Advanced time-accurate simulation capabilities for multidisciplinary problems are vital, e.g. to reliably cover the flight envelope in the design analysis and optimization of aircraft. Simulation efficiency is a key factor, particularly in combination with unsteady and high-fidelity predictions. Adaptive time-integration techniques are well-known and established in disciplinary analyses. The time-step size, which often determines the accuracy (and the computational costs, among others) of the time integration, can be controlled to meet a given error threshold according to the methods of lines. The aim of this work is to carry over the adaptive time-stepping approaches to the analysis and optimization of multidisciplinary problems. To this end, an MDAO framework approach based on OpenMDAO and the FlowSimulator HPC ecosystem was developed to allow for time-accurate multidisciplinary high-order time integration [1]. The modular integration of high-fidelity simulation plugins like the new-generation CFD software CODA plays a central role to arrive at a scalable framework solution. The focus of this paper is on the extension of the MDAO framework approach for adaptive time-stepping of coupled MDAO problems; central aspects are: • Advanced solution capabilities with nested multidisciplinary solvers are necessary to accurately solve large and stiff coupled problems in the context of implicit time stepping. It is deemed a prerequisite for the success of the time-step control. • A computationally cheap error estimator based on Runge-Kutta methods with an embedded low-order scheme for adaptive time-step control [2]. • A scalable framework implementation ready for large-scale multiphysics with high-fidelity simulation components. The suggested multidisciplinary framework approach will be investigated and verified based on numerical experiments for elementary problems first. Subsequently, coupled/error-controlled time integrations will be demonstrated in conjunction with high-fidelity simulations. The method will be evaluated for large coupled problems involving CFD simulations for aerodynamic/-elastic cases of different complexity levels. The presentation will be rounded off with an outlook on the use of the suggested framework approach for design optimization.