Gait-related back movements require coordination of multiple extremities including the flexible trunk. Ageing and chronic back pain influence these adjustments. These complex coordinations can advantageously be quantified by information theoretically based communication measures such as the gait information flow (GIF). Nine back pain patients (aged 61 ± 10yr) and 12 controls (aged 38 ± 10yr) were investigated during normal walking across a distance of 300 m. The back movements were measured as distances between characteristic points (cervical spine CS, thoracic spine TS, lumbar spine LS) by the sonoSens^® Monitor, a system for mobile motion analysis. Gait information flow and regularity indices (RI₁: short prediction horizon of 100 ms, RI₂: longer prediction horizon of walking period) were assessed as communication characteristics. All indices were non-parametrically tested for group differences. Sensitivity and specificity were assessed by bivariate logistic regression models. We found regularity indices systematically dependent on measurement points, information flow horizon and groups. In the patients RI₁ was increased, but RI₂ was decreased in comparison to the control group. These results quantitatively characterize the altered complex communication in the patients. We conclude that ageing and/or chronic back pain related dysfunctions of gait can advantageously be monitored by gait information flow characteristics of back movements measured as distances between characteristics points at the back surface.