The Generalized Data Model for clinical research

Table 2 Example mapping table for conversion to the Sentinel model

concept_id _1	relationship _id	concept_id _2	vocabulary_id_1	concept_code_1	concept_text_1	vocabulary_id_2	concept_code_2	concept_text_2
100000002	is_a	100000092	synpuf_sex	2	Female	sentinel_sex	F	Female
100000001	is_a	100000093	synpuf_sex	1	Male	sentinel_sex	M	Male
100000003	is_a	100000095	synpuf_race	1	White	sentinel_hispanic	N	Not Applicable
100000004	is_a	100000095	synpuf_race	2	Black	sentinel_hispanic	N	Not Applicable
100000005	is_a	100000095	synpuf_race	3	Others	sentinel_hispanic	N	Not Applicable
100000006	is_a	100000097	synpuf_race	5	Hispanic	sentinel_hispanic	Y	Yes
100000005	is_a	100000098	synpuf_race	3	Others	sentinel_race	0	Unknown
100000006	is_a	100000098	synpuf_race	5	Hispanic	sentinel_race	0	Unknown
100000004	is_a	100000101	synpuf_race	2	Black	sentinel_race	3	Black or African American
100000003	is_a	100000103	synpuf_race	1	White	sentinel_race	5	White

Note: This table shows how sex and race/ethnicity would be associated if the GDM were to use the Sentinel definitions as its defaults. Mappings allow users to query the data based on standardized default values for these variables while retaining the values from the original data. The GDM can support more extensive mapping relationships, but these are not required. Note that only the first 3 columns are part of the Mappings table. The remaining columns are created from joining to the Concepts and Vocabularies tables

ISSN: 1472-6947