What is the difference between coded fields and free-text fields in registry data and how are free-text fields handled?

What is the difference between coded fields and free-text fields in registry data and how are free-text fields handled?

• A. Free-text fields are unstructured and must be coded or abstracted into structured fields.
• B. Coded fields use predefined values; free-text fields are unstructured and must be coded or abstracted into structured fields.
• C. All fields are stored as free text.
• D. Free-text fields do not require any coding.

Answer: (B)

In registry data, there are two kinds of fields: coded (structured) fields and free-text (unstructured) fields. Coded fields use predefined values, such as standardized codes for diagnosis site, histology, stage, and other variables. This structure makes data easy to compare, analyze, and aggregate across cases and time. Free-text fields capture detailed narrative information from reports, notes, and descriptions. They are rich and clinically meaningful but not readily analysable until they are converted into structured data. Therefore, free-text must be coded or abstracted into structured fields using standard coding schemes (for example, ICD-O site and morphology, AJCC staging, or registry-specific abstractions). This coding can be done by trained abstractors, sometimes aided by computer-assisted coding or natural language processing, with quality checks to ensure consistency. So the best description is that free-text fields are unstructured and must be coded or abstracted into structured fields, while coded fields already use predefined values for efficient analysis. The other statements are incorrect because they imply all data are free text or that free-text requires no coding.

In registry data, there are two kinds of fields: coded (structured) fields and free-text (unstructured) fields. Coded fields use predefined values, such as standardized codes for diagnosis site, histology, stage, and other variables. This structure makes data easy to compare, analyze, and aggregate across cases and time.

Free-text fields capture detailed narrative information from reports, notes, and descriptions. They are rich and clinically meaningful but not readily analysable until they are converted into structured data. Therefore, free-text must be coded or abstracted into structured fields using standard coding schemes (for example, ICD-O site and morphology, AJCC staging, or registry-specific abstractions). This coding can be done by trained abstractors, sometimes aided by computer-assisted coding or natural language processing, with quality checks to ensure consistency.

So the best description is that free-text fields are unstructured and must be coded or abstracted into structured fields, while coded fields already use predefined values for efficient analysis. The other statements are incorrect because they imply all data are free text or that free-text requires no coding.