BACKGROUND: Previous epidemiological studies have suggested that phthalate exposure may contribute to neurocognitive and neurobehavioral disorders and decreased muscle strength and bone mass, all of which may be associated with reduced physical performance. Walking speed is a reliable assessment tool for measuring physical performance in adults age 60 y and older. OBJECTIVE: We investigated associations between urinary phthalate metabolites and slowness of walking speed in community-dwelling adults ages 60-98 y. METHODS: We analyzed 1,190 older adults [range, 60-98 y of age; formula presented ] from the Korean Elderly Environmental Panel II study and measured repeatedly up to three times between 2012 and 2014. Phthalate exposure was estimated using the following phthalate metabolites in urine samples: mono-(2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), mono-(2-ethyl-5-oxohexyl) phthalate (MEOHP), mono-n-butyl phthalate (MnBP), mono-(2-ethyl-5-carboxypentyl) phthalate (MECPP), and mono-benzyl phthalate (MBzP). Slowness was defined as a walking speed of formula presented . We used logistic and linear regression models to evaluate the association between each urinary phthalate metabolite and slowness or walking-speed change. We also used Bayesian kernel machine regression (BKMR) to examine overall mixture effects on walking speed. RESULTS: At enrollment, MBzP levels were associated with an increased odds of slowness [odds ratio (OR) per doubling increase: 1.15, 95% confidence interval (CI): 1.02, 1.30; OR for the highest vs. lowest quartile: 2.20 (95% CI: 1.12, 4.35) with p-trend across formula presented ]. In longitudinal analyses, MEHHP levels showed an increased risk of slowness [OR per doubling increase: 1.15 (95% CI: 1.02, 1.29), OR for the highest vs. lowest quartile: 1.47 (95% CI: 1.04, 2.06), formula presented ]; whereas those with higher MnBP showed a reduced risk of slowness [OR per doubling increase: 0.84 (95% CI: 0.74, 0.96), OR in the highest (vs. lowest) quartile: 0.64 (95% CI: 0.47, 0.87), formula presented ]. For linear regression models, MBzP quartiles were associated with slower walking speed (formula presented ) at enrollment, whereas MEHHP quartiles were associated with slower walking speed, and MnBP quartiles were associated with faster walking speed in longitudinal analysis (formula presented , respectively). Further, the BKMR analysis revealed negative overall trends between the phthalate metabolite mixtures and walking speed and DEHP group (MEHHP, MEOHP, and MECPP) had the main effect of the overall mixture. DISCUSSION: Urinary concentrations of prevalent phthalates exhibited significant associations with slow walking speed in adults ages 60-98 y.