It is undeniable that healthcare, like virtually every other industry in the US today, faces many significant financial and other challenges. High inflation rates, tight labor markets, increasing supplier costs, and rising drug expenses are all contributing to significant financial losses among hospitals and clinics.
Certainly, stop-gap measures such as stemming reductions in Medicare payments are available to help hospital administrators and public policymakers fight these challenges. But these typically only slow the bleeding without solving the more extensive, longer-term financial problems. More effective solutions can arguably be found in new technologies and services based on artificial intelligence (AI) and machine learning (ML).
It was during 2020-2021, the most challenging years of the COVID-19 pandemic, that—almost out of desperation—some providers began implementing AI- and ML-based technology to help meet growing demand for disease screening and diagnosis, as well as to help address logistical matters like patient outreach and compliance. Indeed, almost overnight, the discussion evolved from whether AI would ever play a major role in health care to what kinds of AI tools were already available and ripe for implementation.
Today, new AI technologies are rapidly coming to market with promises of delivering higher screening and diagnostic accuracy, greater efficiency, earlier intervention, lower radiation exposures, and better automation. Indeed, these tools have the potential to significantly reduce costs and improve overall patient care.
Take breast cancer care, for example. Evidence is mounting in peer-reviewed journals and scientific presentations at society meetings that ML-based tools can help radiologists find cancers that they otherwise might miss.
Several ML technologies have been approved by the US Food and Drug Administration to help radiologists manage caseloads and identify lesions suspected to be cancerous with near-pinpoint accuracy. These breast cancer-screening tools have the potential to provide stakeholders with significant value; eg, to help radiologists identify lesions more confidently, to save payors substantial downstream care and treatment costs, and to give patients better quality of life and perhaps even more hope through earlier intervention.
So why aren’t hospitals beating a path to the doors of AI vendors? The problem isn’t a lack of capable tools or performance data. It isn’t a question of legal accountability. It isn’t even patient concern about artificial intelligence. The real problem is that the brunt of the costs of these technologies is borne by healthcare institutions; herein lies the greatest hurdle to innovation in health care.
To return to the breast cancer-screening example, it was a multi-year, state-by-state advocacy effort that ultimately led to nearly 100 percent Medicare reimbursement for digital breast tomosynthesis (DBT), a screening approach that is superior in nearly every way to standard mammography. As a result, with an average $54 Medicare add-on code for DBT, almost 90% of health systems today leverage the technology.
The potential for even greater adoption of life-saving AI-based technology exists if similar financial incentives are implemented in lung cancer screening, as well as in diagnostics for neurological, pulmonary, and cardiovascular diseases, as well as in breast cancer, prostate cancer, and many other areas.
But, except for perhaps some forward-thinking facilities that are willing to bet on future benefit, waiting for hospitals and clinics to extract more funds out of budgets that are already stretched thin is not the way to encourage mass adoption of new technology. Instead, as with DBT, financial incentives offer the most effective means to help health systems adopt and acquire the advanced technologies needed to improve efficiencies and drive earlier disease detection.
It may seem counterintuitive during these times of cost cutting to increase reimbursements to help fuel adoption of AI in health care. But that is exactly the kind of bold move needed to achieve the dual goals of reducing costs and improving patient outcomes.
The Catalyst to Revive Healthcare. Appl Radiol. 2022;51(6):29.
Mr Baird is chief commercial officer at DeepHealth.