There are several OSM format files for each area in Colorado. The road and trail data comes from the USDA's National Dataset site. The building footprints are from Bing. Address data comes from the state GIS office, and often isn't available, or has a restricted license. I only have address data for a few states, but keep searching. The conversion from ERSI Shapefiles or GeoJSON files to OSM format has been done, and the conversion validated. That process is documented here. Occasionally I find different datasets for the same data, like addresses. The county created ones are usually better, and are in the AddressPoints files.
These files are to encourage people to import this into OpenStreetMap by doing some of tedious preprocessing to make that process easier. These are the original data files broken into smaller files based on the County boundaries to make then easier to work with. You can read my tutorial on importing to learn how to import this data.
While converting the Bing building footprints, I noticed the poor quality of the data. Microsoft used machine-learning to identiy buildings from satellite imagery. This appears to have problems with mountains, and many snow fields and large boulders got identified as a building where there are none. I also found it incorrectly identifies big rocks in the desert as houses. Maybe this isn't a problem elsewhere as it appears to primarily happen with above treeline or rocky areas. I found I had to zoom in far for each building in the data to be sure what it is. If you are also using the address data, it's a little easier as you can always start by just adding buildings that have an associated address.