(1031) What Predicts Longer Time to Clinical Recognition and Documentation of Hypertension for Patients with High Blood Pressure

Objectives: The purpose of this study was to evaluate whether and how high blood pressure measurements observed in routine clinical settings were addressed in real-world practice. We investigated factors predicting whether high blood pressure was documented by clinicians in clinical notes and/or through diagnoses associated with a patient visit. This study focused particularly on variation across hospital departments.

Methods: This was a retrospective cohort study using data from electronic medical records of a large academic health system in the Midwest. Patients with any high blood pressure measurement (systolic >= 140, diastolic >= 90) from 01-January-2017 through 31-December-2021 were identified for the study sample. Patients less than 18 years old or with any diagnosis indicating hypertension during the 365-day period prior to their first high BP during this period were excluded. Study data were created from EMR data for all inpatient and outpatient encounters from 2016 through 2022. The period of observation for each patient was from 365 days prior to index high blood pressure measurement until first date of any hypertension diagnosis, or the end of the study period (31-December-2022). The unit of analysis was the patient-visit-day. The two primary outcomes of interest were documentation of high blood pressure in clinical notes and diagnosis of hypertension. Discussion of high BP or hypertension in notes was measured using tokenization and approximate phrase matching, with a rule-based algorithm considering relative position of terms to determine relevance and exclusion. Additional measures included patient demographics, BMI, vitals, whether the visit was for an acute problem (injury, illness, pain), number of prior elevated BP observations, and visit type/specialty. Linear mixed-effects regression models with varying intercepts for visit department specialty were used to evaluate differences in documenting high BP across departments adjusting for patient and visit features.

Results: The sample for analyses included 646,248 visit-days with blood pressure for 40,621 unique patients. Over the course of the observation period, 66% of patients had documentation of high BP in notes and 38% were eventually diagnosed with hypertension. Average follow-up for patients from index to hypertension diagnosis or study end was 872 days. Among patients diagnosed with hypertension, average time from index to diagnosis was 66 days (IQR: 0–467 days). Higher BP was associated with higher probability of documentation in notes and diagnosis of hypertension. Visits related to problems of pain were less likely to have high BP documented. Visits occurring before noon or 5pm were also associated with lower probability of documentation. BMI was associated with higher probability of diagnosis, but not documentation in notes. Department specialty explained 18% of variation in probability of documenting high BP in notes and 12% of variation in diagnosis after adjusting for modeled patient and visit factors. Visits occurring in departments of Primary Care or Internal Medicine were associated with a 0.25 higher probability of documenting high BP in notes and visits in the hospital inpatient department had a 0.40 higher probability of documentation in notes. Similarly, visits in Nephrology, Endocrinology, and Urgent Care had a 0.20 greater probability. Conversely, visits in PT/OT/ST had 0.24 lower probability of documentation in notes and visits in Audiology had a 0.33 lower probability of documentation in notes.

Conclusions: There appears to be a long period of time between the first time a high blood pressure is observed in a patient and the eventual clinical recognition or diagnosis of hypertension for the patient. Much of this variation seems to be associated with institutional factors and modifiable systems of communication across providers and departments. There is tremendous opportunity for earlier recognition, diagnosis, and control of hypertension that could lead to less disease burden and improved patient outcomes.