Forty-eight patients with chronic neck pain were randomized to receive either conventional treatment alone (n=16), or conventional treatment with either deep cervical flexor training (n=16) or stabilization of the neck and core (n=16). The primary outcome of interest was pain intensity measured on a Verbal Numeric Pain Scale (VNPS). Secondary outcome measures were Reedco's posture score, cervical range of motion, and Neck Disability Index (NDI), all after 4 weeks. Change from baseline scores was significantly better for pain, posture, ROM in flexion, extension, right lateral flexion, left lateral flexion, right rotation, and left rotation between conventional treatment and deep cervical flexor training and conventional treatment against stabilization of the neck and core region. No significant between-group differences were seen in the change from baseline scores between deep cervical flexor training and stabilization of the neck and core region.