Digital Healthcare Innovation. Challenges and Lessons for Digital Health Startups

June, 2024
3278 words / 17 min read
By Melek Somai
786 views
In the ever-evolving landscape of digital health, recent announcements such as OliveAI are an important anchor to review the intricacies and complexities involved in healthcare innovation.

The healthcare industry, representing 17% of the US economy in 2022 and totaling $4.5 trillion in spending, is often viewed as amazingly innovative. This is particularly true in biotech and life sciences, where groundbreaking innovations such as mRNA, CAR-T, and Gene Therapies are revolutionizing how we treat patients and eradicate diseases. Conversely, Digital Health ("Health Tech"), distinct from biotech and life sciences and encompassing various digital solutions and services specifically designed for healthcare providers, is often seen as lagging in innovation. While the potential for digital health to transform patient care, streamline operations, and improve outcomes is immense, the industry faces numerous challenges. These include regulatory hurdles, interoperability issues, and resistance to change within established healthcare systems.

Despite these obstacles, there is a growing push towards leveraging technologies such as artificial intelligence, telemedicine, and electronic health records to modernize healthcare delivery and enhance patient experiences. In the last decade, a massive flux of investments has been made to modernize it. Government incentives such as the Health Information Technology for Economic and Clinical Health (HITECH) Act, which has poured almost $35 billion since 2008, created a near-universal adoption of electronic medical records (EHRs) in the healthcare industry. This signaled an opportunity to build technology solutions needed by the healthcare industry and expand digital innovation to encompass digital infrastructure and data interoperability. This inflection point was soon followed by an influx of venture capital investment, which accelerated its allocation to healthcare information technology and electronic health record companies faster than the rest of the industry, peaking at $25 billion in 2021.

Despite the stream of venture capital in the sector and the onset of digital health startups seeking attractive valuations, healthcare delivery remains an outlier in the digital transformation wave. The year 2023 has served as a stark reality check, marking a significant downturn in healthcare tech investment to just $13 billion (a 48% decline from 2021). This decline, although in line with the overall decrease in venture capital funding from a historic high, warrants a closer consideration of digital health startups' underlying challenges. For instance, Olive AI, which once raised over $800 million in venture capital and was valued at over $4 billion, folded in October 2023 after selling its remaining assets to Waystar and Humata Health. At one point, Olive AI reported a customer base of more than 900 hospitals throughout 40 U.S. states, including over 20 of the top 100 U.S. health systems, and employing as many as 1,400 people. Analogously, Babylon Health, once celebrated as the crown jewel of the UK digital health sector, going public in 2021 through a $4.2 billion SPAC, announced bankruptcy last year. While by no means representative of the digital health sector as a whole, these two examples serve as a practical case study of the fragile nature of scaling and identifying product-market fit in the healthcare tech sector.

Product-market fit (PMF) is a concept in the business and startup world that refers to the degree to which a product satisfies a strong market demand. It is a crucial stage in the lifecycle of a startup and is often considered a key indicator of potential long-term viability. PMF is achieved when a product resonates with a target audience, addresses a real and significant problem or need, and has a sustainable market where customers are willing to pay for the solution. The pursuit of PMF for digital health startups is uniquely complex. Evidently, the healthcare industry comprises multiple stakeholders with divergent incentives and a wide range of regulations. Navigating this ecosystem to develop a product that fits well within the existing workflows can be daunting and necessitates a humble acknowledgment of the intricate web that defines this sector. It's a delicate dance between bold innovation and a grounded understanding of healthcare's complexities. It is also a good reminder of W. Edwards Deming’s quote, “Every system is perfectly engineered for the results it gets.”

Emerging Health Tech Models

With those challenges in sight, digital health startups are trying to adopt a few tactics to achieve success. Two distinct models are the most common, each representing a distinct approach to achieving a true PMF: a Software-as-a-Service (SaaS) model and a Services-as-a-Software model.

Note

There is a trend of companies claiming to be platforms, but many fail to meet the true definition of platform, and as such, we are not including Platform as a Service (PaaS) models in this discussion

Software-as-a-Service

Over the last decade, many SaaS digital health startups have emerged, introducing a new paradigm where cloud-based applications and platforms are provided to healthcare organizations as a service rather than licensing their software to be installed and managed entirely by the healthcare provider's IT department. SaaS in healthcare encompasses a wide range of applications, from electronic health records (EHR) and patient management systems to advanced analytics tools and telemedicine services. These cloud-based solutions enable healthcare providers to access tools and technologies without on-premise infrastructure, reducing IT overhead. By leveraging the SaaS model, healthcare organizations can streamline operations, manage capital costs more elastically, and respond more swiftly to the evolving landscape of technology. For entrepreneurs and investors alike, the SaaS model, with its high margins and scalability, is attractive and scalable.

However, even with those lofty promises, SaaS models are challenging to implement in the industry. In the healthcare sector, where the stakes involve human lives, a natural inclination towards risk aversion prevails among providers. Convincing healthcare stakeholders to embrace new technologies demands a demonstration of innovation and irrefutable proof of safety and efficacy. This is a catch-22, where a SaaS startup must demonstrate key outcomes while validating its product-market fit. Thus, startups have limited ability to pivot, and it becomes complicated to deliver outcomes and identify the right product offering.

This is often complicated by the heavily regulated nature of healthcare. Any new healthcare technology must comply with a plethora of privacy, security, and compliance standards, which can hinder rapid iteration and scaling. While HIPAA compliance is built to safeguard patients' data and ensure that Healthcare organizations implement the appropriate security and privacy guardrails, current security practices across the industry are non-standardized and impractical. This hinders product development and the overall value of the digital health ecosystem to improve care.

Another pitfall for digital health startups is the added complexity of the B2B2C dynamic. There is usually a disconnect between the decision-maker (i.e., the healthcare system) and the end-user (i.e., the healthcare practitioner and the patients). While a startup might identify a cardiologist who shows enthusiasm for their product and champions its implementation as a proof-of-concept, the cardiologist is often not the buyer and is rarely the final decision-maker who drives adoption and procurement at the organizational level. The flip side is also problematic: securing a buyer with funding within an organization doesn’t always guarantee the successful deployment and adoption of a product by the end users. As a result, even if a startup successfully navigates the sales process, its product may be stalling during deployment, requiring additional investments from the organization, or suffering from limited adoption. This disconnect occurs when the product, while financially endorsed, fails to resonate with or delight those who interact with it daily. The bottom line is that an early interest by a healthcare entity does not necessarily equate to a viable path toward commercialization.

Beyond all, health tech companies must often integrate deeply into existing healthcare systems, workflows, and protocols, providing a complete service rather than just a platform for others to build upon. The technology stack of healthcare providers is largely dominated in Health IT by Electronic Health Records systems (EHRs) like Epic and Cerner. Founded in 1979, Epic continues to operate primarily as an on-prem monolithic operating system with a market share of more than 60% of total U.S. hospitals' net patient revenue and 50% of total hospital beds. At its core, Epic's technology is built on MUMPS (Massachusetts General Hospital Utility Multi-Programming System), a database and programming language originally developed back in the 1960s. EHRs typically function as closed operating systems, which poses significant challenges for health tech startups trying to integrate their solutions. Though there has been increasing regulation around data sharing and interoperability, such as the 21st Century Cures Act, these efforts have not been sufficient to overcome the inherent limitations of these legacy systems. Epic and the major EHR players are from another technology era and are too cumbersome and too complex to integrate with. As a result, health tech startups frequently find themselves customizing and integrating solutions for each healthcare system, which weighs heavily on their resources, often before achieving true PMF. The majority of their talent and resources are instead allocated to implementing and customizing their products for specific use cases. This means that while they may be growing in terms of customer acquisition, the resource-intensive nature of these implementations can detract from their ability to invest in and improve their core product, thereby impacting their journey toward achieving a broader PMF.

While the integration challenges with existing EHR systems present immediate obstacles for health tech startups, they may also offer a strategic advantage in the long term. Successfully integrating a product with these complex and closed EHR systems can create significant barriers to entry for competitors. Once a health tech startup has navigated the intricate process of integration and customization for a healthcare provider, it establishes a deeply entrenched position within that system.

However, the bottom line remains that growth in the health tech sector might be slower due to these integration challenges. Startups and investors should anticipate a higher upfront investment in terms of financial resources, time, and effort. The cost of integration and customer acquisition in the health tech space is typically higher than seen in other sectors, especially when dealing with the closed systems of legacy EHR platforms. This phenomenon can be approximated by examining the median time to exit since founding and since the first VC investment, which has gradually increased (though remains in flux) over the past decade with a median of 8 years and 5.6 years, respectively (Exhibit C).

Similarly, the Total Addressable Market (TAM) for Healthcare IT can be misleading to digital health startups. Healthcare IT represents only about 4% of healthcare providers' total operating expenses or an estimated $52 to $68 billion. Despite its dominant position in the Healthcare Provider segment, Epic has an estimated annual revenue of approximately $3.8 billion, a modest figure compared to the healthcare sector's total market size. In reality, a common mistake by health tech startups is to estimate the TAM based on the healthcare systems’ overall budgets and costs, overlooking the reality that IT spending is markedly smaller. This misalignment often originates from founders' belief that their product is transforming rather than competing for the existing IT budget.

The misalignment in Digital Health often originates from founders' belief that their product is transforming rather than competing for the existing IT budget.

Services-as-a-Software

The concept of "Services-as-Software'' in healthcare represents a different approach. Under this lens, care and technology are bundled together, offering a comprehensive care service that is either direct-to-consumer (B2C) or through white labeling and partnerships with existing healthcare providers (B2B2C). Startups embracing this model are, in essence, care providers. The difference is in how they leverage technology as the primary means of customer experience and care delivery. Integrating care delivery with technology gives these startups greater control over the patient experience. This usually leads to faster innovation, quicker product iterations, and, more importantly, better control and adaptability of their services to meet the needs of their patients and customers. It offers a more direct path to finding PMF, as they can directly influence and observe the outcomes of their service, making adjustments as needed along the way. Services-as-Software models in healthcare have the potential to generate significant value. They can offer personalized, efficient, and often more accessible care, appealing to a market increasingly seeking convenient, technology-driven healthcare solutions. In return, the enhanced care experiences can lead to higher patient retention and more robust revenue streams.

However, these models are limited by a major obstacle: scale. While they can generate significant value, they typically do not achieve the operating margins seen in traditional SaaS startups. Integrating care services often involves higher operational costs, including staffing, training, and compliance with healthcare regulations.

Another miscalculation stems from the belief that those models can be more efficient than traditional healthcare systems by orders of magnitude, which might not hold true at scale. Eventually, the operating margins of delivering care will converge. The main reason is that, as these startups scale, a brick-and-mortar interface is often needed. This requirement might arise due to the nature of healthcare delivery, where physical interaction is necessary to complement and enhance the service experience to be more comprehensive. Scaling in this direction can involve 1) establishing partnerships with existing healthcare providers, 2) building their own facilities, or 3) combining both approaches. Each path entails tradeoffs: Significant investment, logistical planning, and an added complexity to the scaling process.

Road Ahead for Health Tech Innovation

The broader market dynamics underline the need for startups to remain grounded in their PMF, avoid the seduction of rapid, unfocused expansion, and pursue sustainable growth tailored to the unique demands of healthcare providers.

In parallel, emergent trends could potentially create a more strategic path for startups to be more successful: (1) Healthcare systems transitioning to becoming builders, (2) Reverse engineering the investment in the Healthcare Tech sector, and (3) AI reshaping the entire technology stack for the current care delivery.

Rethinking the Role of Healthcare Providers in the Health Tech Ecosystem: from Integrators to Builders

As key players in the health tech ecosystem, healthcare providers must critically examine their role in shaping the rapidly-evolving landscape. Currently, healthcare providers are technology integrators rather than technology builders. However, recognizing the increasing importance of technology in healthcare, healthcare organizations must embark on a path to transform themselves from integrators to technology builders. This transition must involve establishing a dedicated product engineering team, a move that represents a significant step towards making technology an integral part of the business of healthcare. Advantageously, healthcare systems can combine in-house disciplines ranging from clinical services and engineering to marketing and user research. The immediate benefit is augmenting their capabilities to leverage non-healthcare SaaS products to develop custom digital care services and to collaborate more effectively with digital health startups.

Reverse Engineering Healthcare Technology Adoption

To take matters a step further, venture funds with a healthcare focus in the likes of General Catalyst (GC) and Aegis Ventures are partnering with healthcare systems directly in a new model of tech integration.

General Catalyst acquired a healthcare provider through a newly established entity, Health Assurance Transformation Corporation (HATCo). The goal is to use this acquisition as a vehicle to test, implement, and scale the products of its digital health portfolio companies. This model reverses the order of innovation: Healthcare systems become the incubation environment for their venture partners and portfolio companies rather than the end customer. For startups, this is an easy path to get their products in the hands of their users and validate their products in a shorter, more iterative, and less costly cycle.

The other important aspect of this model is an effective (in terms of value delivered) and efficient (in terms of low cost) go-to-market strategy that connects healthcare systems through their needs, meaning through the products and solutions provided. GC paraphrases this as creating the “Amazon ecosystem of healthcare” that builds “an interoperability model with technology solutions.”

Aegis’ version of this model is a “Digital Consortium,” a collective of major health systems co-creating and scaling health tech solutions. Because healthcare systems' market power is concentrated within a few players, implementing this model becomes a play on winning the few rather than going after the many. For example, Aegis has started its partnership with Northwell Health, New York’s largest health system.

The question arises whether any of these approaches represent an anti-pattern to an actual market innovation. In this anti-pattern, the relational model becomes capitalhealthcare systemsinnovation rather than capitalinnovationhealthcare systems. Innovation now sits at the edge of the relationship between capital and healthcare systems rather than being the core. It becomes the byproduct of that relationship and is relegated to a peripheral outcome of the capital-healthcare system interaction, potentially undermining its centrality and effectiveness.

AI in Reshaping the Ecosystem

Every decade or so, a new technology trend transcends the hype cycle to reconfigure and transform entire economies. This decade will likely be the decade of AI for reasons thoroughly explained in a blog by NEA.

For the Software-as-a-Service model, AI presents a potential tipping point. Integrating AI into healthcare SaaS solutions can drive a fundamental shift from traditional EHR systems, paving the way for more efficient, effective, and productive tools. AI can enhance these platforms' capabilities, enabling them to process vast amounts of data, offer predictive analytics, and provide more personalized and accurate care suggestions. This transition may encourage healthcare systems to adopt more advanced SaaS solutions, breaking away from the limitations of legacy systems.

AI could be equally transformative for the Services-as-Software model, particularly in overcoming the scaling limitations. By integrating AI, these services can offer more sophisticated virtual care, remote monitoring, and diagnostic capabilities, significantly enhancing their reach and efficiency. In addition, AI can automate routine tasks, analyze patient data for better health outcomes, and provide insights that human practitioners alone might not attain. More importantly, AI integration could enable these models to scale their services without the proportional increase in physical infrastructure or personnel typically required, thus changing the dynamics of scaling productivity and service delivery in healthcare.

With AI, we stand on the precipice of a new era in healthcare disruption, one where traditional paradigms are challenged and fundamentally reinvented, offering unprecedented opportunities for innovation and efficiency. The pivotal question is whether new entrants in the aforementioned models can harness AI to achieve >10x transformative enhancement in care delivery. If such a significant leap is not attainable, AI's role as a critical differentiator in reshaping healthcare models may be questioned. This necessitates new players to not only integrate AI but to do so in a manner that yields groundbreaking advancements, propelling them beyond mere incremental progress and solidifying AI's position as a game-changer in the healthcare sector.

Summary

The current financial climate calls for reevaluating investment strategies in health tech. The end of the once cheap and readily available capital era has led to a more conservative approach, emphasizing sustainable growth and a focus on problem-solving over-aggressive expansion. This is the right time to test the “price is what you pay, value is what you return” thesis in health tech venture investments. This slow and steady paradigm shift signals a maturation of the health tech investment landscape, where the emphasis is on building viable businesses that address tangible challenges in the healthcare sector. The market correction will likely see many startups faltering, breaking up, or consolidating. It also sets the stage for more resilient and impactful companies to emerge.

The booming health tech companies of the future are expected to have common characteristics such as solving critical problems, demonstrating clear ROI, operating in complex ecosystems, focusing on attractive unit economics, and building multidisciplinary teams. As the ecosystem evolves, a new breed of health tech ventures — patient, prudent, and problem-solving — is set to lead the way.

Contributors

I would like to thank Brad Crotty, Eya Somai, and Mike Maschek for revising and reviewing early drafts of this article.

spotify logo
Not Playing - Spotify
Made with love from Milwaukee, and Tunis.