In this paper we introduce an extension of the Malmquist total factor productivity index, which utilizes the Multi-directional Efficiency Analysis approach. This enables variable-specific analysis of productivity change as well as its components (efficiency change and technical change). The new approach is illustrated and compared to the conventional Data Envelopment Analysis Malmquist approach by considering a empirical data set on Lithuanian family farms. The results highlight that important differences in variable-specific performance of the farms can be hidden when using the conventional (radial) Data Envelopment Analysis-based Malmquist index.