Knee and hip osteoarthritis (KHOA) are a source of functional impairment. With aging, the management of osteoarthritis (OA) is a major issue in the search for improved quality of life. Spa treatment provides short- and mid-term symptom relief without serious side effects. This study aimed to identify characteristics of patients with KHOA associated with use of spa treatment. The prospective KHOALA cohort included 878 adults aged 40 to 75 years with symptomatic KHOA. We separately analyzed knee and hip OA data and compared patients who never had spa treatment with those who had at least one or multiple treatments during 5 years of follow-up in terms of socio-demographic characteristics, clinical data, quality of life (OAKHQOL, SF-36), physical activity (MAQ), functional impairment (WOMAC), and health care consumption (pharmacological and non-pharmacological treatments). Factors associated with at least one or multiple spa treatments were evaluated with regression logistic models. In all, 607 (69.1%) patients had knee OA (KOA), 222 (25.3%) hip OA (HOA) and 49 (5.6%) both, 91 (13.9%) with KOA, and 33 (12.2%) with HOA had at least one spa treatment. In the KOA cohort, the probability of at least one, two, or three spa treatments was increased with older age (odds ratio = 1.6 [95% confidence interval 1.2–2.2], 1.8 [1.2–2.8], 2.4 [1.4–4.2], respectively), greater use of physiotherapy (OR = 3.9 [2.1–7.1], 2.7 [1.3–5.6], 2.5 [1.1–5.9]), having a prosthesis (OR = 2.1 [1.2–3.8], 2.2 [1.1–4.3], 2.5 [1.1–5.5]), and low MAQ score (OR = 0.7 [0.6–0.9], 0.7 [0.5–1.0], 0.7 [0.5–1.0]). In the HOA cohort, female sex was associated with at least one (OR = 3.0 [1.1–8.0]) or two (OR = 5.1 [1.2–22.5]) spa treatments. In this cohort of KHOA, repeated spa treatment over 5 years was strongly associated with older age, greater use of physiotherapy and presence of a prothesis with KOA and female sex with HOA. This study may help to better understand spa treatment determinants in OA.