I suppose you could actually get a reduction at high altitudes, but that wouldn't be in the realm of normal building codes- more in aeronautical work.
Wind speed increases with alitude above the ground, not necessarily alitude above sea level. So you should see some minor reduction in areas that are flat and elevated as opposed to sea-level areas of equal flatness.
I may be remembering wrong, but it seems like the older ASCE A58.1 actually had density in the equation.
JStephen and csd72 make good points that you need to consider. However, air is less dense at higher elevations and it is fairly significant. I don't have the formulas on hand for air density with elevation, but they are fairly available, and theoretically you could multiply your wind load by a factor taking this into account. That is if you have a good wind speed value for the area and topological factors are used appropriately. This is likely beyond the scope of any codes at least in the US.
I'm going to assume you are not "up in the air", but are near the ground at high altitude.
John D. Holmes' "Wind Loading of Structures" covers this, if I remember correctly - I do not have my copy handy. You will still be in a boundary layer near the ground, so the wind profile will still be the same as any other location, but you can reduce the velocity pressure according to the ratio of density altitude to air density at "normal" MSL conditions. Forgive me for not remembering the equations, but you'll probably find very little benefit on low rise buildings.
Many years ago a mechanical engineer gave me a table showing the density of air at various altitudes. Mechanical engineers use this to determine the efficiencies of air compressers at various altitudes. I have long since lost the table. But, I'll bet you can find it in a mechanical engineering handbook.
At one time the Denver(elevation 5,280 feet) building code used the density of air as 85% of that at sea level. I haven't checked lately to see if this is still the case. I once designed a structure at 10,000 feet (Winter Park, Colorado)and I think the density of air is on the order of 70% of that at sea level.
Considering all the guess work that goes into wind design such as wind speed, terrain factors, gust factors, and shape factors, I would be hesitant to take much of a reduction for altitude.
The air density is a significant factor. Poppo is fairly close with 85% in Denver. I work in Denver and our office uses the air density factor for everything. I belive it comes ot to .82. But read through ASCE 7. I know it is in there. It has a small section saying you can do it, and then I believe it goes into greater detail in the commentary. It's been a few years since I did this.
If you are using ASCE 7 to compute the wind loads, you can adjust the wind speed due to different air density if sufficient weather data are available to justify this. This is done by adjusting the .00256 constant used to calculate q. Check out the commentary section C6.5.10 (ASCE 7-05) for an explanation and Table C6-13 for air density values at various altitudes.
