Target Audience
- Researchers, healthcare professionals, and policymakers interested in the development and operation of large-scale biobanks such as the UK Biobank and the Million Veteran Program (MVP).
- Professionals working in genomic medicine, precision medicine, healthcare big data, and research involving electronic health records (EHRs).
- Anyone interested in the development, governance, research use, and clinical implementation of large-scale health and biomedical data infrastructures.
Summary
- The Million Veteran Program (MVP) is a large-scale genomic and health research initiative launched in 2011 by the U.S. Department of Veterans Affairs (VA). It reached its original enrollment target of one million participants in November 2023.
- Unlike the UK Biobank, which recruited participants from the general population, MVP is built upon an existing healthcare system and integrates genomic information, medical records, lifestyle data, and military service experiences.
- MVP has generated significant findings through large-scale genetic studies involving diverse populations, producing more than 600 scientific publications as of October 2026.
- Major challenges include participant representativeness, EHR data quality, the availability of research-ready datasets, restrictions on external researcher access, information security, and clinical implementation.
- MVP demonstrates that the value of a large-scale biobank depends not only on the number of participants but also on its integration with healthcare systems, sustainable operations, and the translation of research findings into improved medical care.
1. Introduction: What Is the Million Veteran Program?
The Million Veteran Program (MVP) is a large-scale genomic and health research initiative launched in 2011 by the U.S. Department of Veterans Affairs (VA). It integrates and analyzes veterans' genomic information, medical records, lifestyle factors, military service experiences, and environmental exposure histories over extended periods.[1][2]
Like the UK's UK Biobank, MVP was established as a repository of biological samples and health information that could support a wide range of future scientific studies. However, there is an important difference in how the two initiatives originated.
While the UK Biobank is a large population-based cohort established by recruiting participants from the general public, MVP developed by linking genomic research to electronic health records (EHRs) already maintained by the VA, one of the largest integrated healthcare systems in the United States.
MVP reached its original enrollment target of one million participants in November 2023. By October 2026, it had developed into a major research resource supporting more than 150 ongoing research projects and generating over 600 scientific publications.[3][4]
The program's key milestones and figures are summarized below.
| Item | Figure or Date |
|---|---|
| Initial foundation of the concept | Beginning in 2006 |
| Launch of MVP | 2011 |
| Achievement of one million participants | November 2023 |
| Total enrollment | More than 1.09 million (October 2026) |
| Ongoing research projects | More than 150 (October 2026) |
| Scientific publications | More than 600 (October 2026) |
Five points are particularly important for understanding the development of MVP.
-
Its original vision was the advancement of precision medicine. MVP was conceived and launched before the Precision Medicine Initiative announced by the Obama administration in 2015.
-
Existing healthcare infrastructure provided the foundation for its growth. MVP benefited from the VA's nationwide healthcare network, longitudinal EHRs, research infrastructure, and biospecimen banking capabilities.
-
Genetic diversity is a major scientific strength. MVP has generated important findings that complement genomic studies historically dominated by populations of European ancestry.
-
Enrolling one million participants is not equivalent to having one million fully research-ready genomic and phenotypic datasets. Differences remain in sample processing, quality control, longitudinal data completeness, and availability within research environments.
-
Current challenges include data quality and representativeness, external researcher access, privacy protection, and clinical implementation. The program is transitioning from the accumulation of large datasets toward demonstrating how research discoveries can improve healthcare.
The following sections examine how MVP was conceived and launched, its distinctive features, major scientific achievements, current challenges, and differences from the UK Biobank.
2. How Was MVP Conceived and Launched?
MVP did not emerge suddenly as a nationwide project involving one million participants.
Its establishment was preceded by developments in VA genomic medicine policy, surveys of veterans' attitudes toward research participation, and the development of research ethics frameworks during the mid-2000s.
2006: The VA Launches a Genomic Medicine Initiative
In 2006, the VA began an initiative to apply knowledge of genetic function to healthcare, with the goal of personalizing disease prevention and treatment.
This initiative provided the policy foundation for what would later become MVP.[1]
2007–2009: Developing the Concept of a Large-Scale Genomic Biobank
Dr. Joel Kupersmith, then a senior VA research and development leader, and his colleagues helped advance the concept.
During this period, the VA conducted focus groups and surveys to understand veterans' attitudes toward participating in genetic research.
In May 2009, twelve physicians and scientists gathered in Rockville, Maryland, to discuss the design of the proposed research program.[1]
2010: Approval by the Central Institutional Review Board
In 2010, the VA's Central Institutional Review Board approved the MVP research protocol.
MVP was designed to allow participants' information and biological samples to be used not only for individual studies but also for a broad range of future research projects.
This required appropriate mechanisms for informed consent, privacy protection, the scope of research use, and the governance of future studies.
Approval by the Central Institutional Review Board represented a critical milestone in establishing MVP as a sustainable, long-term research infrastructure.[1]
2011: Participant Enrollment Begins and the Program Expands Nationwide
MVP began enrolling participants in early 2011.
On May 5 of that year, under Secretary of Veterans Affairs Eric Shinseki, the VA announced the nationwide expansion of MVP, formally launching a major research initiative with the ambitious goal of enrolling one million participants.[2]
2012: Enrollment Reaches 100,000
In 2012, MVP reached 100,000 participants.
The milestone demonstrated the effectiveness of recruiting participants through VA medical facilities across the country and engaging directly with veterans.[1]
2016: Enrollment Exceeds 500,000
By 2016, MVP had surpassed 500,000 participants.
It was becoming a major genomic research database and was also positioned as an important research resource supporting the broader U.S. Precision Medicine Initiative announced in 2015.[1]
2023: MVP Reaches Its One-Million-Participant Goal
On November 8, 2023, MVP enrolled its one-millionth participant.
The VA officially announced the achievement on November 15.
Approximately twelve years after its launch, MVP had reached its original enrollment target of one million participants.[4]
2024–2026: From Large-Scale Enrollment to Advanced Analysis and Scientific Discovery
In 2024, a major genome-wide association study (GWAS) involving approximately 636,000 participants and 2,068 traits was published in the journal Science.[5]
By October 2026, MVP had produced more than 600 scientific publications.[3]
During this phase, MVP increasingly focused not only on expanding enrollment but also on advanced data analysis, data quality control, expanding omics research, and translating research findings into clinical applications.
Three Important Decisions Made During the Planning Stage
Three decisions are especially important for understanding how MVP was established.
First, the program was built around the idea of using veterans' existing healthcare data for future biomedical research.
Because the VA already maintained extensive medical records, MVP did not need to collect all healthcare information from scratch. Instead, genomic information and additional survey data could be linked to existing medical records.
Second, participant trust and informed consent were emphasized from the design stage.
Using genetic information for a broad range of future studies, rather than for a single predefined research question, required participant understanding, ethical oversight, and robust confidentiality protections.
The VA's official historical account identifies ethical approval in 2010 as a critical step in overcoming the barriers to establishing the program.[1]
Third, MVP was designed as a shared, long-term research infrastructure rather than a study focused on a single disease.
This approach made it possible to support future investigations of cardiovascular disease, diabetes, psychiatric disorders, cancer, and many other conditions.
Taken together, these decisions show that MVP should not be understood simply as a project to collect DNA from one million individuals.
Rather, it was designed to integrate a healthcare system, research ethics, biospecimen collection, EHRs, and a scientific research community into a unified research infrastructure.
3. Distinctive Structural Features of MVP
3-1. Integration of Healthcare Delivery and Research Infrastructure
One of MVP's most distinctive features is its close integration with the VA healthcare system.
Rather than collecting research data entirely independently, MVP connects newly collected information with medical data already available through routine healthcare delivery.
With informed consent, participants provide biological samples and complete questionnaires.
These data are then linked with their existing VA medical records.
The main categories of data are summarized below.
| Data Category | Contents |
|---|---|
| Genomic data and biospecimens | DNA, plasma, whole-genome sequencing, and other omics data |
| Existing VA medical records | Diagnoses, prescriptions, laboratory tests, inpatient and outpatient visits, imaging, and clinical documentation |
| Lifestyle and self-reported information | Diet, smoking, physical activity, family history, and health status |
| Military service and exposure history | Military service experiences, deployments, and service-related environmental exposures |
These datasets are managed through integrated data systems and secure analytical environments.
They are made available for approved research through processes involving data quality control, disease phenotype definition, and access review.
Broadly speaking, the research activities supported by MVP fall into two categories:
- Genomic and epidemiological research: Identifying disease-associated genetic variants and risk factors.
- Clinical translation research: Applying findings to risk prediction, early detection, and personalized healthcare.
However, findings from genomic and epidemiological research do not automatically translate into clinical practice. Additional research and validation are necessary.
The VA's EHRs contain accumulated information on diagnoses, laboratory results, prescriptions, treatment histories, and clinical notes.
For example, researchers studying diabetes can compare the genetic characteristics of individuals diagnosed with the condition while also examining historical laboratory results and treatment records to investigate factors associated with disease onset and progression.
In addition, some studies can supplement VA records by linking them with information from the U.S. National Death Index and healthcare claims data from Medicare and Medicaid.[6]
3-2. Military Service and Environmental Exposure Data
Another distinctive feature of MVP is its ability to investigate health factors associated with military service that are often difficult to study in general population cohorts.
Research areas include:
- Post-traumatic stress disorder (PTSD)
- Traumatic brain injury
- Substance use disorders
- Health conditions associated with the Gulf War
- Associations between health outcomes, military experiences, and service-related environmental exposures
The ability to examine relationships among military-specific environments, experiences, genetic factors, and disease outcomes contributes to MVP's uniqueness.[7]
However, information about military-related exposures is not available with the same precision or completeness for every participant.
The number of participants available for a particular analysis may vary depending on how exposure information was collected and whether participants provided the relevant responses.
3-3. Diversity of Genetic Ancestry
One of MVP's important contributions is the inclusion of large numbers of participants from populations historically underrepresented in genomic studies, particularly those of non-European ancestry.
According to the VA's self-reported race and ethnicity data for more than one million enrolled participants as of November 2023, approximately 74% identified as White, 18% as Black, and 8% as Hispanic.[8]
Because Hispanic ethnicity may overlap with racial categories, these figures should not be interpreted as mutually exclusive categories that necessarily sum to 100%.
A 2024 study published in Science also described the ancestry composition of 635,969 participants included in genomic analyses.[5]
| Genetic Ancestry Group | Participants | Percentage |
|---|---|---|
| Similar to European reference populations | 449,042 | 70.6% |
| Similar to African reference populations | 121,177 | 19.1% |
| Similar to Admixed American reference populations | 59,048 | 9.3% |
| Similar to East Asian reference populations | 6,702 | 1.1% |
| Total | 635,969 | Approximately 100% |
Note: Percentages are rounded to one decimal place. These groups represent analytical classifications based on genetic similarity and should not be confused with self-reported race or ethnicity.
This diversity is scientifically important for reasons extending beyond demographic representation.
By analyzing populations with different genetic backgrounds together, researchers can identify disease-associated genetic signals that might not be detected in studies dominated by a single ancestry group.
MVP has demonstrated this advantage through major genomic research findings.
4. How Much of MVP's Data Is Currently Available for Research?
According to the VA's official data page, updated in late October 2026, MVP enrollment had exceeded 1.09 million participants.
However, the amount of data available for research varies by data type.[6]
| Data Category | Reported Scale |
|---|---|
| Enrolled research participants | More than 1.09 million |
| Samples genotyped | More than 1 million |
| Quality-controlled and curated genotype datasets | More than 650,000 |
| Whole-genome sequencing (WGS) completed | Approximately 190,000 |
| WGS datasets available for research | 102,000 |
| DNA methylation assays completed | 80,000 |
| DNA methylation datasets available for research | 45,000 |
| Participants completing baseline questionnaires | More than 625,000 |
| Participants completing lifestyle questionnaires | More than 508,000 |
Note: The denominator may differ among categories, and the figures may represent participants, samples, or datasets depending on the item.
These figures are important for understanding MVP's current development.
In particular, the maturity of the research infrastructure can be considered at three different levels.
| Stage | Scale | Significance |
|---|---|---|
| Participant enrollment | More than 1.09 million | Establishing the overall scale of the cohort |
| Research-ready genotype data | More than 650,000 | Availability of curated and quality-controlled genomic data |
| Research-ready WGS data | 102,000 | Availability of high-resolution genomic data |
These figures represent different stages of data processing and different types of data. They should not be interpreted as participant attrition rates.
For example, the published figures alone do not establish how many of the more than one million genotyped samples are awaiting processing or quality control, or how many were excluded from particular analyses.
In this sense, MVP is transitioning from the challenge of enrolling one million participants to the challenge of making data collected at that scale fully usable for scientific research.
5. Major Scientific Achievements of MVP
5-1. Large-Scale Multi-Ancestry Genomic Analysis
One of MVP's most important scientific contributions was reported in a paper published in Science in July 2024.
Verma A, et al. "Diversity and scale: Genetic architecture of 2068 traits in the VA Million Veteran Program." Science, 2024.[5]
The study conducted genome-wide association analyses across 2,068 diseases and traits involving 635,969 participants.
The researchers identified 13,672 genomic association regions.
Among these, 1,608 regions reached statistical significance only when participants from non-European ancestry groups were included.
This study demonstrated the importance of population diversity in discovering genetic associations that might otherwise be missed in analyses focused primarily on European ancestry populations.
5-2. Psychiatric Disorders and Veteran-Specific Health Challenges
MVP supports research into PTSD, depression, anxiety disorders, substance use disorders, and other mental health conditions.
A VA research overview published in May 2026 highlighted findings involving genetic associations shared across opioid use disorder, schizophrenia, bipolar disorder, and major depressive disorder.
It also discussed research investigating relationships between suicide risk and social factors, including food insecurity and marital status.[7]
An important strength of MVP is its ability to examine genetic factors in combination with social environments, lifestyle characteristics, and medical histories.
5-3. Translating Genomic Discoveries into Clinical Care
An increasingly important direction for MVP in 2026 is translating scientific discoveries into improvements in actual healthcare delivery.
Two notable initiatives are MVP-ROAR and ProGRESS.
MVP-ROAR: Returning Medically Actionable Genetic Findings
MVP-ROAR investigates how to identify participants with genetic findings associated with an increased risk of familial hypercholesterolemia and connect them with appropriate clinical follow-up.
The process includes obtaining additional consent, confirming findings through clinical genetic testing, providing genetic counseling, and sharing relevant information with treating physicians.
A pilot study and trial design were published in 2024.[9]
ProGRESS: Genetic Risk-Based Prostate Cancer Screening
ProGRESS evaluates whether a genetic risk prediction approach, developed and validated using data from more than 585,000 male MVP participants, can improve the detection of advanced prostate cancer when used alongside conventional screening methods.
The approach is being evaluated through a clinical trial.[10]
These initiatives represent an important transition.
Historically, MVP has primarily served as a resource for discovering genetic factors associated with disease.
It is now also supporting efforts to translate these discoveries into earlier detection and more personalized medical care.
However, these initiatives remain research and evaluation activities.
They do not mean that MVP-derived genetic information has already been widely integrated into routine clinical practice or that improvements in clinical outcomes have been conclusively established.
6. Current Challenges Facing MVP
The challenges facing MVP can be grouped into six broad areas.
It is useful to distinguish between issues explicitly identified by public authorities and structural limitations that arise from the program's research design.
6-1. Participant Representativeness and Selection Bias
Although MVP is large and includes participants from diverse racial and ethnic backgrounds, it is not a representative cohort of the general U.S. population.
According to enrollment data from November 2023, slightly more than 10% of participants were women, while approximately 55% were between 60 and 79 years of age.[8]
An older VA research document from August 2015 reported a recruitment response rate of 13.2%.
This historical figure should not be interpreted as the current response rate, but it illustrates the need to consider selection bias associated with willingness to participate and health status.
Consequently, careful interpretation is required when studying:
- Diseases specific to women, including pregnancy- and childbirth-related conditions
- Diseases more common among children and younger adults
- The applicability of disease prevalence estimates and risk predictions to the general U.S. population
- Differences between veterans who use VA healthcare services and those who do not
Importantly, genetic diversity and population representativeness are not the same thing.
A cohort can be genetically diverse while remaining unrepresentative of the general population in terms of age, sex, health status, or other characteristics.
6-2. EHR Data Quality and Completeness of Longitudinal Follow-Up
The VA's extensive longitudinal medical records are a major strength of MVP.
However, EHRs were originally designed to support clinical care rather than scientific research.
For example, a diagnosis code does not necessarily reveal the exact date of disease onset or the detailed severity of a condition.
Healthcare services received outside the VA system may also be incompletely captured within VA EHRs.
To address these challenges, MVP uses resources such as CIPHER (Centralized Interactive Phenomics Resource), which supports the standardization and sharing of algorithms used to define disease phenotypes from EHR data.
Where appropriate, VA records can also be supplemented with Medicare and Medicaid information and mortality records.[6]
Nevertheless, standardized phenotype definitions cannot eliminate all missing data or measurement errors.
6-3. The Gap Between Participant Enrollment and Research-Ready Data
As noted earlier, more than one million samples had been genotyped by 2026, but the VA identified more than 650,000 genotype datasets as quality-controlled, curated, and available for research.[6]
The value of a large-scale biobank cannot be measured solely by the number of samples collected.
Other critical factors include:
- Whether the necessary datasets have been adequately processed and curated
- Whether genomic data can be appropriately linked to phenotypic information
- Whether analytical methods and findings can be reproduced
MVP is also expanding whole-genome sequencing, DNA methylation analysis, metabolomics, and other omics studies.
However, the creation, processing, and quality control of these datasets continue to require substantial effort.
6-4. Restrictions on External Researcher Access
One of the most important differences between MVP and the UK Biobank concerns data access.
According to current VA documentation, access to MVP individual-level data is primarily organized around approved projects conducted by VA-affiliated researchers.
The VA explicitly states that there is currently no access pathway for studies led by investigators who are not affiliated with the VA.[6]
On the other hand, some MVP genome-wide association summary statistics may be accessible to external researchers through the U.S. database of Genotypes and Phenotypes (dbGaP).[8]
Access to summary-level statistical results should be distinguished from access to individual-level genomic and EHR data.
This creates important trade-offs.
| Advantages of Restricted Access | Potential Limitations |
|---|---|
| Facilitates strict protection of sensitive genetic and medical information | Makes independent validation by external researchers more difficult |
| Enables more consistent review and governance of research within the VA | Restricts the scope of international collaborative research |
| Supports the maintenance of highly secure analytical environments | Limits the expansion of the broader research community |
These trade-offs reflect the need to balance the protection of sensitive information with broader scientific access.
How to achieve this balance remains an important consideration for MVP and other large-scale biomedical research resources.
6-5. Information Security and Privacy Protection
A particularly important source on this issue is a report published in 2026 by the U.S. Government Accountability Office (GAO).
In its report released on May 21, 2026, the GAO identified weaknesses in information security controls affecting major systems supporting MVP.[11]
The areas examined included:
- Asset and risk management
- System configuration management
- Identity and access management
- Continuous monitoring and log management
According to the GAO, thirteen recommendations were issued in September 2025.
By March 2026, nine recommendations had been implemented, while three additional recommendations had been partially implemented.
| Item | Number |
|---|---|
| Recommendations issued in September 2025 | 13 |
| Implemented by March 2026 | 9 |
| Partially implemented by March 2026 | 3 |
The GAO concluded that these weaknesses reduced assurance regarding the confidentiality and integrity of sensitive health information.
However, the report does not establish that an actual data breach involving MVP had occurred.
It also indicates that the VA had already made progress in addressing many of the recommendations.
This case illustrates a broader issue facing large-scale genomic and healthcare data infrastructures: as the scientific value of accumulated data increases, so does the importance of maintaining rigorous information security over the long term.
6-6. Translating Research Findings into Better Healthcare
MVP has generated more than 600 scientific publications, but the number of publications should not be equated with demonstrated clinical effectiveness.
For example, identifying a genetic variant associated with a disease does not automatically establish that the discovery can improve clinical decision-making.
Additional evidence is required regarding:
- Predictive accuracy
- Clinical utility
- Cost-effectiveness
- Effects on decisions made by physicians and patients
In addition, MVP genomic analyses are primarily conducted for research purposes.
The program does not routinely return individual genetic test results to participants as part of standard clinical practice.
Initiatives such as MVP-ROAR are beginning to bridge this gap by investigating approaches for obtaining additional consent, clinically confirming relevant genetic findings, and communicating actionable results to participants and their physicians.[9][10]
Future evaluations of MVP should therefore consider not only how many participants have been enrolled or how many scientific papers have been published, but also how the program contributes to disease prevention, earlier diagnosis, and improved treatment outcomes.
7. Comparing MVP with the UK Biobank
Comparing MVP with the UK Biobank reveals important differences in the design philosophies of major biomedical research infrastructures in the United States and the United Kingdom.
| Comparison | UK Biobank | Million Veteran Program |
|---|---|---|
| Launch | Participant recruitment began in 2006 | Enrollment began in 2011 |
| Target population | General UK population, aged 40–69 at recruitment | U.S. military veterans |
| Cohort size | Approximately 500,000 | More than one million |
| Research model | Population-based prospective cohort | Healthcare system-linked cohort |
| Health data | Standardized baseline measurements linked to subsequent medical records | Existing EHRs supplemented with questionnaires and biological samples |
| Genomic research | Large-scale genomic and multi-omics research | Large-scale genomic research involving diverse genetic ancestries |
| Imaging research | Systematic multi-organ imaging studies involving more than 100,000 participants | Clinical imaging and examination data available through EHRs |
| Distinctive research areas | Broad range of diseases, lifestyle factors, and environmental influences | Common diseases as well as military service- and exposure-related conditions |
| Researcher access | Broad access for approved researchers in the UK and internationally | Individual-level access primarily for VA-affiliated researchers |
| Connection to clinical research | Linked to NHS and other healthcare records | Directly connected to the VA healthcare delivery system |
The UK Biobank has developed a model that combines standardized research measurements, genomic analyses, omics datasets, and imaging information within a common platform accessible to researchers around the world.[12]
MVP, by contrast, builds upon longitudinal clinical records and a participant population with diverse genetic backgrounds.
Its integration with the VA healthcare system provides opportunities to connect biomedical discoveries with clinical applications within the same healthcare environment.
The two initiatives should therefore be viewed as complementary rather than simply competing research infrastructures.
Their respective strengths and limitations can help address different scientific questions.
Indeed, collaborative research has used the UK Biobank to replicate and validate genetic associations identified through MVP.
This comparison highlights how two large-scale biobanks can differ substantially in their data collection strategies, relationships with existing healthcare systems, and policies governing external researcher access.
8. Understanding MVP's Development in Three Stages
The development of MVP from its original conception to the present can be divided into three broad stages.
These are analytical categories based on the program's history rather than officially designated phases.
Stage 1: Concept Development and Institutional Design (2006–2011)
During this period, the VA developed the framework for using biological samples and medical information from up to one million veterans in future research.
Major activities included:
- Developing the concept of a large-scale genomic research program
- Investigating veterans' attitudes toward participation
- Establishing ethical review processes
- Designing participant recruitment systems
- Building the organizational and research infrastructure
A particularly important aspect of this stage was not simply setting the ambitious target of one million participants.
It was determining how the existing VA healthcare system could be integrated with a long-term research infrastructure.
Stage 2: Participant Recruitment and Research Infrastructure Development (2011–2023)
During the second stage, MVP expanded enrollment through VA medical facilities across the United States.
At the same time, the program advanced EHR linkage, genomic analysis, and the development of its research community.
Enrollment milestones included:
- 100,000 participants in 2012
- More than 500,000 participants in 2016
- One million participants in 2023
Over this period, MVP evolved from a proposed research initiative into a major genomic and health research infrastructure.
Stage 3: Large-Scale Analysis and Clinical Translation (2024 Onward)
Beginning in 2024, MVP increasingly emphasized advanced multi-ancestry GWAS, expanded omics research, and studies evaluating the clinical application of genetic risk assessment.
At the same time, several challenges became increasingly important:
- Quality control of research-ready datasets
- Development of high-resolution genomic and omics data resources
- Governance of research data access
- Information security
- Evaluation of clinical utility
MVP's emphasis is therefore shifting from increasing the number of participants toward determining how accumulated data can produce meaningful scientific discoveries and improvements in healthcare.
9. Future Prospects: What Determines the Value of a Large-Scale Biobank?
Four areas are particularly important when considering the future development of MVP.
9-1. Expansion of High-Resolution Omics Data
An important priority is expanding not only genotyping but also whole-genome sequencing, metabolomics, and epigenomic data.
Combining these datasets with clinical records and lifestyle information could enable researchers to investigate disease mechanisms and risk factors from multiple perspectives.
At the same time, generating these datasets creates additional demands for processing, quality control, storage, and analytical infrastructure.
9-2. Improving the Accuracy of Risk Prediction Across Diverse Populations
An important opportunity is to use MVP's genetic diversity to develop disease risk prediction methods applicable to a broader range of patient populations.
MVP has already contributed findings that complement genomic studies historically dominated by participants of European ancestry.
However, additional research is necessary before these discoveries can be reliably translated into individual risk prediction and clinical decision-making.
9-3. Balancing Data Accessibility and Public Trust
Because MVP manages sensitive genetic and medical information, strict privacy protection and information security are essential.
At the same time, broader scientific collaboration and reproducibility depend on researchers having appropriate access to data.
Balancing the protection of personal information with the expansion of research access is a challenge shared by large-scale biobanks worldwide.
9-4. Demonstrating Improvements in Clinical Outcomes
Initiatives such as MVP-ROAR and ProGRESS will help evaluate whether genomic discoveries can improve actual clinical care and health outcomes.[9][10]
Building a large research infrastructure and producing numerous scientific publications are important achievements.
However, from the perspective of healthcare applications, the ultimate question is how the research contributes to disease prevention, early detection, and improved treatment.
10. Conclusion: What Can We Learn from MVP?
The most distinctive feature of MVP is not simply that it has conducted genomic research involving more than one million participants.
Rather, its significance lies in building a continuous connection among medical records, biospecimens, genetic information, longitudinal follow-up, and clinical translation research within an existing healthcare system.
For countries and organizations that already have nationwide EHR infrastructures or extensive networks of healthcare institutions, MVP offers an important model for designing large-scale biobanks.
At the same time, MVP demonstrates that the development of a major research cohort does not end when participant recruitment is completed.
Long-term investment is necessary to maintain data quality, representativeness, trustworthiness, appropriate research access, and the translation of scientific findings into clinical practice.
By 2026, MVP had grown into a research infrastructure involving more than one million participants and producing over 600 scientific publications.
However, a large participant population does not automatically guarantee the quality of research-ready data or the clinical value of research findings.
Future evaluations of MVP will need to consider the continued development of high-quality datasets, advanced genomic and omics research, privacy and security protections, and evidence demonstrating clinical utility.
It is also important to acknowledge some remaining gaps in the publicly available information reviewed for this article.
The available sources did not provide a consistent historical series covering MVP's total investment since inception, annual operating costs, or data collection costs per participant.
Nor was a comprehensive program-level evaluation identified that quantitatively measures improvements in clinical outcomes resulting from MVP research.
These areas warrant further investigation when assessing the program's overall cost-effectiveness.
From the perspective of comparison with the UK Biobank, perhaps the most important lesson from MVP is that the value of a large-scale cohort depends not only on what data it collects, but also on which healthcare and research systems it connects with and what it aims to achieve over the long term.
A large-scale biobank is more than a repository of biological samples and data.
It should be understood as a long-term scientific and social infrastructure connecting healthcare systems, research ethics, data platforms, scientific communities, and clinical practice.
References
[1] U.S. Department of Veterans Affairs, "Object 33: The Million Veteran Program," VA History. https://department.va.gov/history/object-33-the-million-veteran-program/
[2] U.S. Department of Veterans Affairs, "The Million Veteran Program: VA's Genomics Game-Changer Launches Nationwide," 2011. https://news.va.gov/press-room/the-million-veteran-program-vas-genomics-game-changer-launches-nationwide/
[3] U.S. Department of Veterans Affairs, "Million Veteran Program Research," VA Office of Research and Development. https://www.research.va.gov/mvp/research.cfm
[4] U.S. Department of Veterans Affairs, "Million Veteran Program," Official Program Information and Announcement of the One-Million-Participant Milestone, 2023. https://www.research.va.gov/mvp/
[5] Verma A, et al., "Diversity and scale: Genetic architecture of 2068 traits in the VA Million Veteran Program," Science, 2024. https://doi.org/10.1126/science.adj1182
[6] U.S. Department of Veterans Affairs, "MVP Data Available for Research," Million Veteran Program, 2026. https://www.mvp.va.gov/pwa/mvp-data-available-research
[7] U.S. Department of Veterans Affairs, "Million Veteran Program Research," VA Office of Research and Development. https://www.research.va.gov/mvp/research.cfm
[8] U.S. Department of Veterans Affairs, "Million Veteran Program," Official Information on Participant Demographics and Research Data. https://www.research.va.gov/mvp/
[9] Vassy JL, et al., Article on the MVP-ROAR Study, American Heart Journal, 2024. https://pubmed.ncbi.nlm.nih.gov/38762090/
[10] U.S. Department of Veterans Affairs, "From Health Research to Healthcare," 2026. https://news.va.gov/149786/from-health-research-to-healthcare/
[11] U.S. Government Accountability Office, "GAO-26-108651," Audit Report on Information Security in the Million Veteran Program, 2026. https://www.gao.gov/products/gao-26-108651
[12] UK Biobank, "UK Biobank," Official Information on the Research Infrastructure and Researcher Data Access. https://www.ukbiobank.ac.uk/