Standard PN-EN 1992-1-1 for designing reinforced concrete structures gives a major priority to the issues relating to second-order effects, but presents in detail only two approximate calculation methods: the nominal stiffness method and the nominal curvature method. As regards the general method, only certain requirements and suggestions are provided. In typical situations, when the appropriate assumptions are satisfied, the approximate methods yield satisfactory results. However, in engineering practice one can come across several cases (e.g. very tall columns, columns with a cantilever for a gantry girder, and floor joists) in which the approximate methods will prove unreliable. This paper presents and discusses a procedural algorithm for analysing second-order effects using the general method. The algorithm is employed to perform exemplary calculations and their results are compared with the results yielded by the approximate methods commonly used by engineers. Moreover, areas in which the approximate methods can be unreliable are indicated. The analyses have confirmed the significant advantage of the general method over the approximate methods. Therefore it is worth popularizing this method, the more so that its calculation procedures can be to a large extent automated and dedicated computer programs can be developed.