236T568, Even a sub-score improvement is meaningful for the patient. Patient 2 stated that she feels improvements (despite the near 0 probability for some improvement due to the severity of her injury!)
For patient 4 and 5 NVIV stated that they do not see a plateau.
I think that your statement that there is only an improvement when the score improves is to rough.
And even when you reduce everything on the score: 2 patients out of 5 had tremendous improvements. This means 40%. The historical data benchmark is 16%. Your evaluation that 2 out of 5 is bad is not shared by the FDA that would grant the Device for a probable benefit.
So why dropped the price: Shorts covered the stock because they were uncertain about the outcome. After the news they again dumped the stock as they learnt that they have time till end of 2016 regarding acute and end of 2017 regarding chronic spinal injury
Market Data 
Markets 
AI
