Navigating the future of AI through health economics
During my recent chat on The Healthtech Podcast, I spoke candidly about the important aspects of health economics, touching on early value assessment, budget impact modelling, and why technologies (especially AI) need meticulous evaluation to ensure not only patient safety, but value for money. Without a strong health economic argument, it’s possible to bark up the wrong tree completely and produce tech that has no commercial viability.
The significance of health economics in AI
Health Economics and Outcomes Research (or HEOR to those in the know) explores the value of new healthcare interventions by balancing their costs against benefits, often expressed in Quality Adjusted Life Years (QALYs). This analysis supports key areas such as pricing, commercialisation, and reimbursement. You know, all the things that medtech developers actually want after years of grinding away at building something.
It’s particularly important to consider how we actually use AI in healthcare due to its promises to improve just about everything under the sun (but can it and should it?), especially when it comes to claims around time-saving. Careful validation, regulation and economic modelling ensure that AI not only complies with medical standards and legal requirements, but also provides evidence of cost-effectiveness that positively enhances healthcare systems instead of depleting resources and money without clear benefits.
To start the economic evaluation process, there are two main types of economic models developers can consider:
Beginning with early value assessments (EVA), this is an important strategy employed to evaluate the potential value of a system or technology early in its development cycle. This assessment helps guide investment and development decisions by predicting whether the technology will meet the expected clinical and economic thresholds before reaching the market. EVA also helps mitigate risks by allowing developers and stakeholders to pivot or iterate applications based on early economic feedback.
The second assessment, budget impact modelling (BIM), looks at the financial implications of integrating new technologies into healthcare systems. It forecasts how adopting systems or technologies might alter expenditures within a healthcare setting, providing insights into potential savings or costs over a fixed time period.
By using both EVAs and BIMs, businesses can determine if AI technologies are financially viable before going down the lengthy road of clinical validation and regulation, and help stakeholders make informed decisions. This also allows developers to engage early with payer bodies, presenting preliminary economic models that showcase potential value and cost-effectiveness.
Most importantly, technologies need solid economic evidence to be recommended by NICE and procured into the NHS. By considering both financial and clinical impacts, health economic models support the responsible use of AI and tech in healthcare. In fact, the NICE Evidence Standards Framework recommends both EVA and BIMs for all novel technologies.
The practical implications of AI in clinical settings
We all know that the practical applications of AI have been lauded for their potential benefits. However, these advantages are often accompanied by significant concerns. Take, for example, an AI system designed for lung nodule detection, which is integrated into routine CT scans. While it enhances diagnostic accuracy and facilitates early intervention by identifying nodules that might otherwise go unnoticed, the implications are twofold:
Firstly, introducing such AI tools leads to heightened initial costs due to increased lung nodule detection. It goes without saying, if you find more cancers, someone then has to pay the costs of treating those cancers. While economic models often predict long-term savings from reduced treatments in advanced disease stages, these models depend heavily on ideal scenarios and assumptions that may not hold in real-world settings. The increased frequency of diagnostics can also lead to overdiagnosis – where non-threatening conditions are treated unnecessarily – escalating healthcare costs and potentially causing harm to patients through unneeded interventions.
Moreover, the reliance on AI for diagnostic accuracy raises concerns about the dependency it creates within healthcare systems. There is a risk that the technology's recommendations might overshadow clinical judgement, potentially leading to errors or bias. These systems require rigorous testing and validation to ensure they do not perpetuate existing biases or introduce new ones, particularly in diverse populations.
While seemingly promising, such technologies can pressure healthcare policies and practices to shift prematurely towards more technology-dependent interventions. This shift may prioritise perceived economic efficiencies over patient-centred outcomes (which should remain the primary focus).
While health economic models can demonstrate the cost-effectiveness of AI-driven tools and support policy shifts towards earlier diagnostic interventions, their integration must be approached critically. This ensures that we remain aware of healthcare economics' impact on AI adoption and policy planning while prioritising patient-centred outcomes and addressing AI's risks and ethical considerations.
Future-proofing healthcare responsibly
While AI could be said to have significant potential to improve healthcare, rigorous EVAs and BIM are essential. These evaluations inform stakeholders about AI’s practical benefits and limitations, ensuring that integration is both technologically sound and economically viable.
Strict regulations and controls are also necessary to mitigate risks like systemic errors, biases, and privacy concerns. By guiding digital health developers to focus on measurable benefits that address existing gaps, EVAs and BIM reduce financial risks regarding the integration of AI in healthcare, meeting both current and future demands.
At Hardian Health, our expertise in health economic models can evaluate whether AI-driven tools are cost-effective and aligned with policy and regulations. We work collaboratively and transparently to provide comprehensive support, ensuring that your AI innovations are adopted responsibly and effectively. We achieve this while prioritising patient-centred outcomes and addressing the economic and ethical considerations essential for long-term success.
Hardian Health is a clinical digital consultancy focused on leveraging technology into healthcare markets through clinical strategy, scientific validation, regulation, health economics and intellectual property.
