Publication guidelines

Truveta publication guidelines

When publishing studies based on data from Truveta, please adhere to the following guidelines for how to document Truveta as a method and data source for your research. We welcome your use of the Truveta logo on your research materials when Truveta data is used and appropriately cited.  


Within the Methods section of your paper, Truveta should be cited as a source like: 

Truveta provided the deidentified medical records used in this study on [INSERT DATE OF ANALYSIS OR DATA EXTRACT]. Truveta’s deidentification is attested to by a qualified expert in accordance with the HIPAA Privacy Rule 

You should adhere to any policies of your organization regarding IRB review. In accordance with §46.101 Protection of Human Subjects, our Truveta Research team cites: 

Because this study used only deidentified medical records, it did not require Institutional Review Board approval. 


Within the Limitations section of your paper or a supplement to the paper, all relevant limitations of Truveta’s data should be cited like: 

Data representativeness 

Note: No healthcare organization names should be cited unless a healthcare organization is studying themself.  

Truveta’s data comes from community healthcare organizations within the United States. No data is included from immigration facilities, correctional facilities, military facilities, or Indian health facilities. 

The patient cohort used in this study is similar to the US population in the following ways: 



Data Completeness 

Truveta’s deidentification process excludes patients in groups with a high risk of reidentification (e.g., patients with rare diseases, rare ethnic groups). 

Medical information not recorded by the clinician because they perceive that information as not relevant for billing or care will not be included within a patient’s medical record within Truveta. 

All medication requests are included within the patient’s medical records within Truveta. Not all medication fills are reconciled to the patient’s medical records within Truveta.  

Any at-home tests are unlikely to be reconciled to the patient’s medical records within Truveta.  

Vaccinations received outside of member health care systems may not be reconciled to the patient’s medical records within Truveta.  

Data Validity 

Medical records within Truveta reflect the determination of the clinician which may be inaccurate (e.g., an incorrect diagnosis). 

Unless otherwise specified, Truveta’s medical records have been normalized using an expert-led, AI-driven process described here, not a per-record human chart review. 

Data Timeliness 

There can be delays in clinician documentation (e.g., it may take several days following a discharge for a provider to record a discharge event). 

There can be delays in clinician recording new event types (e.g., it is unlikely clinicians start documenting the ICD-10 code for Long COVID the day the code is created). 

While in-hospital death is recorded daily, out of hospital death is derived from a weekly Social Security database update.