What is data masking?

Data masking, also known as data obfuscation or data anonymization, is a data security technique used to protect sensitive or confidential information by concealing it through encryption, character shuffling, or tokenization.

Data masking plays a role in securing various types of sensitive data, including Personally Identifiable Information (PII), Protected Health Information (PHI), Payment Card Information (PCI), and intellectual property, making it a beneficial tool for organizations seeking to maintain data privacy and meet regulatory requirements.

What is data masking?

The primary purpose of data masking is to safeguard data while preserving its original format and structure. This is primarily during non-production activities, such as software development, testing, or sharing data with third parties, where exposing data could pose significant security and compliance risks. Data masking ensures that even if unauthorized individuals gain access to the masked data, they cannot decipher or reverse-engineer the original information.

What are the main techniques used in data masking?

1. Tokenization: Tokenization replaces sensitive data elements, such as patient IDs or medical record numbers, with non-sensitive placeholder tokens. These tokens have no intrinsic value and cannot be reverse-engineered to reveal the original data. Tokenization helps maintain data relationships and can be reversible when necessary.
2. Pseudonymization: Pseudonymization involves replacing sensitive data with fictional or pseudonymous information. This technique retains the structure and format of the data but makes it challenging to identify individuals. Pseudonymous data can be re-identified using a separate process if needed.
3. Data encryption: Data encryption involves transforming sensitive data into an unreadable format using encryption algorithms. Only authorized users with the decryption key can access and decipher the original information. This is often used in healthcare to secure electronic health records (EHRs) and sensitive patient data.
4. Data redaction: Data redaction is the process of selectively removing or obscuring sensitive information from documents or records, typically in unstructured formats like PDFs, Word documents, or images. Redacted information is replaced with placeholders, black bars, or other forms of obfuscation.
5. Data perturbation: Data perturbation involves introducing slight changes or noise to the original data while retaining its statistical properties. This technique is often used for medical research where the original data must be preserved, but individual identities or specific values must be protected.
6. Data shuffling: Data shuffling involves reordering data records or rows to maintain relationships between data points, but the specific values are randomized. This can help in anonymizing patient data while maintaining the overall structure.
7. Dynamic Data Masking (DDM): Dynamic data masking is a real-time technique used in production environments. It ensures that sensitive patient data is masked or obscured as it is accessed by unauthorized users. DDM is commonly used in EHR systems to control who can view sensitive patient information.

Data masking and HIPAA compliance

Data masking plays a role in ensuring compliance with HIPAA. Like other methods of securing PHI, such as HIPAA compliant email, access controls, and staff training, data masking is helpful for healthcare providers. Data masking techniques, such as pseudonymization, tokenization, and encryption, allow healthcare organizations to share, store, and utilize EHRs and other sensitive patient data while mitigating the risk of unauthorized access or data breaches. 

By masking PHI, organizations can provide access to necessary medical information for research, testing, and training purposes without exposing patients' identifiable details. This helps healthcare entities maintain HIPAA compliance, significantly reducing the potential for breaches and ensuring that patient privacy is upheld. 

See also: Best practices to de-identify PHI

How to implement data masking policies

1. Define data masking rules: Develop data masking rules and policies based on your organization's specific needs. Determine which data elements need to be masked and the masking techniques to be used.
2. Select data masking tools: Choose suitable data masking tools or solutions that align with your organization's requirements. Consider factors such as scalability, performance, and integration capabilities.
3. Data mapping and discovery: Use data mapping and discovery tools to locate sensitive data within your organization's databases, file systems, and other data repositories.
4. Create a masking plan: Develop a comprehensive data masking plan that outlines the specific steps, timeline, and responsibilities for implementing data masking within your organization.
5. Establish test environments: Set up test environments that mirror your production systems. These will be used for masking data and testing the effectiveness of the masking rules.
6. Mask data in test environments: Implement data masking in your test environments according to the established rules. This allows you to validate the masking process and ensure the data remains functional for testing and development purposes.
7. Conduct data masking testing: Thoroughly test the masked data in your test environments to verify that the data retains its utility while sensitive information is effectively obfuscated. Test various scenarios and use cases.

See also: How to de-identify protected health information for privacy