Low back pain (LBP) is a leading cause of disability. However, the processes contributing to disability are not well understood. Therefore, this study (1) empirically derived LBP subgroups and (2) validated these subgroups using walking performance, pain, and disability measures. Seventy adults with LBP underwent testing for a priori determined sensory (temporal summation; conditioned pain modulation), psychological (positive affect/coping; negative coping), and motor (trunk extensor muscle activation during forward bending and walking) measures. A hierarchical cluster analysis determined subgroups that were then validated using walking (walking speed; Timed Up and Go [TUG]; TUG-Cognitive [TUG-Cog]; obstacle negotiation) and clinical (Brief Pain Inventory; Oswestry Disability Index; low back pressure pain threshold) measures. Two subgroups were derived: (1) a "Maladaptive" subgroup (n = 21) characterized by low positive affect/coping, high negative coping, low pain modulation, and atypical trunk extensor activation and (2) an "Adaptive" subgroup (n = 49) characterized by high positive affect/coping, low negative coping, high pain modulation, and typical trunk extensor activation. There were subgroup differences on 7 of 12 validation measures. The Maladaptive subgroup had reduced walking performance (slower self-selected walking speed, TUG completion, and obstacle approach and crossing speed) and worse clinical presentation (higher pain intensity, pain interference, and disability) (moderate to large effect sizes; P's < 0.05). Findings support the construct validity of this multidimensional subgrouping approach. Longitudinal studies are needed to determine whether the Maladaptive subgroup is predictive of poor outcomes, such as pain chronicity or persistent disability.