Gregory House

Emergency Room or Everyday Room?

Picture the scene: you are pregnant, it’s Saturday, and you woke up with one of the worst coughs in your life, your nose is running, and you have a slight fever of 38.5°C. After a few hours of dry cough your chest starts to hurt and you feel unusually tired. Over the course of the day you already tried calling your primary care physician, but during the weekend only the voicemail is available to take your call. Your primary care physician’s office opens on Monday and you are questioning yourself whether you can wait until then.

Familiar, isn’t it? This dilemma plays in people’s houses every day across the country. A trained medical professional can, with high dose of confidence, decide whether a given set of symptoms warrants an Emergency Department visit. Healthcare systems often ask regular people – patients – to make that distinction… and it is often harder than it seems. In the situation above a patient is ostensibly presenting with chest pain, fever, and generalized weakness. Just one of these descriptors – chest pain – usually qualifies patients to be evaluated in an emergent manner. However, in the case described here the chest pain could easily be explained by a full day of coughing… but how does that “chest pain” feel? … … …

Patients are NOT at Fault!

Almost everyone can relate to a feeling of walking to a busy ER, being evaluated promptly, to be then told it will be hours until any further treatment is administered. While we, as healthcare professionals, sometimes blame patients for coming to ERs with – to us – benign matters, the real reason is rather systemic. It is often challenging to balance erring on the side of caution with healthy skepticism. This dilemma is especially crucial in case of pregnant patients, where benign-appearing matters can quickly turn into real emergencies. Observing countless of patients transferred for evaluation from regular Emergency Department to the more specialized, and thus more resource-constrained, OB-ED I seen a whole spectrum of cases.

OB-ED: Like ED, but Even More Urgent

True emergencies in pregnancy can develop in minutes or even seconds. This is why some specialized OB-EDs refrain from using a regular triage, opting for more specialized evaluation right away. This obviously requires more specialized personel and equipment. These resources are expensive. A reliable research data on the matter is difficult to locate. From the conversations with medical providers of one of the local OB-ED, evaluation procedures happening within minutes in the OB-ED are often procedural and almost always involve a specialist in a narrow field.

While upwards of 10% of patients leave regular EDs before being seen ^{(2017-19, JAMA)}, we cannot let that happen in an OB patient. We need to evaluate these patients even more throughly. In one of the local hospitals, an OB-ED patient is almost always evaluated by one or more physicians, one or more advanced practice provider, interacts with multiple midlevel providers, and countless other supporting staff. While the costs may be challenging to reduce on that level, I believe there is a potential for improvement.

The Need for OB Patient Education

A preliminary needs statement should appears obvious to an experience reader:

Pregnant patients have a problem in identifying when to visit OB-ED, causing overcrowding of specialized facilities, and could be better informed about resources usage.

The statement appears to fit industry-standard POO model:

• Population: Pregnant patients
• Opportunity: Overcrowding of specialized facilities
• Outcome: Could be better educated on the resources usage

The idea of improving that aspect also appears to be feasible as well. Everyone who ever visited a healthcare facility most likely remember receiving a stack of papers after the visit. Somewhere in this packet, most facilities also include elements of patient education. There are multiple providers of such standardized and legally-sound materials. HealthWise appears to be one of the main providers in the space. However, in the author’s personal opinion, the materials offered appear long and dry. Another provider, Everwell, takes a more interactive approach but appears to focus more on “coaching”-style resources usable during the visit. All in all, a lot of resources while accurate, give impression of lacking the necessary brevity and attention-grabbing UX.

Wait, Is This Actually a Problem?

The question of viability and need for a solution should always accompany any problem that appears to lack a suitable solution. While the overuse of OB-EDs does not seem to be described sufficiently by literature, the regular EDs were studies extensively. The impact of reduction of preventable visits could be tremendous. Let’s look at just the financial side of the issue, through the perspective of industry-standard TAM:

• A visit in ED is expensive, on average at ~$1800 each
• Every year, there are nearly 140 MILLION ED visits in just United States
• ….and around 50% of ED visits are preventable!

This alone costs us around over 250 Billion USD…. over $250,000,000,000. This is a big number, which could potentially be reduced, even if partially.

The Scope

Looking through the problem, it appears that the original needs statement, could use some improvement. While usually we tend to limit scope, I believe in this case the scope can be expended for the greater good:

Patients with health concerns have trouble identifying appropriate healthcare facility, contributing to extensive healthcare systems operational costs, and should be provided with better resources to make an informed decision.

This statement appears to better address the needs, according to the POO model:

• Population:
  • “Patients with health concerns”
  • This expends scope to all potential ED patients. In addition while being more broad in one way, is also more specific to patients with health concerns and not just any pregnant patients.
• Opportunity:
  • “(…) trouble identifying appropriate healthcare facility (…) contributing to extensive healthcare systems operational costs”
  • This opportunity more clearly defines the actual problem, while not placing blame on patients. In addition, it more clearly shows the impact of *contributing* to higher healthcare *system* cost
• Outcome:
  • “(…) should be provided with better resources to make an informed decision”
  • The outcome, again, attempts to emphasize patient autonomy and need for better resources. This puts the burden of change on the systems rather than individuals, creating a more realistic opportunity.
  • In addition, specifically addressing “better resources” creates an implicit drive towards more agile process, hopefully creating a more continuous closed loop of improvements

But is this even possible?  Stay tuned to learn more.

Technology is everywhere… or is it?

As we explored before, technology is certainly wonderful and offers great possibilities. However, there’s a significant and often overlooked challenge in implementation of modern solutions: equality of access. In the pursuit of modernization of healthcare patients from most vulnerable populations are at times overlooked. In the course of innovation it’s often convenient to see access to a most basic technological thing – the Internet – as universal. Yet, over 22% of Americans living in rural areas lack access to broadband Internet!

Economic and Racial Disparities in Access to Technology

Healthcare disparities based on economic and racial factor are widely described in the literature and recognized by US government agencies. I decided to look more closely at CGM access itself. Our colleagues from Northwestern University, also from Chicago, authored a great paper on the subject. In the work titled “Use of Continuous Glucose Monitors to Manage Type 1 Diabetes Mellitus: Progress, Challenges, and Recommendations” by Friedman JG, Cardona Matos Z, Szmuilowicz ED, and Aleppo G presented a set of multiple difficulties in CGM usage.

Amongst bias in prescribing practices, insurance issues, and CGM adherence challenges, the paper also looked into technological requirements and technology access with respect to overall income.

Technology Requirements Meets Reality

Effective use of CGM technology requires a smartphone with Bluetooth capabilities and consistent internet access for data upload. Many individuals, particularly in low-income or rural areas, lack these resources, creating a barrier to CGM usage. This technological gap exacerbates health disparities, as those without access to these tools cannot fully benefit from CGM technology​.
Beyond the findings of the paper authors, my personal experience shows that the 22% figure mentioned above is just tip of an iceberg. Many patients seen in the clinic technically do have access to broadband Internet. However, their smartphones or computers are often too old or not powerful enough to support required interactions with medical devices.

Global Perspective

Authors of the aforementioned also offer a glimpse into an interesting topic of technology availability in lower- and higher-income countries. It’s clear that access to CGM is limited in lower-income countries, due to public funding not covering them. This further led to a “gray market” where CGMs are resold from higher-income countries to lower-income ones, often not in the way intended by manufacturers​, with questionable practices surrounding that. My personal experience shows a similar pattern in a more local setting as well.

Doing It Cheaper?

The healthcare space is filled with innovation. Naturally such innovation involves heavy legal protection. There are many core patents guarding CGM itself. One of the examples is work by University of Virginia Patent Foundation. The patent itself describes practically a vague core concept of CGM, including but not limited to:

• estimating glucose level based on quick checks
• extrapolating in the future
• denoising output
• calibration

With such a wide scope, and many similar patent application being fileld for other core parts of the CGM technology, entering the market is extremelly challenging. This unfortunately also tends to increase the cost of the device’s R&D, which later will be passed at end-consumers, perpetuating economic disparities.

Oh Technology…

Technology surrounds our lives and penetrates every aspect of it. We often don’t think about the impact its presence has on everything we do at home, work, or during our leisure time. When we tend to begin seeing the technology is when it fails us. Technology in healthcare occupies a very privileged position, as many processes are nearly impossible without a given system or device being present. While medical devices are certainly prepared to fail-safe, processes are often planned without a second thought of human interactions.

Technology & Patients

One of a wonderful pieces of technology are various Continuous Glucose Monitoring (CGM) systems. In an ideal world, a patient gets a small wearable sensor measuring blood glucose. Then, such a sensor reports data wirelessly to a smartphone, where it is being processed and later uploaded to a centralized system. Care providers can later access summarized and contextualized data from a given time period. This works wonderfully and provides valuable insights into patient’s health, not possible before.

Accessibility Challenges

But what if a patient cannot afford a sufficiently modern smartphone, or the data plan, or maybe they have some privacy concerns? The whole system screeches to a halt. Let’s look at the scenario in a more standardized approach, as the manufacturer ostensibly prepared for such a scenario:

• Activities: uploading data collected offline
• Environment: clinic, during patient’s visit
• Interactions: connecting the device to a computer to upload data manually
• Objects: Windows-based computer; collection device
• Users: care provider

Testing in the Real World

The process of offline data upload works perfectly in a laboratory/testing scenario. However, in practice it is severely limited-to-impossible in the real world. As highlighted above, the data is meant to be uploaded in the clinic – a place where data security is of uttermost priority. Computing devices in healthcare setting are often either severely locked down, thus making impossible to physically plug anything into them, or even utilize so-called terminals. This is however just a tip of an iceberg, as most devices on the market require not only plugging the device, but installation of custom drivers only available for a single operating system.

Clinicians are there for patients

So, how care providers solve the problem? Leaving patients without care they deserve is naturally not an option. This leads to care provider’s private devices being used to upload patient’s CGM data. While the solution works, it is very far from optimal. The solution unfortunately doesn’t seem to be readily available.