Equity by Design: When Health Systems Finally See Every Patient
Author: Bahar Folad, Multi-omics Research Analyst, Axia Medicine
Published by Axia Medicine | November 2025
This is the final article of a series. Previous entries:
Precision medicine has never been constrained by scientific potential. It has been constrained by patient visibility and access.
Across healthcare, advances in genomics, multimodal analytics, and population-scale AI continue to collide with systems that were never designed to include everyone. Scientific capability accelerates, yet access to testing, data generation, and coordinated participation does not. Data volumes grow, but representation does not. Insights expand, but equity does not.
This is not a failure of innovation. It is a failure of infrastructure.
Precision medicine can only be as equitable as the systems that determine who is visible in evidence generation and care. When large segments of the population remain structurally invisible, the outputs of even the most advanced technologies reflect that imbalance. Equity does not fail because science falls short, but because the foundations beneath science were never built to reach entire populations.
The question, therefore, is not simply how to make precision medicine fairer downstream, but what becomes possible when health systems are built upstream to see individuals continuously, across settings and across time.
That question defines the patient-centred infrastructure AXIA is building today.
The Status Quo: Precision Medicine Without Equity
Equity and equality are often evoked alongside precision medicine, but the distinction between the two is critical to understanding why progress has stalled. Equality offers the same solutions to all patients, while equity requires tailoring care based on genetic profiles, environments, behaviours, and lived context. This is the core principle of precision medicine.
Yet today’s systems fail to operationalise this principle.
The struggle begins with evidence. Despite the unprecedented advances in sequencing highlighted in The Genomic Paradox, the datasets underpinning precision medicine remain unevenly distributed, poorly integrated, and systemically biased. Even after correcting for repeated sampling, people of European ancestry still constitute two-thirds of participants in genome-wide association studies, with East Asian, African American/Afro-Caribbean, Hispanic or Latin American, and other ancestries remaining underrepresented (Ju et al., 2023). Screening tools and polygenic risk models derived from these datasets inevitably perform best for the populations they were built on.
Access further narrows visibility. In many regions, genetic testing remains concentrated in private laboratories or tertiary centres, placing it beyond the reach of most families. Egypt’s breast cancer screening programme, for example, continues to face structural limitations due to the cost of genetic testing (El-Attar et al., 2022). As a result, modern carrier screening panels disproportionately reflect variants common among European populations (Salari & Larijani, 2017). Entire communities remain invisible to tools explicitly designed to prevent disease.
At its core, this is a visibility problem. Innovation follows visibility, and visibility follows access. This pattern is not unique to healthcare. In its early years, PayPal did not grow by building more complex financial products for already banked users. Its inflection point came from expanding access, enabling millions of previously underserved individuals to transact digitally. As access broadened, new behaviours surfaced, trust accumulated, and unmet needs became visible, catalysing innovation in fraud detection, peer-to-peer payments, and embedded finance.
Precision medicine faces a similar inflection point. Without access, patients remain invisible. Without visibility, discovery concentrates around the same populations.
The consequences are predictable. Precision therapies emerge where infrastructure and capital already exist. High-income systems adopt personalised treatments rapidly, while lower-resource settings lack the financial and technical capacity to deploy them at scale (Morsi et al., 2025). A field designed to personalise care risks scaling inequity faster than it scales impact.
Gender disparities follow a similar pattern. Women represent more than half of healthcare utilisation, yet account for less than a third of medical research participation (Merone et al., 2022). Pregnancy, reproductive conditions, and sex-specific symptom profiles remain underrepresented in genomic datasets and clinical research (Geller et al., 2019; Kannan et al., 2019). Biases in symptom interpretation delay diagnosis and treatment, while AI systems inherit datasets embedded with racial and gender bias, reinforcing these inequities (Samulowitz et al., 2018; Sogancioglu et al., 2022; Sun et al., 2023)..
Digital fragmentation compounds these failures. The Multimodal Shift established that data is abundant but poorly distributed. Non-interoperable EHRs, inconsistent data standards, and poor integration between clinical and genomic information prevent the longitudinal visibility precision medicine requires (Abdelouahid et al., 2023; Prodduturi et al., 2025). Even large-scale investments, such as the UK’s £12 billion national record initiative, have faltered due to infrastructure constraints (Morsi et al., 2025). Where digital records do exist, inconsistent data quality, incomplete longitudinal capture, and bias amplification through advanced analytics continue to undermine their ability to support equitable, patient-centered care (Shen et al., 2025).
These blind spots are most visible in research participation. Up to 20% of trials terminate early due to recruitment failure, yet surveys show 85% of patients were unaware participation was an option at diagnosis, with 75% expressing willingness to enrol (Goodson et al.; NIH 2018).
Patients are willing. They are simply invisible.
This cycle is self-reinforcing. Limited clinical access concentrated participation among the few. Narrow participation generates narrow datasets. Narrow datasets underfeed new technologies, constraining discovery. Innovation slows, and invisibility deepens.
Health inequity, then, is not only an ethical injustice. It is a structural constraint rendering precision medicine scientifically fragile and demanding to scale.
The Bridge: Rebuilding the System Around Individuals
Incremental fixes are not enough. The literature proposes increasing genomic literacy, diversifying datasets, strengthening governance, modernising regulation, and expanding access through community engagement (Dankwa-Mullan, 2024; Thareja et al., 2024). Yet these interventions do not address the underlying problem: infrastructure.
Health data is not episodic, and it does not belong to institutions. It accumulates across a lifetime, across settings, and across modalities. Yet most health systems were designed to optimise local workflows, capture isolated encounters, or monetise fragmented datasets. They were never built to support continuous, individual-centred visibility at population scale.
Equity requires a patient-scale architecture, one in which access enables participation, participation creates visibility, and visibility shapes innovation. In this context, equity is not a downstream corrective. It is a structural prerequisite.
As outlined in The Patient-Centred Infrastructure Precision Medicine Has Been Waiting For, AXIA Medicine provides what health systems cannot build alone: continuous multimodal data collection, longitudinal patient visibility, dynamic consent, and programmatic access for research under a trusted governance framework. Such infrastructure enables decentralised participation, longitudinal evidence generation, and representative discovery without reliance on physical sites or institutional silos.
When patients are visible by design, equity becomes possible by design.
The World That Emerges When Systems Can See Everyone
Imagine a system built to see each patient clearly enough to intervene properly. Visible across their health journeys, patients can provide longitudinal, multimodal representations of biology, behavior, and environment. Participation becomes continuous rather than episodic.
Clinical research no longer depends on narrow, site-bound recruitment pipelines centred on academic institutions. Patient visibility enables decentralised participation by default, expanding recruitment across geography, income levels, working patterns, and physical ability. Trials enrol faster, retain participants more effectively, and generate cohorts that better reflect real-world populations (The Patient-Centred Infrastructure Precision Medicine Has Been Waiting For).
Remote and home-based genomic sampling and monitoring already demonstrate that precision medicine does not require proximity to academic centres; patient-centred infrastructure simply makes this model standard rather than exceptional (Carter et al., 2022; Thacarodi et al., 2023; Brancato et al., 2024). Participation becomes realistic for rural patients, shift workers, people with disabilities, and those far from specialist care, while Trial awareness can be embedded into digital workflows, rather than relying on brief clinical encounters
As participation broadens, datasets become more representative and the economics of precision medicine improve. To expand upon The Multimodal Shift, AI models trained on heterogeneous, longitudinal data generalise more reliably across populations, reducing bias-driven performance degradation and costly recalibration (Norori et al., 2021; Carrasco-Ribelles et al., 2023). Biomarker discovery strengthens as signals reflect diverse ancestries, sexes, and environments, and therapies validated against representative evidence become easier to deploy globally.
Public health gains a different form of visibility. When consented data is safely aggregated across individuals, systems can detect emerging trends earlier, whether infectious threats, environmental exposures, or shifts in chronic disease patterns within specific communities (Ghildayal et al., 2024; Shen et al., 2025). As explored in Precision Public Health, the COVID-19 pandemic demonstrated the cost of fragmented clinical, genomic, and population data; a patient-anchored approach is designed to prevent such failures (Yehudi et al., 2022).
Women’s health illustrates how equity emerges as a functional outcome rather than a parallel initiative. Many conditions that disproportionately affect women, or present differently across sexes, have been underdiagnosed due to a lack of longitudinal, sex-specific data integration (Gemmati et al., 2020). Embedding these dimensions as standard components of health infrastructure strengthens the evidence base for diagnosis, risk stratification, and care.
Finally, the digital landscape itself becomes less fractured. Consent travels with data, provenance is tracked at the level of the individual and each transaction, and interoperability emerges by design rather than negotiation (Torab-Miandoab et al., 2023; Welzel et al., 2025). Health systems no longer depend on monolithic platforms, because they share a common anchor: the patient.
In this world, equity is not something organisations pursue despite structural constraints. It emerges as a property of infrastructure designed to remove those constraints.
Precision medicine does not fail because the science is immature. It fails because this infrastructure exist at the level the field now demands.
Systems that can see everyone do not merely improve equity. They change what precision medicine can do at scale. Discovery becomes representative. Models generalise. Decentralised research becomes routine. Public health moves earlier.
AXIA Medicine exists because precision medicine cannot reach this phase without a new class of infrastructure, one designed around individuals rather than institutions. Once health data systems are built to truly see every patient, health equity will finally transform from an aspiration into the industry's default state.
Reference list
Abdelouahid, R.A., Debauche, O., Mahmoudi, S. and Marzak, A. (2023). Literature Review: Clinical Data Interoperability Models. Information, [online] 14(7), p.364. doi:https://doi.org/10.3390/info14070364.
Brancato, V., Esposito, G., Coppola, L., Cavaliere, C., Mirabelli, P., Scapicchio, C., Borgheresi, R., Neri, E., Salvatore, M. and Aiello, M. (2024). Standardizing digital biobanks: integrating imaging, genomic, and clinical data for precision medicine. Journal of Translational Medicine, [online] 22(1). doi:https://doi.org/10.1186/s12967-024-04891-8.
Carrasco-Ribelles, L.A., Llanes-Jurado, J., Gallego-Moll, C., Cabrera-Bean, M., Monteagudo-Zaragoza, M., Violán, C., and Zabaleta‐del‐Olmo, E., (2023). Prediction models using artificial intelligence and longitudinal data from electronic health records: a systematic methodological review. Journal of the American Medical Informatics Association, [online] 30(12), pp.2072–2082. doi:https://doi.org/10.1093/jamia/ocad168.
Carter, A.B., Abruzzo, L.V., Hirschhorn, J.W., Jones, D., Jordan, D.C., Nassiri, M., Ogino, S., Patel, N.R., Suciu, C.G., Temple-Smolkin, R.L., Zehir, A. and Roy, S. (2021). Electronic health records and genomics. The Journal of Molecular Diagnostics, 24(1). doi:https://doi.org/10.1016/j.jmoldx.2021.09.009.
Dankwa-Mullan, I. (2024). Health Equity and Ethical Considerations in Using Artificial Intelligence in Public Health and Medicine. Preventing Chronic Disease, [online] 21. doi:https://doi.org/10.5888/pcd21.240245.
El-Attar, E.A., Elkaffas, R., Aglan, S., Naga, I.S., Nabil, A. and Abdallah, H.Y. (2022). Genomics in Egypt: Current Status and Future Aspects. frontiers in Genetics, 13. doi:https://doi.org/10.3389/fgene.2022.797465.
Geller, S.E., Koch, A.R., Roesch, P., Filut, A., Hallgren, E. and Carnes, M. (2018). The More Things Change, the More They Stay the Same. Academic Medicine, 93(4), pp.630–635. doi:https://doi.org/10.1097/acm.0000000000002027.
Gemmati, D., Varani, K., Bramanti, B., Piva, R., Bonaccorsi, G., Trentini, A., Manfrinato, M.C., Tisato, V., Carè, A. and Bellini, T. (2019). ‘Bridging the Gap’ Everything that Could Have Been Avoided If We Had Applied Gender Medicine, Pharmacogenetics and Personalized Medicine in the Gender-Omics and Sex-Omics Era. International Journal of Molecular Sciences, 21(1), p.296. doi:https://doi.org/10.3390/ijms21010296.
Ghildayal, N. (2024). Public Health Surveillance in Electronic Health Records: Lessons From PCORnet. Preventing Chronic Disease, [online] 21. doi:https://doi.org/10.5888/pcd21.230417.
Ju, D., Hui, D., Hammond, D.A., Wonkam, A. and Tishkoff, S.A. (2022). Importance of Including Non-European Populations in Large Human Genetic Studies to Enhance Precision Medicine. Annual Review of Biomedical Data Science, 5(1), pp.321–339. doi:https://doi.org/10.1146/annurev-biodatasci-122220-112550.
Merone, L., Tsey, K., Russell, D. and Nagle, C. (2022). Sex Inequalities in Medical Research: A Systematic Scoping Review of the Literature. Women’s Health Reports, [online] 3(1), pp.49–59. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC8812498/.
Morsi, M.H., Elawfi, B., ALsaad, S.A., Nazar, A., Mostafa, H.A., Awwad, S.A., Abdelwahab, M.M., Tarakhan, H. and Baghagho, E. (2025). Unveiling the Disparities in the Field of Precision Medicine: A Perspective. Health Science Reports, [online] 8(8). doi:https://doi.org/10.1002/hsr2.71102.
Norori, N., Hu, Q., Aellen, F.M., Faraci, F.D. and Tzovara, A. (2021). Addressing bias in big data and AI for health care: A call for open science. Patterns, [online] 2(10), p.100347. doi:https://doi.org/10.1016/j.patter.2021.100347.
Otokiti, A.U., Shih, H. and Williams, K.S. (2025). Gender and Racial Bias Unveiled: Clinical Artificial Intelligence (AI) and Machine Learning (ML) Algorithms are Fanning the Flames of Inequity. Oxford Open Digital Health. doi:https://doi.org/10.1093/oodh/oqaf027.
Prodduturi, V.R. (2025). INTEGRATION OF CLINICAL GENOMICS IN ELECTRONIC HEALTH RECORDS: A COMPREHENSIVE FRAMEWORK FOR IMPLEMENTATION AND CLINICAL IMPACT. INTERNATIONAL JOURNAL OF COMPUTER ENGINEERING AND TECHNOLOGY, 16(1), pp.947–963. doi:https://doi.org/10.34218/ijcet_16_01_075.
Saberi, M.A., Mcheick, H. and Adda, M. (2025). From Data Silos to Health Records Without Borders: A Systematic Survey on Patient-Centered Data Interoperability. Information, 16(2), pp.106–106. doi:https://doi.org/10.3390/info16020106.
Salari, P. and Larijani, B. (2017). Ethical Issues Surrounding Personalized Medicine: A Literature Review. Acta Medica Iranica, [online] 55(3), pp.209–217. Available at: https://acta.tums.ac.ir/index.php/acta/article/view/6215.
Samulowitz, A., Gremyr, I., Eriksson, E. and Hensing, G. (2018). ‘Brave Men’ and ‘Emotional Women’: A Theory-Guided Literature Review on Gender Bias in Health Care and Gendered Norms towards Patients with Chronic Pain. Pain Research and Management, [online] 2018(1), pp.1–14. doi:https://doi.org/10.1155/2018/6358624
Shen, Y., Liu, Y., Krafft, T. and Wang, Q. (2025). Progress and challenges in infectious disease surveillance and early warning. Medicine Plus, [online] 2(1), p.100071. doi:https://doi.org/10.1016/j.medp.2025.100071.
Sogancioglu, G., Mijsters, F., Uden, van and Peperzak, J. (2022). Gender bias in (non)-contextual clinical word embeddings for stereotypical medical categories. arXiv (Cornell University). doi:https://doi.org/10.48550/arxiv.2208.01341.
Sun, T.Y., Hardin, J., Nieva, H.R., Natarajan, K., Cheng, R.J., Ryan, P.B. and Elhadad, N., (2023). Large-scale characterization of gender differences in diagnosis prevalence and time to diagnosis. medRxiv (Cold Spring Harbor Laboratory). [online] doi:https://doi.org/10.1101/2023.10.12.23296976.
Thacharodi, A., Singh, P., Meenatchi, R., Tawfeeq Ahmed, Z.H., Kumar, R.R.S., V, N., Kavish, S., Maqbool, M. and Hassan, S. (2024). Revolutionizing healthcare and medicine: The impact of modern technologies for a healthier future—A comprehensive review. Health Care Science, [online] 3(5). doi:https://doi.org/10.1002/hcs2.115.
Thareja, S.K., Yang, X., Paramita Basak Upama, Abdullah, A., Torres, S.P., Cocroft, L.J., Bubolz, M., McGaughey, K., Lou, X., Sailaja Kamaraju, Ahamed, S.I., Praveen Madiraju, Kwitek, A.E., Whittle, J. and Franco, Z. (2024). Equitable community-based participatory research engagement with communities of color drives All of Us Wisconsin genomic research priorities. Journal of the American Medical Informatics Association, [online] 31(12), pp.2940–2951. doi:https://doi.org/10.1093/jamia/ocae265.
Torab-Miandoab, A., Samad-Soltani, T., Jodati, A. and Rezaei-Hachesu, P. (2023). Interoperability of heterogeneous health information systems: A systematic literature review. BMC Medical Informatics and Decision Making, [online] 23(1). doi:https://doi.org/10.1186/s12911-023-02115-5.
Welzel, C., Ostermann, M., Smith, H.L., Timo Minssen, Kirsten, T. and Gilbert, S. (2025). Enabling secure and self determined health data sharing and consent management. npj Digital Medicine, [online] 8(1). doi:https://doi.org/10.1038/s41746-025-01945-z.
