A multicenter clinical trial comparing nifedipine gastrointestinal therapeutic system (GITS) to atenolol was conducted to evaluate the effects of reporter (physician, patient, and spouse) and assessment of environment (clinic vs. take-home), and structure (long vs. short forms). After a 4-week single-blind, placebo washout, 394 male patients with mild-to-moderate hypertension were randomized to receive 20 weeks of either atenolol or nifedipine GITS, titrated to achieve goal blood pressure of < 90 mm hg for 8 weeks and maintained at the final dose for 12 weeks. Quality-of-life assessments included "long-form" clinical assessments by the patient, parallel "short-form," take-home, mail-back assessments by patient and spouse and a side effects and symptom distress evaluation by the physician. Patients completing 20 weeks of therapy indicated a more favorable quality-of-life profile (p < 0.05) for nifedipine GITS compared with atenolol for psychosocial, positive well-being, emotional ties, behavior/emotional control, vitality, and leisure scales using the long-form clinical assessment. The short-form, take-home, mail-back assessments were highly correlated with the clinic assessments, but their responsiveness to treatment differences was less owing to item structure and lower return rates. Spouse reports were more sensitive to treatment differences in sexual functioning with a more favorable response for patients taking nifedipine GITS (p < 0.02). Physicians were less sensitive to the overall impact of side effects, reporting only 14.9% of symptoms reported by the patient. Physician evaluation of symptoms and side effects does not provide the sensitivity required for detecting quality-of-life treatment differences.(ABSTRACT TRUNCATED AT 250 WORDS)