Information Overload in Medicine — It's getting worse

Welcome to the inaugural blog post for MedCurbside! In order to understand our mission, it’s important to understand the problem MedCurbside is trying to solve. It’s a problem of growth of medical information without a matching advancement in information management tools.

This is part 1 of a 3 part intro series.

How it starts

Medicine, perhaps more than most fields, requires you to assimilate large amounts of information. If you’re a physician reading this, this is obvious to you. For any non-healthcare-provider readers, a couple facts to illustrate this point: A physician’s education is at least a 7 year path (after finishing college), and oftentimes more like 8–10 years, before practicing as a fully qualified and independent doctor. The time requirement is largely a result of the extensive knowledge base required to be a competent physician; and the only way to absorb that knowledge base is by experiencing it over a long time (there are non-intellectual components to the training also, like developing procedural skills, patient rapport, personal practice style, etc. But, these are all still some form of knowledge development).

Even more important than the depth of medical knowledge, is the breadth of knowledge. There are far more specialties and subspecialties now than ever before. This trend is not particularly surprising, nor is it unique to medicine. As human knowledge expands individual fields continue to subdivide and specialties are born. We no longer have just “computer scientists”. We have computer scientists who focus on web development, operating systems, artificial intelligence, and so on. No longer do physicians simply go into anesthesiology or internal medicine or surgery. Rather, a large number of trainees will complete these residencies and then further sub-specialize. An anesthesiology resident may train further in pain management or pediatric anesthesiology. An internal medicine resident may further train in gastroenterology, cardiology, or rheumatology. The scope of knowledge in any one of these fields did not used to warrant specialized training, but now it does. This has effectively narrowed any one physician’s focus while still requiring a great depth of knowledge.

With increasing specialties comes a change in how an industry operates. Look again to the engineering industry: A software design company hires individuals with specialized knowledge such that the whole team can collectively work together to make one final and complete product. The final vision can only be realized by relying on expertise from different individuals.

This same trend is evident in medicine. Comprehensive patient care still requires a broad knowledge base. But with the segregation of information amongst specialists, often times the only way to provide this care is for any one patient to have multiple physicians. How these physicians work together varies based on context: In the hospital setting they consult each other and in the outpatient setting they refer to one another. If you are a patient or a physician you’ve likely been involved in this dynamic.

How it gets complicated

In medicine there is a spectrum of information management tools. Consultations and referrals represent one end of the spectrum (left end of the diagram). They are used in one particular context: when a specific patient’s condition clearly becomes complicated enough to necessitate the involvement of a specialist in that case. They allow for a robust two-way exchange of information between healthcare providers.

If consultations represent one end of the spectrum, then the other end is represented by a myriad of literature tools: journal clubs, review articles, summary services, textbooks, etc (far right end of the diagram below). These include several services that are almost household names such as Dynamed, UpToDate, PocketMedicine, Micromedex, etc. These tools all share one thing in common: They take several well-read experts and ask them to create comprehensive reviews in a topic specific fashion. You can easily find an UpToDate article about the management of diabetic ketoacidosis, the outpatient management of hyperlipidemia, and the management of atrial fibrillation. These are excellent resources for physicians both in and out of training. However, they only provide a one-way flow of information. You must simply read and absorb generalized information, whereas consultations involve the active and dynamic exchange of information about a specific (non-generalized) case.

Now that we have addressed the two extremes of the spectrum we get to the most important point: The tools available for the middle of the spectrum. Healthcare professionals often find themselves with clinical questions that aren’t exactly addressed in literature and aren’t exactly complicated enough to warrant a consultation. This state of limbo is so common in fact that we have developed one often-used tool to address these questions: The “curbside” (middle of the diagram). For most physicians this is a familiar term. Frequently, we find ourselves saying we’ll “curbside endo (endocrinology) about this TSH” or “curbside ID (infectious disease) about the antibiotic duration”. This means we will find an appropriate colleague and ask them the question when they have time to discuss it.

In modern medicine there is a lot of available information. There is primary literature, review articles, and the knowledge and experience of specialists. A curbside is so valuable because it can combine all these within the context of a particular question. It offers a robust two way exchange of information that is context specific but doesn’t have to be limited to a particular case or utilize formal resources like a consultation. This middle part of the spectrum is where daily practice takes place, which makes it so relevant to actual patient care. In the example above, an ID physician may provide input on antibiotic duration in a patient with an uncommon implant; something the inquiring provider may not know how to easily lookup. This info is so valuable because it immediately addresses the question, allows discussion, and can potentially be backed by evidence and specialist experience.

But it’s not without its problems. There are generalists, specialists, and mid-level providers nowadays: pharmacists, respiratory therapists, and wound care specialists to name a few. The variety of clinicians, specialists, and routes of communication may make it seem like there are more available resources but it also fuels the segregation of information. This can actually make it more difficult to rely on a curbside because connecting with the right resources at the right times is challenging without tools to facilitate these connections.

On top of all this take into account geographic, economic, and logistical limitations to connecting all these resources together:

• How does a primary care provider in a small town without a large hospital and minimal resources discuss common long-term complications for a new type of procedure performed by interventional radiologists (a specialty which the local hospital may not even have)?

Moreover, modern medicine demands evidence over opinion, so a random answer from one provider may not always be fully well informed or supported by literature. A curbside is also ephemeral. The information exchange is between two people, lost afterwards, never to be preserved or shared for the benefit of others. These reasons, among others, are why so many providers are hesitant about curbsides.

All these issues lead to the heart of the problem with this tool (and other tools) in middle of the spectrum: Despite there being a lot of available information, the dissemination and organization of information is sloppy. The information is so isolated, there is no way to search through this cumulative knowledge to find the one piece of information you are looking for.

How to solve the problem

The tools available today at either end of the spectrum aren’t practical for the middle. Generalized references/primary literature lack context and community/colleague expertise. Consults are impractical for every little question that arises during an average clinic day. A curbside and other middle-spectrum tools have their own problems. In short, if you have a specific question, then you’re stuck, hoping to find a resource that will connect you with the information you need.

What we really need is a system that fills the niche of the curbside and solves its problems… A system that aggregates information and makes it available on a scale comparable to the depth and breadth of available medical knowledge… A system dynamic enough to keep up with the pace of millions of providers’ daily practices… A system that intelligently connects providers to otherwise segregated information… A system that overcomes geographic and other logistical limitations… A system that encourages the modern standard of evidence based decisions while still benefiting from provider experiences…A system that builds on community knowledge and input… A system that encourages proper use of curbsides and dissuades improper use.

The problem boils down to this: we have a lot of information and poor mechanisms to access that information. I didn’t address any tools, other than the curbside, that might be available for this problem. Moreover, I described very generally what a good “system” can do, but didn’t give any specific examples. These omissions are so influential to the design of Medcurbside.com that I discuss them separately in part two of this article. Please check it out to see why we think MedCurbside can help make medicine better!