The quality of routinely collected maternity data

Area of Science:

• Health Informatics
• Clinical Data Management
• Maternal Health

Background:

• Regional maternity databases are crucial for clinical information management.
• Assessing the validity of data within these systems is essential for reliable audits.
• The St Mary's Maternity Information System (SMMIS) is a key regional database for maternity care.

Purpose of the Study:

• To evaluate the accuracy and consistency of clinical data within the SMMIS.
• To determine the validity of the SMMIS as a resource for comparative maternity practice audits.

Main Methods:

• Retrospective review of 892 maternity case notes and matched SMMIS records.
• Data validation performed by a midwife trained in clinical coding.
• Analysis of percentage agreement for 17 SMMIS data items and diagnosis codes.

Main Results:

• High agreement (over 80%) between SMMIS records and case notes for most data items.
• Ten of 17 data items demonstrated 95% or greater agreement.
• SMMIS generated fewer and a narrower range of diagnosis codes compared to manual abstraction from notes.

Conclusions:

• The clinical data in the SMMIS is largely accurate and consistently recorded across different units.
• SMMIS is a valuable resource for comparative audits of maternity practice.
• The direct data collection method used by SMMIS could serve as a model for other NHS specialties.