Objective: The aim of this study was to systematically evaluate prevalence based on variations in case definitions used for epidemiological studies of musculoskeletal disorders (MSDs). Background: Prior studies of MSDs have mostly relied on a single case definition based on questionnaires. Method: In a multicenter prospective cohort study, we systematically collected data to evaluate impacts of differences in case definitions of MSDs on prevalence of three common musculoskeletal disorders: (a) shoulder tendinosis, (b) lateral epicondylalgia, and (c) carpal tunnel syndrome. Production workers were from 21 employment settings in three diverse U.S. states and performed widely varying work. All workers completed laptop-administered structured interviews, two standardized physical examinations, and nerve conduction studies (NCS). Case definitions included symptoms only, and symptoms plus physical examinations and/or NCS. Results: A total of 1,227 subjects had complete health data at baseline. The prevalence for shoulder tendinosis is 23.0% if only glenohumeral pain is used for a case definition, compared with 8.0% if a combination of pain plus a positive supraspinatus test is used. The prevalence for lateral epicondylalgia varied on the basis of lateral elbow pain (12.0%), pain plus tenderness on palpation (9.9%), or pain plus tenderness on palpation plus resisted wrist or middle finger extension (3.5%). Carpal tunnel syndrome prevalence varied on the basis of tingling or numbness in a median nerve–served digit (29.9%) or tingling or numbness plus NCS abnormalities consistent with carpal tunnel syndrome (9.0%). Conclusion: Variations in epidemiological case definitions have major impacts on prevalence of common MSDs. Wide-ranging differences in prevalence may have impacts on purported risk factors that need to be determined.