Atmospheric water vapour is an efficient greenhouse gas and therefore important in the climate feedback process. Ground-based GPS networks, such as the IGS network, have been identified as a useful technique to obtain long-term trends in the integrated amount of water vapour (IWV) in the atmosphere [Wang et al., 2016]. The selection of the elevation cutoff angle used in the GPS data processing will have a significant impact on the resulting IWV trend if there are elevation-angle-dependent errors that vary with time. Using 14 years of data from 12 GPS sites in Sweden and Finland , Ning and Elgered  found that a higher elevation cutoff angle (25 o ) gives the best agreement between the GPS- derived IWV trends and the ones obtained from profiles measured by radiosondes at nearby launching sites. In order to address this result, we have now processed 20 years of GPS data from 8 sites in Sweden and 5 sites in Finland, using two different elevation cutoff angels, namely 10 o and 25 o , to estimate the IWV. We have also studied the impact of three additional elevation-angle- dependent effects in the GPS data processing, i.e., (1) use of two different mapping functions, (2) with or without applying second order corrections for ionospheric effects, and (3) with or without applying elevation dependent data weighting. The results show that all these three parameters have insignificant impacts the resulting linear IWV trends. We compared the GPS-derived IWV trends to the corresponding trends from radiosonde data at 7 nearby (< 120 km) sites and the trends inferred from the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis data (ERA-Interim). The IWV trends given by GPS elevation 10 o and 25 o solutions show similar results when compared to the trends from the nearby radiosonde data, with correlation coefficients of 0.71 and 0.74, respectively. When compared to the IWV trends obtained from ERA-Interim, the GPS solution for the 25 o elevation cutoff angle gives a higher correlation coefficient (0.90) than the one obtained for the 10 o solution (0.53). The results indicate that a higher elevation cutoff angle is meaningful when estimating long term trends, and that the use of different elevation cutoff angles in the GPS data processing is a valuable diagnostic tool for detection of any time varying systematic effects (unwanted), such as multipath at the receiving antenna.