ABSTRACT: BACKGROUND: Variations in corticosteroid/anesthetic doses for injecting shoulder conditions were examined among orthopaedic surgeons, rheumatologists, and primary-care sports medicine (PCSMs) and physical medicine and rehabilitation (PMRs) physicians to provide data needed for documenting inter-group differences for establishing uniform injection guidelines. METHODS: 264 surveys, sent to these physicians in our tri-state area of the western United States, addressed corticosteroid/anesthetic doses and types used for subacromial impingement, degenerative glenohumeral and acromioclavicular arthritis, biceps tendinitis, and peri-scapular trigger points. They were asked about preferences regarding: 1) fluorinated vs. non-fluorinated corticosteroids, 2) acetate vs. phosphate types, 3) patient age, and 4) adjustments for special considerations including young athletes and diabetics. RESULTS: 169 (64% response rate, RR) surveys were returned: 105/163 orthopaedic surgeons (64%RR), 44/77 PCSMs/PMRs (57%RR), 20/24 rheumatologists (83%RR). Although corticosteroid doses do not differ significantly between specialties (p > 0.3), anesthetic volumes show broad variations, with surgeons using larger volumes. Although 29% of PCSMs/PMRs, 44% rheumatologists, and 41% surgeons exceed "recommended" doses for the acromioclavicular joint, >98% were within recommendations for the subacromial bursa and glenohumeral joint. Depo-Medrol(R) (methylprednisolone acetate) and Kenalog(R) (triamcinolone acetonide) are most commonly used. More rheumatologists (80%) were aware that there are acetate and phosphate types of corticosteroids as compared to PCSMs/PMRs (76%) and orthopaedists (60%). However, relatively fewer rheumatologists (25%) than PCSMs/PMRs (32%) or orthopaedists (32%) knew that phosphate types are more soluble. Fluorinated corticosteroids, which can be deleterious to soft tissues, were used with these frequencies for the biceps sheath: 17% rheumatologists, 8% PCSMs/PMRs, 37% orthopaedists. Nearly 85% use the same non-fluorinated corticosteroid for all injections; <10% make adjustments for diabetic patients. CONCLUSION: Variations between specialists in anesthetic doses suggest that surgeons (who use significantly larger volumes) emphasize determining the percentage of pain attributable to the injected region. Alternatively, this might reflect a more profound knowledge that non-surgeons specialists have of the potentially adverse cardiovascular effects of these agents. Variations between these specialists in corticosteroid/anesthetic doses and/or types, and their use in some special situations (e.g., diabetics), bespeak the need for additional investigations aimed at establishing uniform injection guidelines, and for identifying knowledge deficiencies that warrant advanced education.