliges the coordinating team to share adaptation authority too. Local teams that must adapt the model to their context need the standing to propose changes, and the process evaluation should capture their experience, not only patient outcomes. ## Common Mix-Ups **Confusing donation with partnership.** A shipment of equipment is not a collaboration. Partnership requires host decision-making power, as the mission literature emphasizes in its call for sustained partnership with host institutions and local teams [3]. **Confusing open access with consent.** Data posted openly may violate the consent participants actually gave. Consent scope governs what can be shared, and publication ambition does not override it. **Confusing cost recovery with profit.** Charging documented cost is fair. Charging rates tuned to what a grant will bear, without transparency, is not. **Confusing authorship with acknowledgment.** A technician who designed the sampling strategy deserves authorship. A courier does not. Decide before submission, not at proofing stage. **Confusing data sharing with benefit sharing.** Returning a dataset to contributors is not the same as returning benefit. Benefit can include capacity, infrastructure, authorship, access to interventions, or negotiated financial terms. **Confusing digital adaptation with sharing.** Translating a guideline into another country's digital system, as the WHO SMART Guidelines digital adaptation kits do when localized into national platforms, is resource intensive and can compromise fidelity to the recommended content if done as a copy-paste exercise [12]. Shared digital tools need local adaptation resources, not just local installation. ## Common Mistakes and Limitations ![Four-step sequence: agree ownership, agree benefit sharing, agree publication and credit, then agree wind-down and maintenance terms.](/article-images/illustrations/resource-sharing-ethics-in-global-research-agreement-before-transfer.webp) - **Negotiating terms after resources have moved.** Once samples or data have shipped, the receiving side holds all leverage. Ownership, use, publication, and wind-down terms belong in writing before transfer. - **Leaving maintenance unfunded.** Donated equipment and shared platforms need service contracts, spare parts, and trained operators in the budget, or the donation fails within a year or two. - **Treating ethics review as a one-time checkpoint.** The ecosystem view argues for distributed ethical scrutiny and reflexivity across funders, institutions, teams, and committees throughout a project [1]. Sharing agreements need review at renewal points, not just at initiation. - **Ignoring power asymmetry in contracts.** Standard material transfer templates drafted by the better-resourced side often assign all downstream value to the receiver. Low-resource partners should negotiate publication rights and benefit terms explicitly. - **Assuming training flows solve everything.** One-way training can accelerate brain drain if trainees have no local career path. Durable capacity building pairs training with local infrastructure and joint research roles. - **Underestimating the cost of equitable design.** Co-design phases, participatory workshops, and process evaluations take real time and money. Budgeting zero for them guarantees they will not happen. - **Assuming published ethical principles equal ethical practice.** Studies can cite inclusiveness and equity extensively while the underlying collaboration remains extractive [7]. Judge arrangements by their terms, not their ethics statements. One final limitation applies to every situation described here. Specific agreements, especially those involving human samples, identifiable data, or cross-border funding, require review by your institutional ethics committee and legal counsel, because requirements vary by institution, funder, and jurisdiction. General principles guide the conversation. They do not replace it. ## Frequently Asked Questions ### What does resource sharing ethics in global research actually mean? It is the set of principles governing fair, transparent, and accountable transfer of samples, data, equipment, funding, and expertise between research partners in different countries. The central test is whether the arrangement builds mutual capacity or transfers value from the lower-resource partner without fair return. ### Do I need a material transfer agreement to share biological samples internationally? In almost every case, yes. A material transfer agreement defines ownership, permitted use, publication rights, and liability, and most institutions require one before materials leave or arrive. Negotiating its terms equitably is your ethical obligation, not just a legal formality. ### Who owns data or samples collected in a lower-resource setting? Ownership depends on local law, participant consent, and the agreements signed before collection. The ethically sound default is that the originating institution and community retain a significant stake, and any downstream use beyond the original consent requires new agreement from them. ### What is fair benefit sharing in an international collaboration? Fair benefit sharing means value flows back to the communities and institutions that contributed resources, whether through co-authorship, training, infrastructure, access to the resulting interventions, or negotiated financial terms. The benefit type should be decided with local partners, not chosen for them. ### Is open data sharing always the most ethical option? No. Open sharing maximizes reuse and access, but it is inappropriate when consent does not cover open release, when re-identification risk is real, or when open release transfers value away from contributing communities with no return. Tiered access models often serve ethics better than full openness. ### How can early-career researchers in lower-resource settings get fair credit? Insist on authorship criteria being written into the collaboration agreement at the start, including for data collectors, site leads, and analysts. Programs that combine training with real project ownership, such as workshop-based curricula where participants develop their own abstracts and manuscripts, create the pathway to first and senior authorship rather than permanent acknowledgment lines. ## Sources 1. [Rethinking ethical reflexivity and oversight in health research through an ecosystem approach: A workshop report.](https://pubmed.ncbi.nlm.nih.gov/42454216/) 2. [Global variation in acquisition costs of single-use flexible ureteroscopes and flexible navigable access sheaths: an international collaborative analysis from the YAU & EAU Endourology groups.](https://pubmed.ncbi.nlm.nih.gov/42625061/) 3. [Spine Surgery Missions in Low-Resource Settings: Practical Considerations and Lessons Learned.](https://pubmed.ncbi.nlm.nih.gov/42777256/) 4. [Clinical Research Capacity Building in Pediatric Oncology in Resource-Limited Settings: Application and Early Results of Regional Educational Workshops.](https://pubmed.ncbi.nlm.nih.gov/42525911/) 5. [Advocacy and policy leadership training for primary care: a global environmental scan of competency frameworks, interventions, and strategic priorities.](https://pubmed.ncbi.nlm.nih.gov/42877277/) 6. [Accelerating equitable cancer genomics and precision oncology in health care and research: a Lancet Oncology Commission.](https://pubmed.ncbi.nlm.nih.gov/42790448/) 7. [Responsible artificial intelligence in healthcare: a systematic review on the use of ethical principles in the development and deployment of artificial intelligence.](https://pubmed.ncbi.nlm.nih.gov/42712331/) 8. [Feasibility of integrating thermotherapy (TT) for the treatment of Cutaneous Leishmaniasis into the primary healthcare facilities in east gurage zone, ethiopia: a protocol for an implementation research study.](https://pubmed.ncbi.nlm.nih.gov/42934502/) 9. [The Global Initiative for Obstructive Sleep Apnea (GIFOSA): Rationale, Strategic Framework, and Call to Action.](https://pubmed.ncbi.nlm.nih.gov/42768286/) 10. ['Taking care of your mind': participatory research to identify priorities and co-design wellbeing resources with the Muslim community following COVID-19.](https://pubmed.ncbi.nlm.nih.gov/42958217/) 11. [Implementation of a complex intervention for hip fracture care: a protocol for an embedded mixed-methods process evaluation of a multicentre cluster randomised trial in India, Nepal, the Philippines, Thailand and Vietnam.](https://pubmed.ncbi.nlm.nih.gov/42901556/) 12. [The Guideline Uptake in Digital Ecosystems (GUIDE) study: protocol for implementation research on the impact of WHO SMART guidelines digital adaptation kits to improve quality of care.](https://pubmed.ncbi.nlm.nih.gov/41068964/)