Mexico Medical Student |

Dr. Val Jones at Revolution Health yesterday wrote a scathing review of freeMD.com, a site that offers to triage a person (for free!) and determine if they need to see a doctor. Our conversation actually started on Twitter where I read that last part to mean, “alerts you to seek medical attention.” I had reviewed a site like freeMD.com about 6 months ago and thought I’d already seen it (I can’t find the URL for the site I thought it was, but it was vastly different, focusing more on disease education than decision-making). I played devil’s advocate for Dr. Val’s test case, a female with abdominal pain (read her post for details), by saying that using a tool like freeMD could very well be a good thing, causing someone to seek medical attention for a condition they’d not thought of (ectopic pregnancy) that could very well warrant medical attention. She has since made some more comments, all extremely critical, and I thought I should finally go and see what the hubbub was about.

The first thing that struck me (apart from the realization that it was a different website than I thought) was the founder of the site, Dr. Schueler, in a video stating, “I’ll help you decide if you need to see a doctor and why.” Already, my alarm bells are blaring. We’re not even talking about a physician encounter via IM, email or other faceless medium–we’re actually talking about a disembodied algorithm given the look and feel of a doctor “there for you.” The whole thing makes every attempt, down to the good doctor reading the text questions in a video (marketed as “Breakthrough technology”) to give you the false security that a real doctor is guiding your care. To make matters more slimy, the footer of the website has the following disclaimer:

freeMD is provided for information purposes only and should not be used as a substitute for evaluation and treatment by a physician. Please review our terms of use.

Why doesn’t Dr. Schueler speak THOSE words just as clearly as “I’ll help you decide…?” Perhaps because the whole incongruous nature of the site, the entire clashing dichotomy is that FreeMD markets itself to be a unique service in healthcare yet disavows itself of the veracity of its own advice.

The Happy Hospitalist did his own mock patient for this site, and it was a doozy. Happy and others are intelligent, experienced physicians who can come up with all sorts of clever zebras to outwit the system. I’m just a med student unable to draw from a well of complicated experiences; nevertheless, I find this approach pointless. Happy’s mock 70-year-old pt in ARF is unlikely to be at home clicking on a website, as would anybody worthy of being in an ICU, volume depleted, electrolyte deranged, in respiratory distress, etc. FreeMD is obviously for 1) a younger, presumably ambulatory crowd, 2) people who would already go to the Internet to research their problem in the first place. I fit that bill, so I put it through its paces. My findings after a few rounds:

• Questions are often “yes” or “no” with no ability to say “I don’t know” when that lack of knowledge is significant. Asking a woman “are you pregnant” or “did you miss your last period” when they are irregular in the first place (and without asking if they are regular) is such an example.
• Questions such as “Do you use cocaine (yes/no)” can be replied “no” when the person uses meth or any other number of stimulant drugs that would elicit similar results.
• As in the above, questions don’t go from general to specific as a normal H&P would proceed but pulled out heuristically from a database of questions at random levels of specificity. Had I been asked if I used illicit drugs first, I could then be asked which ones, or “no” would skip the whole hierarchy.
• Most egregiously, in no example was I ever asked about prescribed, pharmaceutical drugs. How can this thing triage anybody without taking to account the medications they are on? Give me a break. This is a dealbreaker in the first degree.
• A severe headache that’s described as the “worst I’ve experienced” is immediately shortcut to the end of the interview with a possible subarachnoid hemorrhage. Whoah! While I applaud the good doc for not letting a serious one like this fall through the cracks, no question of hypertension, trauma, or other medical history had been made yet. I even said the headache did not come on suddenly (and how “sudden” is that? one day? one hour?). And shockingly, no attempt to quantify pain (”worst experienced”) seems to be a real oversight for an EM physician where pain control is at the forefront of almost any triage.

OK, like I said, I’m just a med student. I’m not trying to outwit the system, yet even I can see how this thing needs serious work. It smacks of a tool that could do some good someday (yes, I see the “beta” in the upper right corner) but is a hammer looking for a nail at best, and irresponsible at worst. While writing this, I saw that Dr. Schueler responded on Dr. Val’s site and said that this tool is not attempting to arrive at a true diagnosis and while far from being perfect, it’s much better than an uncontrolled search of the internet at large. I say that for a large portion of the population, this is probably true. But again, one of the biggest offenses this site makes is an unmistakeable intent to make the patient feel like they were evaluated more thoroughly because a doctor in a white coat is “talking” to them. Take away the touchy-feely schtick and my respect for a tool being transparent about what it offers jumps up several notches. Add to that the glaring holes in logic–clinical and heuristic–that even I can see on first blush, and I’d ratchet back the “breakthrough” rhetoric in a serious way.

I haven’t been as fire-breathing as Dr. Val mostly because I want a site like this to have a place–a proper place–in the ability of patients to have an informed choice. Too many people use the Interwebs and find tripe from the likes of Kevin Trudeau, Reiki shamans, and other snake-oil scammers. At least this is based on some real protocols/algorithms used in medicine. The difference is that those same algorithms in a real clinical setting are never removed from the human element wielding them with clinical judgment. The marketed need of a physician connection is something that needs to be filled by primary care. They should be the ones an unsure patient calls, receiving the expertise of a “live” doctor that knows their history and medications to say, “Come in right away,” “Go to the ER,” or “Come see me about this in the next few days.” I’m sure I’m being naive about this point as a student not able to take into account all the financial issues of this, but Dr. Schueler sure isn’t doing this for complete altruism, so I can’t be completely wrong. Leave a comment telling me what you think.

(P.S. I commented on Happy’s post above wondering if Dr. Schueler would trust his system to triage his wife/mother or other loved ones exclusively. That certainly would cut to the meat of the marketing matter.)