Shoulder complaints are among the commonest causes of musculoskeletal pain. They are potentially disabling and frequently cause absenteeism from work and claims for sickness benefits. One of the most prevalent subtypes of shoulder pain is impingement. This is often managed physiotherapeutically, with 'hands-on' manual therapy and exercises being mainstays. Objectives: To assess the effectiveness of manual therapy and exercises to improve pain, disability and function in people with shoulder impingement. Methods: A systematic review was conducted including systematic reviews, quasi-randomized trials and randomized controlled trials published up to October 2008. Searches included the Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, Medline, Embase, Cinahl and PEDro. Methodological quality and risk of bias were assessed with appropriate instruments. All findings were critically analysed and discussed, and summary conclusions formulated. Results: Eight systematic reviews and six randomized controlled trials were included. Methodological quality and risk of bias as well as population and treatment parameters varied. Clinical heterogeneity prevented meta-analysis, thus all findings were synthesized narratively. The included research provides limited evidence to support the use of manual therapy and exercise interventions for shoulder impingement. This primarily relates to subacute and chronic conditions and short to medium-term effectiveness. Conclusions: Manual therapy and exercise seem effective for shoulder impingement, but varying methodological quality and risk of bias in reviews and trials warrant caution in the interpretation of the results. There is a need for further good-quality primary research.