Some research undertaken here in Australia and published in 2011 showed that pumps will deliver a slight excess of insulin during the climb phase of a commercial flight due to dissolved gasses coming out of solution and forming bubbles. Conversely, on descent, not enough insulin is delivered due to the bubbles dissolving again. The pressure change on a commercial flight is similar to flying a light plane from sea level to 7000 feet.
In this study, the amount of excess insulin delivered on climb was proportional to the amount of insulin remaining in the cartridge, and averaged about 0.6% or a bit over 1 unit of insulin. The effect of this on subjects blood sugar level of course varied due to their insulin sensitivity. (If it was me it'd lower my BSL by 3mmol/L). The deficit of insulin delivered on descent was about 0.7 of a unit, leaving a net excess of 0.3 units.
The recommendations made by the authors, and widely adopted by endocrinologists and diabetes educators here now is:
1) The cartridge should only contain 1.5 mL of insulin.
2) Disconnect the pump before takeoff.
3) At cruising altitude, take the cartridge out of the pump and remove any air bubbles before reconnecting.
4) After the airplane lands, disconnect the pump and prime the line with 2 units. Then reconnect the pump.
5) During flight emergencies involving cabin decompression, disconnect the insulin pump.
In your son's case I think the effect will be minimal. If the flight is up and down in 20 minutes, and the maximum altitude is 3-4000 feet, any bubbles will be very small and short lived. If you're worried, just keep a closer eye on his BSLs in the one to two hours after the flight, as this is when a hypo will occur due to any excess insulin delivery. The benefits of keeping the pump connected probably outweigh the very small risk of a hypo.