Cohort profile: St. Michael's Hospital Tuberculosis Database (SMH-TB), a retrospective cohort of electronic health record data and variables extracted using natural language processing.

BACKGROUND: Tuberculosis (TB) is a major cause of death worldwide. TB research draws heavily on clinical cohorts which can be generated using electronic health records (EHR), but granular information extracted from unstructured EHR data is limited. The St. Michael's Hospital TB database (SMH-TB) was established to address gaps in EHR-derived TB clinical cohorts and provide researchers and clinicians with detailed, granular data related to TB management and treatment. METHODS: We collected and validated multiple layers of EHR data from the TB outpatient clinic at St. Michael's Hospital, Toronto, Ontario, Canada to generate the SMH-TB database. SMH-TB contains structured data directly from the EHR, and variables generated using natural language processing (NLP) by extracting relevant information from free-text within clinic, radiology, and other notes. NLP performance was assessed using recall, precision and F1 score averaged across variable labels. We present characteristics of the cohort population using binomial proportions and 95% confidence intervals (CI), with and without adjusting for NLP misclassification errors. RESULTS: SMH-TB currently contains retrospective patient data spanning 2011 to 2018, for a total of 3298 patients (N = 3237 with at least 1 associated dictation). Performance of TB diagnosis and medication NLP rulesets surpasses 93% in recall, precision and F1 metrics, indicating good generalizability. We estimated 20% (95% CI: 18.4-21.2%) were diagnosed with active TB and 46% (95% CI: 43.8-47.2%) were diagnosed with latent TB. After adjusting for potential misclassification, the proportion of patients diagnosed with active and latent TB was 18% (95% CI: 16.8-19.7%) and 40% (95% CI: 37.8-41.6%) respectively. CONCLUSION: SMH-TB is a unique database that includes a breadth of structured data derived from structured and unstructured EHR data by using NLP rulesets. The data are available for a variety of research applications, such as clinical epidemiology, quality improvement and mathematical modeling studies.

Estimated surge in hospital and intensive care admission because of the coronavirus disease 2019 pandemic in the Greater Toronto Area, Canada: a mathematical modelling study.

BACKGROUND: In pandemics, local hospitals need to anticipate a surge in health care needs. We examined the modelled surge because of the coronavirus disease 2019 (COVID-19) pandemic that was used to inform the early hospital-level response against cases as they transpired. METHODS: To estimate hospital-level surge in March and April 2020, we simulated a range of scenarios of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spread in the Greater Toronto Area (GTA), Canada, using the best available data at the time. We applied outputs to hospital-specific data to estimate surge over 6 weeks at 2 hospitals (St. Michael's Hospital and St. Joseph's Health Centre). We examined multiple scenarios, wherein the default (R0 = 2.4) resembled the early trajectory (to Mar. 25, 2020), and compared the default model projections with observed COVID-19 admissions in each hospital from Mar. 25 to May 6, 2020. RESULTS: For the hospitals to remain below non-ICU bed capacity, the default pessimistic scenario required a reduction in non-COVID-19 inpatient care by 38% and 28%, respectively, with St. Michael's Hospital requiring 40 new ICU beds and St. Joseph's Health Centre reducing its ICU beds for non-COVID-19 care by 6%. The absolute difference between default-projected and observed census of inpatients with COVID-19 at each hospital was less than 20 from Mar. 25 to Apr. 11; projected and observed cases diverged widely thereafter. Uncertainty in local epidemiological features was more influential than uncertainty in clinical severity. INTERPRETATION: Scenario-based analyses were reliable in estimating short-term cases, but would require frequent re-analyses. Distribution of the city's surge was expected to vary across hospitals, and community-level strategies were key to mitigating each hospital's surge.