Metabolic dysfunction has been suggested to be involved in musculoskeletal pain; however, few studies have identified metabolic markers associated with multisite musculoskeletal pain (MSMP). This study sought to identify metabolic marker(s) for MSMP by metabolomic analysis. The Tasmanian Older Adult Cohort Study (TASOAC) provided the discovery cohort with the Newfoundland Osteoarthritis Study (NFOAS) providing the replication cohort. MSMP was assessed by a self-reported pain questionnaire and defined as painful sites ≥4 in both TASOAC and NFOAS. Further, MSMP was also defined as painful sites ≥7, while non-MSMP was defined as either painful sites <7 or ≤1 in NFOAS. Serum samples of TASOAC received metabolic profiling using The Metabolomics Innovation Centre (TMIC) Prime Metabolomics Profiling Assay. The data on the identified metabolites were retrieved from NFOAS metabolomic database for the purpose of replication. A total of 409 participants were included in TASOAC, 38% of them had MSMP. Among the 143 metabolites assessed, 129 passed quality control and were included in the analysis. Sphingomyelin (SM) C18:1 was significantly associated with MSMP (OR per log µM increase=3.96, 95%CI, 1.95-8.22; p=0.0002). The significance remained in multivariable analysis (OR per log µM increase=2.70, 95%CI, 1.25-5.95). A total of 610 participants were included in NFOAS and the association with SM C18:1 was successfully replicated with three MSMP definitions (OR ranging from 1.89 to 2.82; all p<0.03). Our findings suggest that sphingomyelin metabolism is involved in the pathogenesis of MSMP and circulating level of SM C18:1 could serve as a potential marker in the management of MSMP.