Differences

This shows you the differences between two versions of the page.

--- documentation:next_cdm:place_of_service_source_concept_id [2017/03/14 01:54] – gowtham_rao
+++ documentation:next_cdm:place_of_service_source_concept_id [2017/03/14 17:49] (current) – gowtham_rao
@@ Line 4: / Line 4: @@
 **Proposal:** This is a small proposal to add place_of_service_source_concept_id. It is a general OMOP convention to represent X_source_value as X_source_concept_id and X_concept_id, where X_concept_id is always standard-vocabulary for the domain, and X_source_concept_id maybe standard/non-standard vocabulary that is an integer representation of X_source_value. We don't have X_source_concept_id in care_site and this proposal is to add it.
-**Use cases:** Many place of services such as Acquired immune deficiency syndrome clinic, Acute pain clinic, Bone marrow transplant clinic, Hospital-based outpatient emergency care center, Medical Intensive Care unit -- does not always have 1:1 mapping with standard-concept ids. We have use cases where we want to use local place of service vocabulary for analytics e.g. how many persons attended the Acute Pain Clinic on Saturday's vs Mondays. This cannot be addressed without adding the field place_of_service_source_concept_id
+**Use cases:** Many place of services such as Acquired immune deficiency syndrome clinic, Acute pain clinic, Bone marrow transplant clinic, Hospital-based outpatient emergency care center, Medical Intensive Care unit -- does not always have 1:1 mapping with standard-concept ids. We have use cases where we want to use local place of service vocabulary for analytics e.g. how many persons attended the Acute Pain Clinic on Saturday's vs Mondays. This cannot be addressed without adding the field place_of_service_source_concept_id. Also adding source_concept_id allows to have good data lineage and provenance to the source.
 **Discussion:** http://forums.ohdsi.org/t/proposal-add-place-of-service-concept-id-to-care-site-table/2404
 **Backward compatibility:** Yes
@@ Line 22: / Line 21: @@
-Add the following to the descriptions in conventions:
+**Other considerations:**
+"Place_of_service_" is a long descriptor. "Place_of_service_source_concept_id" is very long and is more than 32 - breaks database column naming rules such as 32 character limit. To overcome, we will have to systematically replace "Place_of_service_" to "Place_".
+**What is a place**
+The word 'place of service' is interpreted differently by different people. Which one is more important - 'place' or 'service' in place of service. In this proposal, we interpret place as a **place 'type'** e.g. it is a ward-unit within a location, or an operating room or icu. Source data may have non-standard source code value for it such as medical icu, or surgical icu, or Gold-wing, Red-wing, Blue-wing etc to represent local hospital vocabulary for 'places' within the location. These are "settings" or "operational units" within a location where service is provided. In United States claims, CMS defines "Place of Service Codes" as //"two-digit codes placed on health care professional claims to indicate the **setting** in which a service was provided."// It does not represent the actual service provided. In-fact the type of service and level/intensity of service may be independent of the "place of service" e.g. a patients with medical complication such as sepsis who needs ICU care, may be temporarily managed in surgical ICU, pending availability of beds in medical icu (place of service)
+**Add the following to the descriptions in conventions:**
   * It represents the site information needed for use in visit_occurrence table.