If the article on the same topic in the WSJ is accurate, this decision was portrayed as a strategic move by CVS to move further toward becoming a healthcare provider rather than simply a drugstore. CVS already has nurse practitioners providing basic care in clinics in some of their stores, and apparently CVS tries to have their pharmacists counsel people on some health concerns. It was reported that, in trying to deepen their relationship with the medical community and some medical organizations, that CVS's continued sale of cigarettes was a point of distrust among the medical community. CVS wants to prove that it is earnest in trying to become a healthcare provider.
There is a large shortage of primary care physicians in the US that is only expected to get worse (which continues to result in the utilization of physician assistants and nurse practitioners for primary care). Furthermore, slipping cigarette sales in recent years (and renewed campaigns to curb smoking in the US) make this look like a long-term tactical bet by CVS.
It's an interesting idea, but I wonder if people would expend the effort to accurately log their symptoms/recent behavior without some sort of reward. A paper published earlier this year investigated using machine learning with Twitter data to try to do something similar to what you suggested, though. (the paper: http://www.cs.rochester.edu/~sadilek/publications/Sadilek-Br...).
While concerns over an interruption in salary are certainly legitimate, just uncoupling health insurance from employment will decrease the barrier to switching jobs and increase mobility.
It's worth noting that the "job lock" that you're referring to is a rather recent phenomenon. Employer-sponsored health insurance was essentially an accident of WWII wage and price controls (source: http://www.nber.org/bah/2009no2/w14839.html). Because health benefits were exempt from WWII-era wage controls, employers used them to attract employees. Furthermore, because the employer contribution is paid in pre-tax dollars (i.e., employer-sponsored health insurance is effectively subsidized by the government), it became an attractive way to provide greater overall compensation to employees than otherwise possible.
However, the issue of coupling employment to insurance was quickly apparent, which is why we have COBRA (since 1985), but even that is more of a patch than a fix. Regardless of the merits or demerits of Obamacare, uncoupling employment from insurance is (theoretically) a positive outcome for everyone, whether one is an entrepreneur or not. We're already seeing firms shift their cheapest plans to exchanges (http://online.wsj.com/article/SB1000142405270230479580457909...). Increasing health insurance portability should make jobs less "sticky" and result in a more efficient distribution of labor.
I'm not sure I know what you mean by an externality being "covered" by property rights. An externality arises when the social cost or benefit is different from the private cost of an action. I assume you are talking about having the actor internalize the costs of the externality by expanding the scope of his property rights. However, I think the parent post was just referring to a result of the Coase theorem. Namely that, no matter the initial allocation of property rights, externalities can be traded away, as long as there are no transaction costs. In practice, negotiating away externalities is questionable for other reasons too, like contract credibility. E.g, if I were to pay someone in front of me on a plane not to recline their seat, they might just pocket the cash and recline anyway.
There is a large shortage of primary care physicians in the US that is only expected to get worse (which continues to result in the utilization of physician assistants and nurse practitioners for primary care). Furthermore, slipping cigarette sales in recent years (and renewed campaigns to curb smoking in the US) make this look like a long-term tactical bet by CVS.