OBJECTIVES: To describe and compare symptom burden, extracted using Natural Language Processing (NLP) techniques from large-scale real-world data, and healthcare resource utilization among patients diagnosed with Major Depressive Disorder (MDD) with and without comorbid substance use disorder (SUD).

METHODS: A retrospective cohort of adults (≥18 years old) with ≥2 records of ICD-9/10 MDD diagnoses and no lifetime diagnosis of schizophrenia were analyzed using NeuroBlu, a real-world data (RWD) platform comprises US electronic health records data from psychiatric specialty clinics. Two sub-cohorts were developed based on the presence of ≥1 ICD-9/10 SUD diagnoses within 3 months before the date of the first recorded MDD diagnosis (index date). NLP techniques were used to identify psychiatric symptoms from semi-structured Mental Status Examination recorded within 30 days from the index date. Post-index psychiatric hospitalizations within 1 year were analyzed.

RESULTS: Out of 81,518 MDD patients, 18,218 (22.3%) had comorbid SUD (M+S) while 50,426 (61.9%) had no diagnosis of SUD in their lifetime (M-S). Baseline symptoms related to mood, judgement, affect, attitude, memory, and suicidality had higher prevalence in the M+S sub-cohort than M-S. Notable differences include judgement issues (57.6% vs 33.2%; p<.001), suicidal ideation (22.7% vs 18.9%; p<.001) and memory issues/impaired memory (24.4% vs 10.0%; p<.001). No significant difference in insight-related issues was observed (77.2% vs 77.7%; p=0.17). Furthermore, post-index psychiatric hospitalizations were more prevalent in the M+S group (27.3% vs 12.1%; p<.001).

CONCLUSIONS: Analysis of NLP-derived symptom labels revealed that adult MDD patients with comorbid SUD in the US had a higher symptom burden and were hospitalized more frequently, leading to a higher medical cost. NLP-derived data provide an opportunity to derive insights from unstructured clinical notes which are otherwise difficult to summarize in large-scale real-world data, leading to a more holistic view about the overall disease burden and economic impact.