Leverage granular, NLP-enriched data on negative symptoms to understand treatment effectiveness in schizophrenia and support development of treatments to target negative symptoms.
Existing assessments developed to measure negative symptoms in people with schizophrenia are too complex to be used routinely in clinical care delivery
Because of this, few structured measures for negative symptoms exist in electronic health records (EHRs)
Much existing information about negative symptoms is contained within free text in the EHR; this information is unstructured and unusable for research
Data within our NeuroBlu Database have been enriched by our proprietary natural language processing (NLP) models. These first-in-kind models, developed specifically for behavioral health, extract information about negative symptoms from clinical notes, which contain valuable unstructured information that is readily usable for analysis in its initial format. Our NLP models transform this information into fit-for-purpose quantitative variables that can be used in a wide array of studies, including those comparing the effectiveness of different treatments.
Although evidence has shown that people who have schizophrenia and exhibit negative symptoms are at higher risk for poor outcomes, there is no consensus about how to efficiently measure negative symptoms.
Few treatments have been developed to target negative symptoms, and easily identifying people with negative symptoms could aid in the discovery of new treatments for this high-risk group.
People with schizophrenia who experience negative symptoms may react differently to treatments than those who experience positive symptoms. The NeuroBlu Database enables analysis at a granular level to support development of individualized treatments.