You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Originally posted by XcrigX October 10, 2023
I am looking at some recent data I generated with Synthea and noticing that a high percentage of Conditions generated seems to be SDOH-related.
These all have Condition.category = encounter-diagnosis.
Primary question:
Should these use a different Condition category (problem-list-item, health-concern, or one of the SDOH profile values) ?
Secondary question:
Is it realistic to have so much of this data? I generated 20k patients, and around 50% of the Conditions generated are not what OMOP/OHDSI consider to be in the SNOMED "condition" domain. I'd agree these factors are important, but it surprised me to see so much of it.
Here is a listing of all the generated condition codes for one sample patient to illustrate the prevalence of these:
224295006 Only received primary school education (finding) **
160903007 Full-time employment (finding) **
1.24171E+14 Chronic intractable migraine without aura
82423001 Chronic pain
196416002 Impacted molars
15777000 Prediabetes
162864005 Body mass index 30+ - obesity (finding)
73595000 Stress (finding) **
160903007 Full-time employment (finding) **
6525002 Dependent drug abuse (disorder) **
55680006 Drug overdose
160903007 Full-time employment (finding) **
73595000 Stress (finding) **
73595000 Stress (finding) **
160903007 Full-time employment (finding) **
314529007 Medication review due (situation)
237602007 Metabolic syndrome X (disorder)
73595000 Stress (finding) **
424393004 Reports of violence in the environment (finding) **
278860009 Chronic low back pain (finding)
1.121E+12 Chronic neck pain (finding)
59621000 Essential hypertension (disorder)
741062008 Not in labor force (finding) **
160903007 Full-time employment (finding) **
314529007 Medication review due (situation)
314529007 Medication review due (situation)
44054006 Diabetes mellitus type 2 (disorder)
55822004 Hyperlipidemia
160904001 Part-time employment (finding) **
73595000 Stress (finding) **
444814009 Viral sinusitis (disorder)
314529007 Medication review due (situation)
302870006 Hypertriglyceridemia (disorder)
160903007 Full-time employment (finding) **
195662009 Acute viral pharyngitis (disorder)
73595000 Stress (finding) **
314529007 Medication review due (situation)
414545008 Ischemic heart disease (disorder)
274531002 Abnormal findings diagnostic imaging heart+coronary circulat (finding)
399261000 History of coronary artery bypass grafting (situation)
```</div>
The text was updated successfully, but these errors were encountered:
Discussed in #1370
Originally posted by XcrigX October 10, 2023
I am looking at some recent data I generated with Synthea and noticing that a high percentage of Conditions generated seems to be SDOH-related.
These all have Condition.category = encounter-diagnosis.
Primary question:
Should these use a different Condition category (problem-list-item, health-concern, or one of the SDOH profile values) ?
Secondary question:
Is it realistic to have so much of this data? I generated 20k patients, and around 50% of the Conditions generated are not what OMOP/OHDSI consider to be in the SNOMED "condition" domain. I'd agree these factors are important, but it surprised me to see so much of it.
Here is a listing of all the generated condition codes for one sample patient to illustrate the prevalence of these:
The text was updated successfully, but these errors were encountered: