Nyman, M.

Background: The diagnosis of hypertension requires systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg on at least 2 different occasions. Therefore, there is the possibility of patients with elevated BP remaining undiagnosed if not seen by the same provider. Purpose: To utilize electronic medical records (EMR) to identify patients not previously diagnosed with hypertension that have elevated blood pressure meeting criteria for a diagnosis of hypertension using a threshold of SBP ≥ 140mmHg or DBP ≥ 90 mmHg (high BP) on 2 or more occasions. Methods: This was a cross sectional design utilizing retrospective multi-year billing and clinical data from a large multi-specialty center in the Midwest. Using electronic records of all outpatient visits in each year from 2009 - 2011, patients with at least 2 visits with SBP ≥ 140mmHg or DBP ≥ 90 mmHg (2HBP) in the measurement year were identified. These patients were compared with previously identified cohorts of known hypertension patients (Known HTN) compiled using the EMR problem list. A sensitivity analysis was done using patients with high BP at 2 consecutive visits (2CHBP) and also those with high BP at 3 consecutive visits (3CHBP). We compared proportions of patients with high BP without a prior diagnosis of hypertension (UDHTN). Results: The proportion of patients with SBP ≥ 140mmHg or DBP ≥ 90 mmHg without a prior diagnosis of hypertension (UDHTN) when compared to the cohort of known patients with a diagnosis of hypertension (Known HTN) was 25% in 2009, 26% in 2010, and 28% in 2011 in the 2CHBP cohort compared to 27% in 2009, 28% in 2010, and 30% in 2011 in the 2HBP cohort and 18% in 2009, 19% in 2010 and 23% in 2011 in the 3CHBP cohort. Conclusion: About a quarter of patients meeting current thresholds on multiple and consecutive visits did not have a known diagnosis of hypertension. The use of EMR can identify these patients for commencement of appropriate management. Table 1

Background: Hypertension (HTN) control rate is a core clinical quality measure. It is presently assessed by manual review of a random sample of patients with HTN billing claims. There is an increasing push by regulators towards the use of electronic medical records (EMR) to assess quality of care. Purpose: To use EMR to obtain a more in-depth assessment of quality of hypertension care. Methods: Cross sectional study involving adults with HTN seen at 2 or more office visits in 2009 by eligible primary or specialist providers at a large medical group practice. We measured the proportion of HTN patients with blood pressure (BP) at goal (BP < 140/90 mm Hg) at their last hypertension visit (LHV). Results: Of a total of 10,401 patients, 5970 (57%) were controlled and 1959 (19%) were uncontrolled at their LHV. Control could not be assessed in 2472 (24%) as BP values were unavailable from the vital signs section of the visit notes in the EMR. Of a random sample of 250 patients with unavailable BP, only 94 (37.6%) had BP values documented in other parts of the visit note. In 29 patients (11.6%) billing for HTN was done without adequate evidence that it was addressed at that visit. Control rates were highest for primary care providers where measurement and documentation of BP was a focus - Family Medicine 77% controlled, 1% unavailable; Primary Care Internal Medicine 71% controlled, 4% unavailable. A considerable proportion of patients in the other four provider groups did not have BP available from their last hypertension visit - Cardiology 59% controlled, 22% unavailable; Nephrology 38% controlled, 47% unavailable; Preventive Medicine 48% controlled, 39% unavailable; General Internal Medicine 41% controlled, 45% unavailable. Control rates were similar across provider groups when patients with unavailable BP were excluded: Family Medicine 78%, Primary Care 74%, Cardiology 75%, Nephrology 73%, Preventive Medicine 79%, and General Medicine 75%. Conclusions: Up to a quarter of HTN patients did not have properly documented BP in the EMR at their LHV. For those with measurements recorded, control rates were similar across primary care and specialty clinics, but measurement rates were higher for primary care practices. Use of the EMR facilitates efficient and granular assessment of the measurement process and control rates for HTN among various provider groups.

Introduction: Patients with hypertension (HTN) may remain undiagnosed as the diagnosis requires systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg on ≥ 2 different occasions. Hypothesis: We sought to evaluate outcomes of patients without diagnosed hypertension who have elevated BP meeting current thresholds for hypertension on 2 consecutive episodes. Methods: Together with retrospective billing and clinical data, electronic records of all outpatient visits in 2009 were used to identify 24,277 patients with 2 consecutive visits (2CHBP) with SBP ≥ 140mmHg or DBP ≥ 90 mmHg and were compared with a previously identified cohort of 53,998 known hypertension patients. Among the 2CHBP group, 6,071 (25%) were identified without a prior diagnosis of hypertension and were categorized as undiagnosed hypertension. They were compared with patients with known hypertension who had ≥ 1 visit in 2009. Cross-sectional analysis of cardiovascular comorbidities and proportional hazards for survival were done. Results: Of 5,978 undiagnosed HTN patients, 49% were female compared to 51% of 10,311 known HTN. In the undiagnosed HTN group, 19.5% were ≥ 80 years compared to 30.4% of known HTN patients. Cardiovascular comorbidities were generally lower in the undiagnosed HTN group compared to the known HTN group - heart failure (7% v 8%; OR 0.88, 95% CI: 0.77 - 0.99; p < 0.038), atrial fibrillation (11% v 12%, OR 0.88 95% CI 0.79 - 0.97) and coronary artery disease (17% v 26%, OR 0.57 95% CI 0.52 - 0.61). However after a four year followup, mortality was higher in the undiagnosed HTN group with unadjusted hazard ratio HR 1.4 (95% CI 1.29 - 1.60, p<0.0001) and HR 2.41 (2.16 -2.68, p<0.0001) when adjusted for age, sex, heart failure and diabetes. Conclusions: Over a quarter of patients meeting current thresholds on 2 consecutive visits did not carry a known diagnosis of hypertension. Although younger and with less cardiovascular comorbidities than those with diagnosed hypertension, mortality after 4 years was significantly higher in those with undiagnosed hypertension. Electronic records can be used to identify these high risk patients.