Reasonable people can disagree about the meaning of these ambiguous data, and it's not surprising that supporters of smoking bans like Carmona are inclined to see a clear causal relationship, while opponents (like me) are inclined to be more skeptical. But there is no excuse for the kind of scare mongering in which Carmona engaged when he implied that you could drop dead from the slightest whiff of tobacco smoke.
Even supporters of smoking bans, such as longtime anti-smoking activist Michael Siegel, faulted Carmona for gilding the lily (blackening the lung?) by saying things such as, "There is NO risk-free level of secondhand smoke exposure." This position contradicts the basic toxicological principle that the dose makes the poison. Since it's hard to measure even the health consequences of heavy, long-term exposure to secondhand smoke, how could one possibly demonstrate an effect from, say, a few molecules? "No risk-free level" is an article of faith, not a scientific statement.
Speaking of which, Carmona was at pains to say he was merely summarizing the science, not making policy recommendations, even though he emphasized that smoking bans are the only way to eliminate the "serious public health hazard" posed by secondhand smoke. He is right about this much: The issue of what the government should do about secondhand smoke is independent from the issue of exactly how risky it is. Whether smoking bans are a good idea is a question not of science but of values, of whether we want to live in a country where a majority forcibly imposes its preferences on everyone else or one where there is room for choice and diversity.