The increasing adoption of Electronic Health Records (EHRs) in healthcare aims to enhance patient care and operational efficiency. However, the sensitivity of medical data necessitates stringent privacy measures, especially when sharing records across facilities. This study addresses the secure aggregation of EHR data from multiple sites while preserving privacy. We propose a novel approach leveraging a Chosen Ciphertext Attack secure (CCA-secure) variant of the El-Gamal encryption scheme with homomorphic properties and threshold cryptography, complemented by Zero-Knowledge Proofs (ZKPs). Our method facilitates secure and private data aggregation, enabling encrypted data computation and key distribution among multiple parties. The study delves into the theoretical foundations of these cryptographic techniques and explores their practical application in e-health systems. We focus on horizontally partitioned databases, where datasets are distributed across multiple locations, each holding a subset of the entire dataset. This partitioning presents unique challenges for data aggregation and privacy preservation. Additionally, we conduct a comprehensive analysis of our proposed approach, comparing it with existing methods in terms of security, efficiency, and scalability. The findings demonstrate that our approach not only provides robust defenses against various attack vectors but also improves overall scalability and efficiency.