Figure. The likelihood of new blood pressure (BP) medication prescription by stage of BP elevation, number of current BP medications, and visit reason. Each bar represents the mean likelihood of new BP medication prescription, stratified by stage 1 (systolic BP ≥140 but ≤159 mm Hg or diastolic BP ≥90 but ≤99 mm Hg) (A) and stage 2 (systolic BP ≥160 mm Hg or diastolic BP ≥100 mm Hg) BP elevation, number of current BP medications, and whether the patient identified BP as a visit reason. Error bars represent 95% confidence intervals.
Khanna RR, Victor RG, Bibbins-Domingo K, Shapiro MF, Pletcher MJ. Missed opportunities for treatment of uncontrolled hypertension at physician office visits in the United States, 2005 Through 2009. Arch Intern Med. Published online August 6, 2012. doi:10.1001/archinternmed.2012.2749
eTable 1. Adults with Hypertension Presenting with High Blood Pressure: Visit Characteristics for NAMCS 2005-2009
eTable 2. Likelihood of New Blood Pressure Medication Prescription by Visit Characteristics
eFigure. Likelihood of New Blood Pressure Medication Prescribing, by Blood Pressure
Khanna RR, Victor RG, Bibbins-Domingo K, Shapiro MF, Pletcher MJ. Missed Opportunities for Treatment of Uncontrolled Hypertension at Physician Office Visits in the United States, 2005 Through 2009. Arch Intern Med. 2012;172(17):1344-1345. doi:10.1001/archinternmed.2012.2749
Author Affiliations: Divisions of Hospital Medicine (Dr Khanna) and General Internal Medicine (Dr Pletcher), and Department of Epidemiology and Biostatistics (Dr Pletcher), University of California, San Francisco; The Heart Institute and Hypertension Center of Excellence, Cedars-Sinai Medical Center, Los Angeles, California (Dr Victor); Division of General Internal Medicine, San Francisco General Hospital, San Francisco (Dr Bibbins-Domingo); and Departments of Medicine and Health Services, University of California, Los Angeles (Dr Shapiro).
Hypertension affects nearly one-third of all US adults and is a leading treatable cause of cardiovascular death.1,2 National guidelines recommend prescribing new blood pressure (BP) medication for all untreated patients with an established diagnosis of hypertension and for treated patients with stage 2 BP elevation.2 It is unclear how often these guidelines are followed nationally. We analyzed new BP medication prescriptions for patients with uncontrolled hypertension using the National Ambulatory Medical Care Survey (NAMCS).
NAMCS is an annual, nationally representative survey of all visits to nonfederal employed office-based physicians who are primarily engaged in direct patient care.3 Using NAMCS data from 2005 through 2009, we identified visits by nonpregnant adults with an established hypertension diagnosis and elevated office BP of 140/90 mm Hg or greater, seen by physicians who usually manage BP, including general and family practitioners, internists, or cardiologists. NAMCS records at most 8 medications; we excluded patients taking 8 “continued” medications because additional “new” (ie, ninth) medications could not be assessed at these visits.
We measured new BP medication prescription (presence of a BP medication marked as “new” vs “continued” on the medication list) in our sample. We then conducted multivariate regression to determine the independent association between new BP medication prescription and age, sex, race/ethnicity, payer, systolic and diastolic BP, patient-described visit reason, number of “continued” BP medications, comorbidities, physician specialty and degree, primary and established care status, and survey year. Finally, we selected the strongest predictors from our regression model and plotted their association with new BP medication prescription. All analyses were survey weighted and conducted using Stata 11 statistical software (StataCorp).
NAMCS data from 2005 through 2009 included 148 857 observations, representing 4.85 billion visits. Of these, 16 473 (representing 662 million visits) were made by nonpregnant adults with diagnosed hypertension and a measured office BP, who were currently using less than 8 medications and seeing physicians who usually manage BP. Of these, 7153 (representing 261 million visits; 41.7% of the weighted sample) had an elevated BP. Among these visits, 19.5% were prescribed a new BP medication.
In multivariable regression (eTable 1, eTable 2, and eFigure), the strongest predictors of new medication prescription were degree of systolic (odds ratio [OR], 1.24 [95% CI, 1.17-1.31] per 10 mm Hg) and diastolic (OR, 1.23 [95% CI, 1.11-1.36] per 10 mm Hg) BP elevation; whether the patient stated BP to be a visit reason (OR, 2.57 [95% CI, 2.14-3.09]); and the number of current BP medications (OR, 0.45 [95% CI, 0.37-0.56] for 1 medication [vs 0]; OR, 0.23 [95% CI, 0.18-0.28] for ≥2 medications [vs 0]). After dichotomizing BP elevation to stage 1 (systolic BP ≥140 but ≤159 mm Hg or diastolic BP ≥90 but ≤99 mm Hg) or stage 2 (systolic BP ≥160 mm Hg or diastolic BP ≥100 mm Hg),2 we plotted the likelihood of new BP medication prescription by BP elevation, patient identification of BP as a visit reason, and number of current BP medications (Figure). New BP medication prescription exceeded 50% only among visits in which untreated patients with stage 2 BP elevation identified BP as a visit reason.
New medication prescription did not increase from 2005 through 2009 (P value for trend, .35).
Among patients with diagnosed hypertension who were seeing physicians who manage BP, nearly 60% had well-controlled hypertension. Among those with uncontrolled hypertension, however, new BP medication was prescribed at just 19.5% of visits—with no improvement from 2005 through 2009. New prescriptions were substantially more common when patients identified hypertension as a visit reason, but remained under 50%, except among previously untreated patients with stage 2 BP elevation. Our findings are consistent with studies in more specific clinical settings4,5 and demonstrate that the low likelihood of new BP medication prescription for uncontrolled hypertension is a national problem.
We also found that patient self-identification of BP as a visit reason more than doubled the odds of new medication prescription. Empowering patients to discuss BP with their physicians through previsit forms, patient portals, education, and incentives may be a way to achieve better BP control, as shown in other studies.6,7
Our analysis has several limitations. Most importantly, NAMCS does not record medication dose; thus, our study speaks only to new medication prescription, not to overall intensification. If NAMCS physicians escalated medication dose without adding a new medication in the same ratio to overall intensification as in other studies (1:3),5 the likelihood of intensification overall in our sample would have been 26%. However, new medication prescription was often the only option (eg, among untreated patients) or the best option (eg, low-dose combination therapy for stage 2 BP elevation, which tends to lower BP more effectively, with less adverse effects, than dose escalation8,9). In addition, NAMCS does not record the reasons new medications are or are not prescribed; many (eg, white coat hypertension, patient preference, nonadherence, regimen complexity, and acute competing concerns) might be clinically appropriate.
Despite these limitations, our analysis suggests that missed opportunities for new BP medication prescription are common in the United States. Taking advantage of these opportunities could result in improved BP control among US patients.
Correspondence: Dr Khanna, Division of Hospital Medicine, University of California, San Francisco, 533 Parnassus Ave, PO Box 0131, Office U136, San Francisco, CA 94143 (firstname.lastname@example.org).
Published Online: August 6, 2012. doi:10.1001 /archinternmed.2012.2749
Author Contributions:Study concept and design: Khanna, Victor, Shapiro, and Pletcher. Acquisition of data: Khanna. Analysis and interpretation of data: Khanna, Victor, Bibbins-Domingo, Shapiro, and Pletcher. Drafting of the manuscript: Khanna, Victor, Bibbins-Domingo, and Pletcher. Critical revision of the manuscript for important intellectual content: Khanna, Victor, Bibbins-Domingo, Shapiro, and Pletcher. Statistical analysis: Khanna. Obtained funding: Victor and Pletcher. Administrative, technical, and material support: Khanna and Victor. Study supervision: Victor, Shapiro, and Pletcher.
Financial Disclosure: None reported.
Funding/Support: This study was supported by grant R01 HL080582-04 from the National Heart, Lung, and Blood Institute (principal investigator: Dr Victor) and a grant “Barbershop Model of Community-Based Care in Los Angeles, and Clinical Research Projects on the Disparities in Cardiovascular Care in Minorities” from the Lincy Foundation (principal investigator: Dr Victor).
Additional Contributions: Farzaneh Pour Ansari, MS, provided orientation to the NAMCS data set; Premere Knowles, MS, provided administrative support; Eric Vittinghoff, PhD, helped in formatting the Figure; Pamela G. Coxson, PhD, suggested additional analyses; and Larissa Thomas, MD, MPH, provided revisions to the manuscript.