A random sample of patients taking two or more drugs, at least one of which was digoxin or methyldopa, was drawn from a medical clinic population. After giving informed consent, the patients were randomized into control and experimental groups. The experimental group was seen by a specially prepared nurse interventionist who attempted to improve medication compliance. The levels of the drugs in the blood were taken as indicators of medication compliance. Our results document that patients exposed to nurse intervention were more compliant than the general clinic population, but were not more compliant than a nonintervention control group. Problems encountered in the collection and interpretation of compliance study data were identified and discussed.
(Arch Intern Med 138:36-40, 1978)