In the US there are 3 or 4 standards in use.
One of them is the EPA Hydronic Heater Testing Program Requirements (Phase I and II). The test program is available for free; see this link
http://www.epa.gov/burnwise/pdfs/owhhphase2agreement.pdf
Then there are some ASTM standards which I can not provide to you, they are copy righted.
All US testing protocols are more or less the same in essence, some details are different.
The OUTPUT BTU's are the BTU's transfered to the water. This is measured by logging the Flow (GPM), Deta T and Time.
The INPUT BTU's are the BTU's available in the fuel. For the EPA hydronic heater program red oak with 8550 BTU/Lbs on a dry base is taken as the reference value (higher heating value).
The efficiency is the Output divided by the input.
Also for the EPA hydronic Heater testing protocol, the amount of fuel is determined as follows: you calculate the firebox volume in cubic feet and multipy this by 10. This gives you the total weight of wood you need to put into the boiler to perform the high burn and all other test runs. Example: say your firebox is 6 cubic feet x 10 = 60 lbs of red oak plus/minus a tolerance. 60 Lbs of red oak translate into 60 Lbs x 8550 BTU/Lb = 513,000 BTU's (wet). Take into account your moisture content, lets say 20%, this gives you roughly 410,400 BTU on a dry base and this is your reference value.
Then you burn the wood at full speed for the high burn, let say during 4 hours. This gives you around 102,600 BTU/hour input.
During the test you log delta T, Flow and you calculate the heat to the water, lets say 80,000 BTU's/Hour to the water.
So your efficiency is 80,000 / 102,600 = 78%
The "funny" thing about the EPA protocol is that you can achieve nearly 100% efficiencies. See list of phaseII wood boilers.
So to measure efficiencies it's not the best testing protocol out there.
In general, a test protocol based on "stack Loss" method is a better, more accurate protocol, but more expensive to test.
Hope this clarifies the topic a little bit.