X-Prize’s Dumb Idea to Develop A Medical Tricorder Directed At Consumers

January 19, 2012 in Health Policy, Medicine, Misc by RangelMD

Comedian Dane Cook told a great joke about the future in which “everything will be instantaneous, but the DMV will still take like nine seconds.”   So will medical care.

Americans invented the idea of  “McDonaldization” in which consumer services are standardized to be efficient, predictable, and controllable.  When we want something, we want it now!  We want to be able to understand it, predict it, and control it.  However, healthcare isn’t a service that’s amendable to McDonaldization.  Cooking and serving up fast food has a relative small number of variables that can easily be controlled and predicted such as the size of the hamburger patty and the heat and time required to cook it. The diagnosis and treatment of human illness has literally hundreds of millions of variables to deal with. This is why medical care is not anywhere near as “convenient” as fast food.

But the folks at X-prize want to change this.  Their solution is to open a competition for inventors to develop a Star Trek style “tricorder” to detect and diagnose disease just as the fictional medical tricorder was used in the science fiction series. What is the specific problem to their tricorder solution? According to their promotional video, consumers are faced with a paradox.  From the video;

You want to know what’s wrong. Is this normal?  Is this urgent?  Do I need to see a doctor? But, today, the only way to know if you need to see a doctor is . . . to see a doctor. And that’s not fast, not convenient, not easy. So you wait . . .

The average time to get an appointment is 21 days. The average visit to the doctor takes nearly 2 hours.  Cause you’ve got nothing better to do, right? And then you only receive the right diagnosis or treatment 55% of the time.

Correct. Medical care is nether efficient, convenient, predictable, or controllable and this drives Americans crazy. Part of the reason is the massive complexity of human disease. Part of the reason is how our health care system is organized with way too much emphasis on advanced medical care and procedures and too little emphasis on health care access, primary care, and prevention.

The X-prize foundation’s approach to this problem is to empower consumers and patients with a quick and easy home or outpatient mobile  device to provide real time data on critical health metrics such as vital signs to accurately diagnose disease.  The requirements appear to be only that the device is mobile and be able to accurately diagnose a set of 15 diseases. There is no information on what these “diseases” would be.  The device should also be able to give information to the patient as to whether everything is “OK” or not.

The push to develop a more patient friendly health monitoring device is certainly laudable.  However, it’s not likely that such a device will work in the way that they intend it to. The problem won’t necessarily be with limitations in technology. The problem will be in how we interpret and act on information. The limitations and complexities of the diagnostic and treatment process is something that doctors have to deal with every day. Doctors take a limited set of data and formulate what they believe is the best diagnostic and therapeutic course of action that is uniquely tailored to each patient and each  situation.  They take into account not only the data but their training, experiences, statistical probabilities, and “educated guessing”. This is why medicine is still an art.

And data is not the end point of the diagnostic process. Each abnormal data point could have multiple meanings depending on what you are looking for (the pre-test probability) and the statistical likelihood of disease in your particular population. For example, an elevated blood pressure could mean that you have hypertension or it could mean that you are anxious or because of medications or because of renal artery stenosis. A high heart rate could mean anxiety or a pulmonary embolism. Abnormal laboratory results have the same problem. A low sodium level could be caused by anxiety or a brain tumor. Anemia could be caused by a poor diet or colon cancer. Even imaging is tricky. A mass seen on an XRay or CAT scan could be benign or malignant. There is even the fixed and known possibility of false positives and false negatives for each test that has nothing to do with error or technical variability. Detecting abnormalities is easy. It’s the interpretation that is hard. To say with confidence that your device will be able to diagnose a specific disease with “accuracy” is dangerous thinking.

And who or what will interpret the data for the tricorder user? A computer algorithm can be used but is likely to come up with a diagnostic differential list that may not be accurate and will certainly not be exhaustive. This is because the initial evaluation of a patient is often followed up by more advanced evaluation and it is currently not possible to put every diagnostic modality (Xray, CT scan, PET scan, nuclear imaging, EKG, EEG, biopsy, stress testing, and exploratory surgery) into a mobile phone. Maybe in the far future we will develop a “universal” diagnostic device that includes everything but this is not one of the stated goals of this X-prize.

Even worse than diagnostic uncertainty is the false sense of security that could be provided by normal data in the setting or real disease (i.e. a false negative). The FAQ page from the X-prize foundation states that the tricorder should be able to “give confirmation that everything is ok with a consumer and notify that something is not ok (a “check engine light”)“. This approach is simplistic and absurd to the extreme. Show me a person with completely normal vital signs and blood tests and I’ll show you a patient who is has undiagnosed HIV or a smoker who has a malignant growth in their lung. The danger is that whatever limited data this device does provide will, if normal, give consumers a false sense of security and lead them to cancel that trip to the doctor.

The biggest problem is that such a device is either not required for it’s intended purpose or won’t make any difference in outcomes. Will it really be able to help people make a decision on whether or not to see a doctor or go to an ER? True medical emergencies almost never present subtlety. Ironically, PAIN is one of the biggest indicators that something is seriously wrong and the measurement of pain is not something that can currently be directly done by a tricorder or any other device.  Bleeding, passing out, shortness of breath, vomiting, decreased consciousness; patients rarely have a problem knowing when something is seriously wrong because their bodies do a much better job than any tricorder could do in letting them know. And when should you go to see a doctor in an non-emergent setting? The best way is to use a simple rule of thumb. If you have to consider it then likely you should go. And as far as prevention, do we really need a device to tell us to stop smoking, loose weight, and get some exercise?

There yet may be some uses for such a device. More frequent monitoring of metrics like blood pressure and blood glucose levels in the outpatient setting may, if accurate, help doctors to better control such chronic conditions as hypertension and diabetes and better control can lead to better outcomes. Cheaper and mobile medical devices would help health care workers enormously in areas of the world that are remote and/or lack access to modern medical care.

But to develop a device with the stated goal of “empowering” consumers in their health care decisions by helping them know when they can avoid the inconvenience of having to see a doctor is amazingly shortsighted and ridiculous. And it’s potentially dangerous by giving people a false sense of security that they do not need to see a doctor.

I foresee one main outcome of the development of this tricorder. The typical civilian users of such a device are likely to be suburban, middle aged, relatively healthy and health obsessed. These are the types of patients that doctors in affluent areas dread. These patients Google the differential diagnosis for every single symptom or strange noise coming from their bodies and make frequent trips to their doctor with pages of printouts about their “condition”.  Congratulations X-prize! You’ve begun the process to develop the most important enabler for hypochondriacs that the world has ever known.

Please share.