Much has been written about the power of mobile technology to transform patient care, and many innovative developers have produced apps that really do move us forward. But sometimes our enthusiasm can blind us to the trash that people continue to download onto their tablets and smartphones.

A case in point: iSAD, an app that allegedly reduces the signs and symptoms of Seasonal Affective Disorder (SAD), a serious form of depression that strikes during the winter months. While there's solid evidence to show that light therapy can improve this condition, there's no way an app that lights up your cell phone will have any effect on the disorder. A 3G iPhone, for instance, can generate only 200 lux of light, but SAD patients need at least 2,000 lux to experience any benefits.

Apps that claim to benefit patients suffering from clinical depression join a long list of other scientifically groundless online tools for weight loss, stress management and the like, a topic I recently addressed in a slideshow titled "Is That Healthcare Website Making You Sick?"

[ Wearable devices equipped with sensors and Web connections help consumers track health and fitness. Take a look at what's possible now. 10 Wearable Devices To Keep Patients Healthy. ]

Unfortunately, consumers aren't the only ones who need to be cautious. Clinicians may be tempted to load an app that on first blush will make their jobs easier, such as a smartphone app that lets them view X-rays -- not a good idea considering the fact that the screen is simply too small. Some experts even recommend against viewing medical images on an iPad because of its size and relatively poor resolution.

All this criticism begs the question: How do you choose an effective mobile health app? The best approach is to use the same criteria that discriminating physicians have been using for years when evaluating new medical treatments. Here are three criteria worth applying:

--Look for controlled clinical research that supports the app. Many developers haven't done enough homework or linked up with a medical school faculty that can test their product, but some have. The WellDoc Mobile Diabetes Management app comes to mind because the company has done research to prove its value. Similarly, apps classified as medical devices by the FDA require rigorous testing. AirStrip's fetal monitoring and cardiac monitoring apps fall into that category.

--Demand physiological plausibility. The claims made for any app must, at the very least, be consistent with the laws of physics and known facts about how the body functions. Granted, for a buyer to make a judgment call on physiological plausibility sometimes requires in-depth knowledge of health science. If you feel you're out of your comfort zone, turn to a medical expert you trust.

--Don't rely too heavily on anecdotal evidence. Testimonials from famous athletes and actors turn heads, but they rarely prove that a treatment protocol, or a mobile app, can cure your baldness or lift your depression.

That said, let me take a moment to contradict myself. The history of medicine is filled with innovative scientists who developed invaluable treatments that work but that were initially condemned as snake oil. No doubt there are health apps on the market that fall into this category. They may have no clinical research to support them and don't make a whole lot of sense conceptually -- but they still work.

But such unappreciated geniuses and their apps are few and far between. The vast majority of apps that find their way onto your radar screen call for a healthy dose of skepticism.

Clinical, patient engagement, and consumer apps promise to re-energize healthcare. Also in the new, all-digital Mobile Power issue of InformationWeek Healthcare: Comparative effectiveness research taps the IT toolbox to compare treatments to determine which ones are most effective. (Free registration required.)