A study collects blood pressure. Which is the best way to collect the data?
Blood pressure is a quantitative physiological measurement, typically consisting of two continuous numeric values: systolic and diastolic pressure. Therefore, the most appropriate and scientifically valid method of data collection is to use two continuous variables (e.g., systolic = 120 mmHg, diastolic = 80 mmHg).
According to the GCDMP (Chapter: CRF Design and Data Collection), data fields must be designed to capture the most precise, accurate, and analyzable form of clinical data. Numeric data should be collected using numeric data types to allow for range checks, calculations (e.g., mean arterial pressure), and statistical analysis.
Options such as categorical representations (radio buttons or check boxes) introduce rounding, data loss, and analytic limitations. Coding a verbatim diagnosis (option A) is inappropriate for numeric vital sign data and violates the principle of capturing data at the most granular level.
Thus, the correct and validated method per CCDM standards is two continuous variables, ensuring accuracy, traceability, and analytical flexibility.
Reference (CCDM-Verified Sources):
SCDM Good Clinical Data Management Practices (GCDMP), Chapter: CRF Design and Data Collection, Section 4.2 -- Best Practices for Quantitative Data Capture
ICH E6 (R2) Good Clinical Practice, Section 5.5.3 -- Data Accuracy and Collection Standards
FDA Guidance for Industry: Electronic Source Data in Clinical Investigations, Section 4.3 -- Data Format and Structure Requirements
Currently there are no comments in this discussion, be the first to comment!