Over the past few decades, hand-based multimodal biometrics systems have achieved significant attention because of their high security, accuracy, and anti-counterfeiting. Various hand physiological biometric modalities have been explored for identity authentication, i.e., fingerprint, finger knuckle print, palmprint, palm vein, and dorsal hand vein traits. This study provides a comprehensive review focusing on the interface of different hand biometric traits and presents an overview of hand-based multimodal biometrics methods. The framework of this paper is divided into three main categories. Firstly, we introduce the characteristics of four levels of hand-based biometrics in detail. Following this, several typical image capturing devices and image preprocessing techniques of various hand-based biometrics are reviewed. Moreover, existing publicly available and widely used hand-based multimodal biometrics databases are then summarized. Subsequently, the hand-based multimodal biometrics methods are categorized into sensor-level fusion, feature-level fusion, score-level fusion, rank-level fusion, and decision-level fusion. Additionally, the recent hybrid fusion-based and deep learning-based hand multimodal biometrics approaches are analyzed and discussed. Furthermore, we conduct a performance analysis of the abovementioned algorithms from the recent literature. At last, challenges, trends, and some recommendations related to hand-based multimodal biometrics are drawn to give some research directions.