Medical anomaly detection is a crucial yet challenging task aiming at recognizing abnormal images to assist diagnosis. Due to the high-cost annotations of abnormal images, most methods utilize only known normal images during training and identify samples not conforming to the normal profile as anomalies in the testing phase. A large number of readily available unlabeled images containing anomalies are thus ignored in the training phase, restricting their performance. To solve this problem, we propose the Dual-distribution Discrepancy for Anomaly Detection (DDAD), utilizing both known normal images and unlabeled images. Two modules are designed to model the normative distribution of normal images and the unknown distribution of both normal and unlabeled images, respectively, using ensembles of reconstruction networks. Subsequently, intra-discrepancy of the normative distribution module, and inter-discrepancy between the two modules are designed as anomaly scores. Furthermore, an Anormal Score Refinement Net (ASR-Net) trained via self-supervised learning is proposed to refine the two anomaly scores. For evaluation, five medical datasets including chest X-rays, brain MRIs and retinal fundus images are organized as benchmarks. Experiments on these benchmarks demonstrate our method achieves significant gains and outperforms state-of-the-art methods. Code and organized benchmarks will be available at https://github.com/caiyu6666/DDAD-ASR