To investigate the effect of a neuro-muscular exercise (NEMEX) therapy program compared with instructions in optimized analgesics and anti-inflammatory drug use (PHARMA), on measures of knee-joint load in people with mild to moderate knee osteoarthritis (OA). We hypothesized that knee joint loading during walking would be reduced by NEMEX and potentially increased by PHARMA.Single-blind, randomized controlled trial (RCT) comparing NEMEX therapy twice a week with PHARMA. Participants with mild-to-moderate medial tibiofemoral knee OA were randomly allocated (1:1) to one of two 8-week treatments. Primary outcome was change in knee load during walking (Knee Index, a composite score from all three planes based on 3D movement analysis) after 8 weeks of intervention. Secondary outcomes were frontal plane peak knee adduction moment (KAM), Knee Injury and Osteoarthritis Outcome Scores (KOOS) and functional performance tests.Ninety three participants (57% women, 58 ± 8 years with a body mass index [BMI] of 27 ± 4 kg/m2 (mean ± standard deviation [SD])) were randomized to NEMEX group (n = 47) or PHARMA (n = 46); data from 44 (94%) and 41 (89%) participants respectively, were available at follow-up. 49% of the participants in NEMEX and only 7% in PHARMA demonstrated good compliance. We found no difference in the primary outcome as evaluated by the Knee Index -0.07 [-0.17; 0.04] Nm/%BW HT. Secondary outcomes largely supported this finding.We found no difference in the primary outcome; knee joint load change during walking from a NEMEX program vs information on the recommended use of analgesics and anti-inflammatory drugs. ClinicalTrials.gov Identifier: NCT01638962 (July 3, 2012). Ethical Committee: S-20110153.