Global Digi-Health Monitor: Calendar Week 15, 2025
Germany’s Digital Medical Record Delays Highlight EU-Wide Health Data Challenges
Germany’s electronic patient file (ePA) implementation has been postponed until at least April 2025, highlighting the challenges in the EU’s ambitious cross-border health data sharing initiative. While the European Health Data Space (EHDS) framework officially took effect on March 26, Germany—the EU’s most populous country—is struggling with cybersecurity concerns, technical limitations, and data privacy issues. A recent data breach before the January test phase even began prompted the German Medical Association to call for halting the rollout. The delay contrasts sharply with digital health leaders like Estonia and Finland, where citizens show greater willingness to share health data. The EHDS aims to standardize medical information exchange across the EU by 2029, with full implementation planned by 2034. Despite setbacks, German health officials maintain they’re working toward EHDS compliance, with Bavaria’s Health Minister Judith Gerlach suggesting Germany should learn from more advanced European neighbors rather than worry about catching up.
PROMISE: Sweden’s National Initiative to Advance Data-Driven Precision Medicine
Sweden has launched the Precision Omics Initiative Sweden (PROMISE), an ambitious national program that aims to integrate research with healthcare by leveraging the country’s unique advantages in biomedical infrastructure. The initiative will create large-scale molecular datasets from hundreds of thousands of individuals, connecting genomic and other omics data with Sweden’s comprehensive health registries and electronic medical records. This approach is made possible by Sweden’s distinctive combination of a publicly funded healthcare system, high public trust in research, universal personal identity numbers for longitudinal tracking, and world-class technology platforms through SciLifeLab and Genomic Medicine Sweden. By building deeper and more extensive datasets than currently available, PROMISE seeks to enable breakthrough discoveries in cancer and rare disease research, personalize patient care, attract industry partnerships for clinical trials, and ultimately establish Sweden as a global leader in precision medicine.
UK Launches £600 Million Health Data Research Service to Streamline Medical Innovation
Prime Minister Keir Starmer has announced the creation of a centralized Health Data Research Service with £600 million in government funding to improve access to NHS patient data for research purposes. The service, to be operational by the end of next year at the Wellcome Genome Campus in Cambridgeshire, will consolidate currently fragmented data access systems while implementing a cost-recovery pricing model for researchers and companies. This initiative follows recommendations from Professor Cathie Sudlow’s government-commissioned review and aligns with broader goals to enhance data flows across departments while capitalizing on valuable public health information. Alongside this data service, Starmer pledged to reduce clinical trial setup times from 250 to 150 days by March 2026, with both measures aimed at positioning the UK as a global leader in medical research while maintaining appropriate data security and patient trust.
The Legal Frontier of AI in Medicine: Balancing Innovation and Patient Protection
Harvard Law Professor Glenn Cohen outlines the complex legal and ethical landscape as artificial intelligence transforms healthcare. AI is already being used across medicine—from analyzing colonoscopy lesions and optimizing IVF hormone dosages to mental health chatbots and mortality prediction algorithms. While these technologies can process vast datasets and potentially democratize expertise, they raise significant concerns about patient data governance, model validation, informed consent, and equitable access. Current legal frameworks like medical malpractice law may actually discourage innovative AI applications by reinforcing conservative standards of care, and regulatory gaps exist where many AI tools fall outside FDA jurisdiction. Cohen notes that while the law typically adapts to disruptive technologies over time—as it did with medical imaging and genetic testing—the interim period brings critical questions about privacy, liability, and patient rights that require thoughtful legal solutions to ensure AI advances medicine without compromising safety or equity.
Study Reveals Demographic Bias in AI Medical Decision-Making
Researchers at Mount Sinai have discovered that generative AI models can recommend different medical treatments for identical clinical cases based solely on patients’ socioeconomic and demographic backgrounds. In a comprehensive study published in Nature Medicine, the team stress-tested nine large language models across 1,000 emergency department cases replicated with 32 different patient profiles, generating over 1.7 million recommendations. The results showed concerning patterns: high-income patients were more frequently recommended advanced imaging tests like CT scans and MRIs, while low-income patients were often advised to forgo further testing. Mental health evaluations were also inconsistently recommended based on demographics rather than clinical need. The researchers emphasize that these findings highlight the urgent need for robust AI assurance frameworks, validation processes, and oversight mechanisms to ensure AI-driven healthcare remains fair, safe, and effective for all patients regardless of their background.
