Autonomic dysfunction is common in patients with cancer and may have considerable negative effects on quality of life and mortality. This study retrospectively compared heart rate variability measured by the standard deviation of normal-to-normal intervals (SDNN) to Ewing test score, a composite score from a battery of five defined autonomic tests, in detection of autonomic dysfunction in 47 men with advanced cancer. The Ewing test score has been validated for diagnosis of autonomic dysfunction but is time-consuming and requires considerable patient co-operation; we hypothesised that SDNN, a much simpler test, is a useful alternative. The patients were categorised into three groups according to Ewing score: ≤2 (mild or no autonomic dysfunction), 2.5–3 (moderate) and ≥3.5 (severe). The SDNN (mean ± SD) for the three groups were 57.1 ± 26.9 ms 62.3 ± 22.4 ms and 37.7 ± 20.3 ms respectively. A significant negative correlation was found between Ewing score and SDNN (r = −0.40, P = 0.005). A SDNN of ≤40 ms had 63% sensitivity and 75% specificity in the diagnosis of severe autonomic dysfunction (i.e. Ewing score ≥3.5). The positive predictive value of SDNN ≤40 ms in predicting moderate/severe autonomic dysfunction was 89%.