OBJECTIVES: Musculoskeletal pain is experienced by more than a third of the adult population. Alleviation of pain is an important aspect of care in these patients. The aim of the present study was to assess the number of NSAID prescriptions prescribed to patients with musculoskeletal pain during 2004-2008 at 4 primary health care centres in Sweden. METHODS: This retrospective longitudinal study was based on primary care electronic medical records review and data for patients with any ICD diagnosis code of musculoskeletal pain and  prescription of any ATC code for NSAID, selective or non-selective and weak opioids (codeine, tramadol, dextropropoxyphene) was extracted and analysed. RESULTS: A total of 23 456 prescriptions were analysed, 18 187 prescribed to adults [18-64 years] and 5 269 to elderly ≥65 years of age [65-101 years]. The proportion of NSAID prescriptions for non-selective plus selective Cox-2-inhibitors (Coxibs), did not change during the 5- year period, 91.1, 91.3, 91.4, 92 and 91.5 % of patients were prescribed an NSAID. The proportion of prescriptions for traditional non-selective NSAIDs increased from 83.9% to 90.3 % from year 2004 to 2008 among adult patients and from 69.7% to 78.7% among elderly. The proportion of prescriptions for Coxibs decreased between 2004 and 2008 and were prescribed to 3.6 % of patients in this study in 2008. Co-prescription of gastric protective therapy was made in 12 - 20 % of prescriptions with no differences between NSAID, Coxibs and non-NSAID. CONCLUSIONS: In this review of NSAID prescription utilization among patients with a diagnosis of musculoskeletal pain shows an extensive use of anti-inflammatory agents without co-prescription of gastro-protective medication. The impact of this high utilization of NSAIDs without co-prescription of gastro-protective agents on the risk for upper gastro-intestinal complications warrants further evaluation.