Anaesthesiologist here.
I'm extremely doubtful these things will produce data that can be validated against lab data. If they can't, at best they're worthless, at worst they're misleading & potentially dangerous for those that trust them. It would be like determining the size of the waves in the middle of the ocean by looking at the waves on the beach.
There is precious little real data here to go by, so I'll be thrilled to be proven wrong of course!
Currently in the medical world we use no such superficial, peripheral optical sensors to detect*:
- Core body temperature (you need a temperature probe close to your actual 'core' - usually deep nasopharyngeal or rectal). Making assumptions about temperature drops from the core to the periphery is unacceptable.
- Blood pressure. Again, you can make assumptions about someones blood pressure using the pulse oximeter waveform, but it's just that... an assumption. Even a physical pressure transducer inside the radial artery at the wrist has measuring errors dependant on the transducer, positioning or anatomical & pathophysiological variables.
- Alcohol, lactate & glucose. I'm not aware of any medical device currently in use, or being developed that can do any of these things accurately producing data with which you can make clinical decisions. There are some optical glucose measuring devices in the research stage I think, but their accuracy isn't great.
Honestly, the game changer would be blood glucose measuring. Not because it would be a good idea in everyone's Apple Watch (I think too much data for the average punter is actually unhelpful) but for diabetics or those with impaired glucose tolerance it would be fantastic for them to have it linked to their insulin delivery devices for a truly 'digital pancreas'.
Press releases, renders & industry sponsored 'studies' mean very little when we're talking about critical health data which people use to make decisions...
* Pulse oximetry is a different beast. It's been around for 40 years and it's possible due to the very interesting differential absorption spectra of oxygenated vs deoxygenated haemoglobin at two very specific and convenient wavelengths (red & infrared). Measuring pulse alone is best done with a green wavelength as it's less prone to external interference so better for sport/movement, etc. but can still be affected by poor pulse signal and excessive movement.