Table 1 Correspondence between structured and unstructured codes.

From: Predictive structured–unstructured interactions in EHR models: A case study of suicide prediction

Concept

Struct.

Unstruct.

Both

Total

Impulse control disorder

145 (19%)

688 (86%)

37 (5%)

796

Unspecified bipolar disorder

1,322 (30%)

4,053 (94%)

1,051 (24%)

4,324

Schizo-affective disorder

250 (42%)

522 (88%)

177 (30%)

595

Opioid dependence or abuse

1,183 (27%)

3,893 (90%)

761 (17%)

4,315

  1. The number of patients that have a structured EHR code for a given concept (first column), an NLP code (based on a free-text mention of that concept in their unstructured clinician notes, second column), and both a structured code and an NLP code for the given concept. Since NLP concepts are more general, each row includes one NLP code but several structured codes with similar descriptions. Furthermore, “opioid dependence” and “opioid abuse” codes were merged into one code since many EHR codes mention both opioid dependence and abuse.
Back to article page