Abstract: In this paper we propose a new approach to detect changes in persistence in fractionally integrated models based on recursive forward and backward estimation of the Breitung and Hassler (2002) test, in the spirit of the approach suggested by Leybourne, Kim, Smith and Newbold (2003). Asymptotic results are derived and the performance of the new procedures evaluated in a Monte Carlo exercise. The finite sample size and power performance of the procedures are very encouraging and compare very favourably with available tests, such as those proposed by Hassler and Sheithauer (2009) and Sibbertsen and Kruse (2007). The performance of the test together with its simplicity of application make it an interesting approach for empirical analysis.