The need for cost-effective innovative eHealth solutions utilising AI technology is greater than ever and is set to continue to increase as more people require care and support and as health systems look to shift care out of the in-patient hospital setting and into the out-patient and primary care setting. As a result, new AI driven eHealth solutions are being developed and created for use in the life sciences and healthcare sector at a fast rate and regulators, payers and prescribers are struggling to keep up with the pace of change.

The growing demand for eHealth and AI solutions in life sciences and healthcare

The last decade has witnessed an explosion of investment in eHealth devices and solutions following a rapidly increasing demand for utilisation of AI technology. According to analysis the AI in healthcare market is expected to grow from USD 2.1 billion in 2018 to USD 36.1 billion by 2025, at a CAGR of 50.2% during the forecast period [1].

This demand is driven by the fact that health systems across the globe are facing the challenge of an increasing demand for healthcare due to growing and ageing populations, a rising number of people with long term chronic diseases, the rising cost of medical treatments and severely constrained budgets. Specifically, in the UK, healthcare needs reform if it is to remain a high-quality health service free at the point of care [2].

Life sciences and healthcare businesses are set to benefit significantly from eHealth and AI

eHealth can be described as “the cost-effective and secure use of information communication technologies (ICT) in support of health and health related fields, including health-care services, health surveillance, health literature, and health education, knowledge and research” [3] and more and more eHealth solutions are utilising AI technology. By prioritising high-risk patients for screening, improving disease prevention, predicting health-outcomes and optimising treatment for patients, enabling earlier diagnoses and improving patient monitoring, eHealth solutions, in particular when empowered by AI algorithms, have the potential to make care more effective and affordable.

Medical imaging is one of the fast-moving areas of discovery and one of the obvious segments in which eHealth solutions and AI technology can provide benefit to healthcare systems. Solutions such as the well-known IBM Watson Health Imaging to lesser known start-ups such as MaxQ AI, augment radiologists and ophthalmologists capabilities with AI powered eHealth solutions which will not only optimise their workflow but also have the potential to increase productivity and most importantly enhance the diagnosis of the disease, reduce physician burnout and improve patient outcomes.

In the future, eHealth solutions will be a means of better integrated and personalised health care, with a greater focus on disease prevention and on the promotion of health and wellbeing. For example, solutions will be developed that enable patients to monitor their risk of deterioration of long-term chronic conditions such as cardiovascular disease (CVD) or hypertension. Through continuous monitoring of a patient’s heart function, lifestyle behaviours and specific biomarker signatures, AI-enhanced eHealth solutions have the potential to detect escalation of disease sooner. This will enable patients and practitioners to respond promptly and appropriately to optimise treatment and improve patient outcomes.

These types of eHealth solutions have the potential to positively contribute to the sustainability of the healthcare system by removing inefficiencies in care processes, reducing waste and improving patient management and clinical decisions. For example, to come to a clinical decision, a considerable amount of time is currently spent by physicians sifting through information provided by patients, their own knowledge and test results. In the future, AI powered eHealth solutions will have the ability to streamline the whole process and save time by assisting physicians in interpreting all the data and streamlining the decision-making process.

Market Access Barriers and Adoption Hurdles for AI powered eHealth solutions


At present, only 16% of healthcare facilities in Europe use AI tools [4], as currently there are still many barriers preventing widespread access and adoption. A fundamental issue pertaining to this lack of adoption is that healthcare systems are not yet set up to approve or incorporate such solutions. A lack of interoperability between eHealth solutions, compatibility issues with current IT systems, a lack of legal clarity and agreed standards for data protection, privacy and security of both the data and devices and the lack of clear regulatory framework have reduced access and uptake of eHealth solutions, not to mention the high start-up costs. Companies looking to launch new eHealth solutions will need to acknowledge the following barriers to market entry and develop appropriate access and communication strategies to overcome them:

1. Inappropriate methods of evaluation for eHealth solutions

Appropriate methods and valuation criteria to appraise and reward performance in delivering health outcomes are either not in place or are in a nascent state for eHealth solutions. Most often, eHealth solutions will still be evaluated like any other new technology or medicine via HTAs and modelling of health outcomes and benefits or using cost-effectiveness and cost-utility models.

The drawback to this approach when assessing eHealth solutions is that the benefits of an eHealth solution may extend beyond health outcomes and direct costs to healthcare systems and also include access, information, waiting time, time saved, convenience for patients and avoidance of burdensome travels [5].

2. Inadequate funding and reimbursement models

 Inadequate funding and reimbursement models currently exist and there is no specific budget set aside for such solutions therefore most eHealth solutions are not reimbursed. Incorporating or implementing new eHealth solutions requires significant investment from healthcare systems and to justify this, new solutions are expected to demonstrate that they perform at least as well as the standard approach to patient management.

Limited large-scale evidence of the cost-effectiveness of eHealth solutions exists and new players entering this space will need to plan alternative value communication strategies focusing on improved patient outcomes and quality of life and improved physician performance and efficiency.   

 3. Substantial clinical testing is required

Any new eHealth solution incorporating machine  learning or AI algorithms to diagnose or monitor the health of patients will need to undergo substantial clinical testing under a wide variety of situations and patient types to ascertain the clinical performance. Without significant evidence demonstrating the accuracy, reliability, safety and effectiveness of AI powered eHealth solutions, concerns over patient safety, quality and liability will arise and access and funding will be limited. It is also important to consider that by its very nature, eHealth solutions incorporating machine learning will be constantly evolving.

The current clinical and regulatory appraisal environment works on the assumption that a pharmaceutical or device’s characteristics are set, and access is granted based on this explicit set of evidence and product description. In light of any new available data, a change in device specification or reformulation, manufactures must resubmit new evidence on the safety and effectiveness of the pharmaceutical/medical device. Given the nature of AI powered eHealth solutions, many products will adapt in response to patient outcomes, any glitches, adverse events, or other safety concerns. As a result, it is not practical to constantly reassess AI powered eHealth solutions which react to changes in the traditional fashion. To overcome  this, the FDA has introduced its Software Precertification (Pre-Cert) Pilot Programme which a takes a Total Product Lifecycle (TPLC) approach. The proposed approach aims to first assess the software developer and/or digital health technology developer, rather than the product. The manufacturer/developer will be assessed on patient safety, product quality, clinical responsibility, cybersecurity responsibility, and proactive culture. The programme will then leverage unique post market opportunities to verify the continued safety, effectiveness, and performance of AI powered eHealth devices in the real-world without having to prevent patient access. As a result, the program takes on a Total Product Lifecycle (TPLC) approach to regulatory assessment, enabling the evaluation and monitoring of AI powered eHealth solutions from premarket development to post market performance, along with continued demonstration of the organization’s excellence [6].

 4. Lack of physician trust and awareness

Regulatory and HTA expectations are only one side of the story for successful adoption and uptake of eHealth solutions. In addition, clinicians or healthcare professionals (HCPs) must be willing to incorporate these new solutions into their patient management algorithms. A lack of awareness of, and confidence in eHealth solutions among professionals will hinder uptake and companies therefore should look to mitigate these concerns by involving stakeholders in the development of new solutions and provide training and education to users on the benefits and safety of the solutions.

5. Disrupting traditional methods of patient management and increasing patient anxiety

Many eHealth solutions utilising AI technology aim to disrupt the traditional management of patients with long term chronic conditions whereby (for example) patients may be currently expected to visit their HCP maybe every 6-12 months for a regular check-up. eHealth solutions of the future will empower patients to monitor their own disease state on a more continuous basis and identify their risk of further deterioration. Despite recognising many of the benefits of empowering patients to monitor and manage their disease, there is a lot of concern around misinterpretation of the results and the impact this will have on patient anxiety. To mitigate this concern, it will be essential for companies to ensure that any new solutions incorporating AI are clinically validated with a high specificity and sensitivity and that patients are educated and trained on the meaning of their results.

6. A data overload for physicians

Currently physicians spend a vast amount of time interpreting test results and patient medical records. Concern amongst some physicians exists, that eHealth solutions used to continually monitor long term chronic diseases may increase physician’s workload due to an increased amount of data and test results that will need to be interpreted. Companies must therefore develop eHealth solutions from the perspectives of both the end user (patient) and HCP in mind. Developing a solution that in the end only improves patient outcomes and produces large amounts of patient medical data will find strong resistance from HCPs if it does not also help to streamline or reduce their workload.

Looking to the future

In the current environment market access and adoption is going to be challenging for eHealth solutions. Despite the overwhelming need for such devices and solutions, significant barriers to uptake, from a regulatory, payer, HCP and patient perspective, exist. When looking to invest in the multitude of start-ups operating in this space, it is essential that companies and investors conduct the appropriate strategic analysis and due diligence, assess the environment through a holistic lens and think beyond the technical, legal, IT and regulatory environment and understand that success in this space is achievable but the role of both the payer and prescriber carry significant weight. Companies looking to enter this space and overcome these challenges need to prepare communication strategies tailored to each stakeholder. Understanding the complex decision-making process, identifying key stakeholders (national and local payers, HCPs, patients and policy makers) and mapping out their level of influence on the decision making and funding process will form the foundation for successful market access.

Authors:

Rechercher
Toggle location