Scaling Innovation in Health Care Wisely: Lessons from Jevons Paradox

Jevons paradox strikes again! As AI gets more efficient and accessible, we will see its use skyrocket, turning it into a commodity we just can't get enough of.

- Satya Nadella, Microsoft CEO

A few weeks ago, the release of DeepSeek’s new AI model shook the US tech industry. Despite export restrictions limiting China’s access to the best AI chips, DeepSeek’s models outperformed many others in the US that invested vastly more in infrastructure. Instead of relying on raw computing power, DeepSeek’s breakthrough came from a novel reasoning algorithm—an approach not previously seen in AI models. DeepSeek’s success generated immense interest, quickly making waves across the AI and tech industries

This news sparked conversations about Jevons Paradox—the idea that as efficiency improves, overall consumption often increases rather than decreases. This paradox, originally rooted in the industrial revolution, has surprising and profound implications for scaling innovation wisely in healthcare.

First described by economist William Stanley Jevons in the 1860s, Jevons Paradox arose from observations of coal consumption in England. As steam engines became more efficient, one might have expected coal usage to decline. Instead, improved efficiency made coal more economically viable, increasing its adoption across industries and leading to even greater consumption - fueling the Industrial Revolution. This counterintuitive principle—where efficiency gains lead to higher demand rather than conservation—has played out across multiple sectors, from energy to computing.

In healthcare, the paradox is strikingly evident. One would assume that innovations designed to make healthcare delivery more efficient—electronic health records, AI-driven diagnostics, telemedicine—would lead to a reduction in resource use. Instead, they often expand access, surface unmet demand, and ultimately increase utilization rather than streamline it.

Consider the promise of telemedicine. Initially hailed as a way to reduce the strain on in-person visits and make care delivery more efficient, telehealth has paradoxically led to more visits, not fewer. By lowering the barriers to entry, telemedicine enables patients who might have otherwise skipped a visit to seek care more frequently. This isn’t necessarily a bad thing—it improves access—but it does illustrate how efficiency in healthcare can trigger greater utilization rather than reducing strain on the system.

Let’s take a look at a couple examples.

Electronic Health Records (EHRs) were meant to streamline documentation and reduce administrative work. Yet, in practice, they have led to more documentation, not less. EHRs have facilitated increased ordering of tests, a way to a method to meet more detailed documentation requirements, and a rise in administrative burden for physicians. Clinicians now spend more time than ever on documentation, contributing to burnout and inefficiencies rather than solving them.

AI’s role in healthcare diagnostics is supposed to improve accuracy and speed up decision-making. However, by making it easier to detect potential abnormalities, AI could inadvertently lead to an increase in additional testing, specialist referrals, and follow-ups. If AI flags even minor anomalies, clinicians often feel compelled to investigate further, leading to more follow-ups, additional testing, and sometimes unnecessary medical interventions

The rise of retail clinics, urgent care centers, and virtual-first care models was expected to reduce the burden on emergency rooms and primary care providers. Instead, these models have expanded access and surfaced unmet demand, increasing total encounters rather than shifting them. When care becomes more convenient, people use it more frequently, further fueling demand.

If Jevons Paradox tells us anything, it’s that efficiency alone is not the answer. Scaling innovation in healthcare requires careful planning to ensure that efficiency gains do not simply add more work to the system but instead are absorbed in ways that actually improve care delivery and sustainability.

Many AI tools today serve as augmentation rather than automation, meaning they generate insights that require additional human intervention. These will be important but we also need AI to replace redundant administrative tasks, not just surface more information that requires human processing. A few examples:

• AI-powered clinical documentation assistants that write notes in real time, reducing clinician workload rather than increasing documentation complexity.

• AI-driven billing and coding automation that eliminates tedious administrative tasks rather than creating new ones.

To avoid overwhelming clinicians with increased demand, the system will also need to embrace asynchronous care models that allow healthcare professionals to manage patient needs more efficiently. A few examples:

• AI-powered patient portals that handle common inquiries without requiring human intervention.

• Structured asynchronous telehealth visits where patients submit structured data, and clinicians respond when necessary, rather than every case requiring a full real-time interaction.

Healthcare has to shift move away from traditional volume-based models that are heavily transactional toward more meaningful patient interactions that are designed to prevent unnecessary visits while still ensuring quality care. A few examples:

• Longitudinal remote monitoring for chronic conditions, using wearables and AI-driven alerts to proactively manage health without requiring constant in-person checkups.

• AI-assisted triage systems that direct patients to the right level of care without unnecessary physician involvement.

Jevons Paradox teaches us that efficiency alone doesn’t solve the problem—it often fuels greater utilization rather than reducing demand. In healthcare, every efficiency gain risks becoming a new driver of increased engagement, more interventions, and higher utilization.

So, the question isn’t just: How do we make healthcare more efficient? The real question is: How do we ensure that increased efficiency translates into better, more sustainable care?

The future of healthcare innovation depends on finding ways to scale wisely—not by adding volume, but by designing better systems that truly improve outcomes while easing strain on providers. Jevons Paradox warns us that innovation alone doesn’t reduce demand—it amplifies it. The real challenge isn’t just implementing new technology, but rethinking how we integrate it into workflows, ensuring it absorbs demand rather than exacerbates it.

The healthcare systems that will thrive in the future won’t be the ones that simply chase efficiency—they’ll be the ones that redefine sustainability, designing innovations that don’t just generate more care, but deliver the right care, at the right time, in the right way.

This post is not an endorsement or investing advice. It is personal opinions and does not reflect the views of my past, present or future employers, clients, or colleagues.