Background: Terminal digit preference in blood pressure (BP) measurement has been reported in both clinical and research settings. This article examines the prevalence of terminal digit preference (TDP) in primary care practices and the effect that a practice's level of TDP influences patients' BP measurements and management.
Methods: Data were obtained in cross-sectional fashion from the electronic medical records of active patients from 85 practices around the United States. The TDP prevalence was calculated, and statistical techniques were used to examine the influence of a practice's TDP on patients' BP measurements and on the likelihood that patients had an active prescription for selected antihypertensive medications.
Results: The TDP was common, with zero being recorded 44.6% and 47.5% of the time for systolic BP and diastolic BP, respectively. Patients belonging to practices with higher TDP levels had significantly (P < .01) lower systolic BP measurements than patients in practices with lower TDP levels. Patients belonging to practices with higher TDP levels also had significantly lower odds (odds ratio [OR] = 0.92, 95% confidence interval [CI] [0.85, 0.99], P = .036) of having an active prescription for an antihypertensive medication, an association that was stronger in women (OR = 0.91, P = .023) than in men (OR = 0.95, P = .21).
Conclusions: The TDP for BP measurements is common. Although TDP effects on patients' BP measurements may appear modest, treatment of patients, especially women, with antihypertensive medication may be systematically affected by this preference.