We analyzed a random sample of general medicine clinic patients to determine the natural history of newly treated hypertensive (NH) patients: discontinuation patterns, critical intervention periods, and hypertension's (HBP) utility as an indicator condition. The NH patients exhibited a 48% dropout rate in the first year and better continuation adherence than new nonhypertensive (NNH) patients. Patients with HBP and other chronic diseases had better continuation adherence than those with HBP alone, although no predictive patterns emerged. New patients displayed rapid early discontinuation, with further linear decline by four months for NNH and by eight months for NH patients. All patients showed similar subsequent falloff: linear annual decline at 13% to 36%. We conclude that discontinuation rates are unacceptably high, that interventions must be continued throughout treatment, and that HBP has limited utility as an indicator chronic disease.
(Arch Intern Med 139:545-549, 1979)