The literature on the health effects of marriage status is substantial and of varying levels of reliability.
Kiecolt-Glaser and Newton (2001) provide a decent review. Generally, men get more health benefit from marriage than do women, but women in unsatisfactory marriages complain of more health problems than do men (or women in happy marriages).
A major question in more recent research is whether non-marital cohabitation has the same benefits as marriage. Studies on this question have had mixed results. Age seems to be a factor.
Brown et al (2005) examined an older population (over 50) and found that marriage was associated with lower levels of depression as compared to cohabitation, but only in males.
Mastekaasa (2006) studied a younger population (19-30) and found that marriage or cohabitation was associated with more distress in males 19-23, when compared to living alone. In females 19-30, and males 23-30, living alone was associated with more distress than either partnered condition. Unsurprisingly, both Mastekaasa and Brown et al found that breaking up was a source of major stress and distress for both cohabiting and married couples.
I don't think marriage and cohabitation are exactly the same; it seems foolish to argue that. However, I have been accused of shacking up at the drop of a hat. I enjoy living with my partner (in the general and specific senses). I have cohabited with four partners over the past 10 years, and if I'd been through 4 marriages in that span of time, I'd say something was very, very wrong. However, I do think there's probably a point in cohabitation where it approaches the significance of marriage, and I haven't seen any studies examining a way to divide up cohabitating couples into categories.
Musick and Bumpas (2006) comes closest with their "trajectory" analysis of relationships.