Digital health: technology as a catalyst for transformation in healthcare

Technological advances are revolutionizing several sectors, including the healthcare sector. Digital health has allowed the development of a wide range of possibilities to improve the quality of healthcare in a variety of ways, such as promoting access to medical care, accelerating diagnosis and decision-making processes, improving patient outcomes and reduce costs.

Data is transforming healthcare by changing the way medical care is delivered and managed. The convergence of big data, artificial intelligence, wearables and advanced analytics is transforming the paradigm from a traditional reactive system into a more proactive, personalized and predictive healthcare approach. This article focuses on the different ways in which data and emerging technologies are transforming healthcare, including the Portuguese perspective on the digitalization of healthcare.

Big data

In recent decades, data collection and storage has been a common practice in healthcare, along with a growing interest in the secondary use of large amounts of data – big data – to improve clinical care and public health. Gartner proposed the “3V” definition of Big Data: “Big Data are high-volume, high-velocity, and high-variety information assets that require innovative and cost-effective ways of processing information for better insight and decision making” (Beyer et al).

Collected primarily from electronic health records, health insurance records, pharmacy data, medical devices and sensors and biomedical research, clinical data and real-world data help healthcare professionals better understand patient populations, comprehend the effectiveness of treatments, and support regulatory decisions, leading to:

Personalized medicine: Based on genomic data, medical history, biomarkers, lifestyle information and patient health metrics, healthcare will become tailored to individual needs, increasing treatment effectiveness, reducing side effects and anticipating response to medications.

Population health management: Aggregating and analyzing clinical data allows caregivers to identify populations at risk for certain diseases and create preventive public health interventions.

Value-based healthcare: By analyzing patients' clinical indicators and costs, care providers and insurers can evaluate health outcomes and focus on delivering quality care that reduces hospital readmissions and promotes health in the long term, contributing to a value-based care approach.

Artificial intelligence

The consistent collection of clinical data facilitates the use of technologies that systematize not only the amount of health information collected and stored electronically, but also the detail of the conclusions that can be drawn from these data (Adamson et al.).

By identifying patterns and trends from large complex data sets, artificial intelligence (AI) has the ability to increase the accuracy and speed of disease diagnosis and drug development, optimize hospital workflows and manage healthcare systems, and strengthen disease surveillance and outbreak response.

The use of AI prediction models has widespread use in healthcare, whether for identifying patients at high risk of readmission or developing targeted interventions to reduce readmissions (Mohanty et al.), predicting patient admission rates and timing length of hospital stay through improving capacity planning and resource allocation (Davis et al.), forecasting hospital bed occupancy rates and demand (Tello et al.), as well as forecasting and diagnosing several diseases, such as cardiomyopathy (Xia et al.), diabetes (Lu et al.) or Alzheimer's (Uysal and Ozturk). Ultimately, AI can have a huge impact on:

Predictive and preventive care: By analyzing patterns in patient data, algorithms can identify early warning signs in chronic diseases, enabling interventions and preventative measures, reducing healthcare costs and improving patient outcomes.

Clinical decision support systems: Data-driven tools can provide clinicians with real-time case-based decision support systems and predictive models. These tools can help doctors make more informed decisions, reduce diagnostic errors, and choose the most effective treatment options.

Unstructured data analysis: The collected health data can be structured and unstructured, as integrating data stored in both formats can add significant value to healthcare. Natural Language Processing can extract meaningful information from unstructured healthcare data such as medical notes, clinical reports, speech and research articles, helping to automate data entry, discover patterns in patient narratives and improve the quality of documentation in health systems.

Telemedicine and remote monitoring

The COVID-19 pandemic has been a catalyst for the development and adoption of a wide range of telehealth technologies by the health systems, such as virtual facilities and telecare platforms (Bouabida et al). The Internet of Things (IoT) and wearables have played a significant role in the expansion of telehealth, as data collected from these network of devices play a key role in enabling remote patient monitoring and remote care. Such devices can collect metrics – heart rate, blood pressure, glucose levels, activity levels and sleep patterns – which can be automatically transmitted to healthcare professionals. Thus, telemedicine and wearables enable continuous, real-time monitoring of patients, both for preventive healthcare and chronic disease management, reducing hospital admissions and enabling provision of care to remote populations. Additionally, telemedicine and wearables can significantly improve:

Patient engagement: These technologies enable patients to take a more active role in managing their healthcare and monitor and track their progress, promoting self-care and adherence to treatment plans.

Data-driven clinical trials: By aggregating data from real-world, wearable, and patient-reported outcomes and analyzing data from larger, more diverse populations, AI can help identify eligible patients for clinical trials, optimize trial designs, and generate insights about the efficacy and safety of drugs, creating virtual or hybrid clinical trials that will be faster, cheaper and more inclusive.

Furthermore, technologies such as blockchain are revolutionizing the way health data is stored and shared, ensuring that patient information is secure, immutable and accessible only by authorized parties, eliminating issues related to fraud and missing information, and ensuring accuracy in health records while giving patients more control to grant access to their records.

On the other hand, ensuring compliance with regulations such as the General Data Protection Regulation (GDPR) is critical to protect patient data and maintain trust. As privacy concerns grow, synthetic data is being used to simulate real-world patient data without exposing sensitive information. AI can generate synthetic datasets that reflect real patient behaviors for research and model training purposes, enabling the development of new treatments and public health strategies without exposing sensitive patient information, ensuring compliance with privacy laws.

Implications of data intelligence in healthcare management

Conclusion

Beyer, M., & Laney, D. (2012). The importance of ‘Big Data’: A definition. Gartner report, 1–9.

Adamson, B., et al. (2023). Approach to machine learning for extraction of real-world data variables from electronic health records. Frontiers in Pharmacology, 14, 1180962.

Mohanty, S. D., et al. (2022). Machine learning for predicting readmission risk among the frail: Explainable AI for healthcare. Patterns, 3(1).

Davis, S., et al. (2022). Effective hospital readmission prediction models using machine-learned features. BMC Health Services Research, 22(1), 1415.

Tello, M., et al. (2022). Machine learning-based forecast for the prediction of inpatient bed demand. BMC Medical Informatics and Decision Making, 22(1), 55.

Xia, C., et al. (2019). A multi-modality network for cardiomyopathy death risk prediction with CMR images and clinical information. In D. Shen et al. (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2019 (pp. 577–585). Springer International Publishing.

Lu, H., et al. (2022). A patient network-based machine learning model for disease prediction: The case of type 2 diabetes mellitus. Applied Intelligence, 52(3), 2411–2422.

Uysal, G., & Ozturk, M. (2020). Hippocampal atrophy-based Alzheimer’s disease diagnosis via machine learning methods. Journal of Neuroscience Methods, 337, 108669.

Bouabida, K., Lebouché, B., & Pomey, M. P. (2022). Telehealth and COVID-19 pandemic: An overview of the telehealth use, advantages, challenges, and opportunities during COVID-19 pandemic. Healthcare (Basel), 10(11), 2293.

Tang, J., et al. (2020). Pharmacist and data-driven quality improvement in primary care (PDQIP): A qualitative study of anticipated implementation factors informed by the Theoretical Domains Framework. BMJ Open, 10(2), e033574.

Klein, B., et al. (2023). Forecasting hospital-level COVID-19 admissions using real-time mobility data. Communications Medicine, 3(1), 1–9.

Pinto, C. S., et al. (2019). From big data to smart health: Putting data to work for the public’s health.

European Commission: Directorate-General for Communications Networks, Content and Technology, Page, M., Winkel, R., Behrooz, A., & Bussink, R. (2024). 2024 digital decade eHealth indicator study – Final report. Publications Office of the European Union


Bárbara Patrício

Program Manager – Health and Pharma