(152) Use of Large-Scale Electronic Health Record Data to Examine Precise Antihypertensive Medication Exposures Immediately Before a Fall Among US Nursing Home Residents

Publication Number: 826

Background: The relationship between blood pressure (BP) control and falls is complex, with low BP and BP variability increasing risk. BP management with antihypertensive medications may have worsened during the COVID-19 pandemic due to competing care priorities.

Objectives: Among US NH residents, assess whether antihypertensive medication use in the 7 days before a fall differed according to BP control, variability, and COVID-19 diagnosis.

Methods: We leveraged electronic health record data for 280 NHs within a single chain to identify all residents with falls (July 1, 2020-February 12, 2022). We included residents with ≥1 antihypertensive medication and ≥1 systolic BP (SBP) measurement within 7 days before the fall. We categorized BP control according to the average and lowest SBP in the 7 days before the fall as low ( < 120 millimeters of mercury [mmHg]), moderate (120 to < 140 mmHg), or high (≥140 mmHg). Among residents with ≥1 SBP measurement, we also calculated the difference between the highest and lowest SBPs (SBP variability). We compared the number of antihypertensives used by SBP category and quartiles of variability using standardized mean differences (SMDs). Results were stratified according to whether residents had a COVID-19 diagnosis code prior to the date of the fall (after January 1, 2020).

Results: We identified 13,865 NH residents with a fall (median age 78 years, 54% female, 18% with a COVID-19 diagnosis code). Overall, 26% of average SBPs were low, 56% moderate, and 18% high; 67% had ≥1 low SBP. Most residents (60%) received 1 antihypertensive; 33% received 2 and 8.1% received ≥3. The most frequently prescribed antihypertensives were beta-blockers (70% of residents) and angiotensin-related agents (48%). Those with vs. without low average SBP were less likely to be exposed to ≥2 antihypertensives (35% vs. 43%, SMD 0.19). The median SBP variability was 35 mmHg. Those in the top quartile of SBP variability were twice as likely to be exposed to ≥3 antihypertensives than those in the lowest (12% vs. 6%; SMD 0.22). Average SBP, variability, and antihypertensive use were similar between residents with and without a COVID-19 code.

Conclusions: Low BP occurred in two-thirds of NH residents on antihypertensives before a fall, regardless of COVID-19 history. BP variability among NH residents with a fall was also higher than expected. Future research should examine the effects of lowering BP below 120 mmHg and controlling BP variability among NH residents.