Abstract The objective of this study was to assess compliance with a 10-day treatment of antibiotics or placebo once-daily (OD) and three-times-daily (TD) for lower respiratory tract infections (LRTIs) using electronic monitoring, and to evaluate whether compliance depends on time since the start of treatment and weekday. Taking compliance, timing compliance, correct dosing compliance and mean interdose intervals were assessed using data from 155 LRTI patients who received either a 10-day treatment of amoxicillin TD and placebo OD or roxithromycin OD and placebo TD using a double-dummy technique. Compliance was assessed by electronic monitoring. Taking compliance was 98.0% for the OD regimen and 91.0% for the TD regimen. Correct dosing was 98.1% for the OD regimen and 91.1% for the TD regimen and timing compliance was 48.2% and 10.9%, respectively. The mean interdose interval before the first daily dose for the TD group was particularly prolonged to >13 h. Correct dosing over time showed fewer patients with correct dosing compliance, reaching a low of 79% for the TD group towards the end of the 10-day treatment. Compliance was not influenced by weekday. This study adds important information to the limited evidence on compliance with antibiotics for LRTI, one of the most common reasons for consultation in primary care. Taking compliance was high for both regimens, yet timing compliance was poor. The prolonged mean interdose intervals provide striking new insights into understanding non-compliance with more-than-once-daily regimens. These findings require consideration when exploring ways to improve future compliance in short-term antibiotic treatment for respiratory tract infections.