It's pretty clear at this point that any health care reform which passes is going to have more voters against it than for it.  You can argue that voters aren't educated enough and that you can generate good poll numbers for individual components of the plan, but that's not really relevant.  You can generate support for nearly any program, if you poll it without mentioning the associated costs.  When voters think about the health care plan, they're not thinking public option + Medicare advantage cuts + etc.  They're making a judgment about what the entire package will mean.  And the entire package has risks as well as benefits:  higher taxes, less generous health coverage for the majority of Americans who already have it.

Jay Cost thinks risk aversion is a major factor:

The final factor is risk aversion. Recent polling has shown that most people are satisfied with the health care system. Rasmussen recently found that 49% rate it as "good or excellent" while just 27% rate it as "poor." Gallup's numbers are not as positive, but still suggest that most Americans are generally all right with the system as it is.

This might make them especially nervous about the risks inherent to the reforms. If somebody has a 50-50 shot at winning $5,000 or can take $2,500 for certain, what will he do? A risk neutral person will be indifferent between the options. But a risk averse person is acutely uncomfortable with the uncertainty, so will instead take the sure thing. A similar psychological phenomenon might be in place here. If somebody really abhors the uncertainty inherent to a comprehensive overhaul of a system that he thinks is generally all right - he might count it as one of the costs.

Importantly, risk aversion can vary according to the stakes. If somebody has a 50/50 shot at winning $1 million or can take $500,000 for certain, what will he do? A risk neutral person would be indifferent. But most people's risk aversion will make them eager to take the sure thing. People are extremely risk averse when it comes to health care precisely because the stakes are so high.

He doesn't mention some powerful evidence in favor of his claim:  loss aversion.  People react much more strongly to potential losses than they do to potential gains.  That was probably a pretty good heuristic on the veldt, but it's not as effective in modern society.  People are willing to take wild gambles when they are facing the certainty of a large loss, but otherwise, they tend to want to hunker down and stick with the devil they know.  That makes it hard to pass programs, and even harder to reform them once they have been passed.