Atrial fibrillation (AF) is the most common cardiac arrhythmia managed in clinical practice. Maintenance of intended rate or rhythm control following hospitalisation is a key therapeutic goal.The purpose of this study was to assess post-discharge maintenance of intended AF control and classify potentially predictive heart rate (HR) phenotypes via electrocardiogram (ECG) Holter monitoring.In a sub-study of a multicentre randomised controlled trial comparing AF-specific management with usual care, 24-hour ECG Holter monitoring was undertaken in 133 patients 7-14 days post-discharge. Intended rate and rhythm control were compared to Holter data. Analysis of the frequency distribution of mean hour-to-hour differences identified those with labile HRs.Mean age was 71 ± 10 years, 67 (50%) were male and mean HR was 72 ± 14 bpm. Most (89%) had persistent AF (median time in AF=39% (IQR 0-100%)). Uncontrolled HR (>90 bpm for >10% of recording) occurred in 35 (26%) patients and 49 (37%) patients did not achieve their intended rate (n=26) or rhythm control (n=23). Patients in the upper quartile of mean hour-to-hour HR variability were identified as persistently labile (n=33). A further group (n=22) with periodically labile HRs was identified. Those with coronary artery disease (OR 0.34; 95% CI 0.13-0.91, p=0.033) or renal disease/dysfunction (OR 0.24; 95% CI 0.06-0.98, p=0.047) were less likely to demonstrate HR stability (n=78).Post-discharge ECG Holter monitoring of AF patients represents a valuable tool to identify deviations in intended rhythm/rate control and adjust therapeutic management accordingly. It may also identify individuals who demonstrate labile HRs.