While there's no question that inhalation of contaminated air can have negative long-term health consequences, it's difficult to directly link ambient exposure to particulates with acute hospitalization. It's also interesting to note that not all PM2.5 (airborne particulates 2.5 micrometers or smaller) are the same, and can be made up of various different compounds with (theoretically) different health implications.
To determine which of the heterogeneous PM2.5 sources/components were most closely related to hospitalization, a group of researchers analyzed filter samples various air quality testing stations, identified sources and major contributors of PM2.5, and then looked at whether there was any correlation between hospitalization and PM2.5, and whether specific particulate sources were more likely to contribute to hospitalization.
Unsurprisingly, they found that total PM2.5 was positively correlated with hospitalizations. Interestingly, they also found that specific PM2.5 constituents were more likely to cause specific negative health events than others: PM2.5 originating from road dust, calcium, black carbon, vanadium, and zink were positively correlated with cardiovascular hospitalizations, while PM2.5 road dust, sea salt, aluminum, calcium, chlorine, black carbon, nickel, silicon, titanium, and vanadium were all positively correlated with respiratory hospitalizations. An increase in PM2.5 concentrations of 1.73µg/m3 led to an increase in cardiovascular hospitalizations of 2.11%, and an increase of 3.47% in respiratory hospitalizations.
This is the type of research that helps move environmental health legislation forward. Rather than focusing on the well-known idea that air pollution is bad for you, and that smaller particles are more likely to cause negative health events, the researchers did some fairly extensive mathematical modeling to identify what specific compounds cause which specific types of health events. This type of data can allow legislators and health professionals to focus their efforts and (limited) resources on mitigating health events that are more likely to occur, given the knowledge of what compounds are prevalent in their specific geographic regions.
Source: Bell ML, Ebisu K, Leaderer BP, Gent JF, Lee HJ, Koutrakis P, Wang Y, Dominici F, Peng RD. 2014. Associations of PM2.5 constituents and sources with hospital admissions: analysis of four counties in Connecticut and Massachusetts (USA) for persons ≥ 65 years of age. Environmental Health Perspectives 122:138–144.