BACKGROUND: Ultraviolet (UV) B radiation increases serum 25-hydroxyvitamin-D3 [25(OH)D], but the influence of UVA1 and UVA/narrowband UVB (UVBnb) phototherapy on serum vitamin D is unknown. OBJECTIVE: We sought to investigate the influence of UVBnb, UVA1, and UVA/UVBnb phototherapy on serum levels of 25(OH)D and related parameters in patients with an inflammatory skin condition. METHODS: 25(OH)D, as well as calcium, parathormone, phosphate, and albumin were measured before therapy, 2 weeks after start, and after completion of the phototherapy. Diagnoses were divided in 4 groups: atopic dermatitis, psoriasis, morphea, and others. RESULTS: We surveyed 116 dermatologic patients undergoing phototherapy with UVA1 (n = 38), UVA/UVBnb (n = 30), or UVBnb (n = 48) 2 to 3 times a week for 53 to 90 days. UVBnb phototherapy increased serum 25(OH)D from 22.1 to 39.5 ng/mL after the therapy (P < .001). The lower the baseline 25(OH)D level was, the steeper the increase in 25(OH)D was upon application of UVBnb phototherapy. UVA/UVBnb therapy also increased serum 25(OH)D, from 23.9 to 50.3 ng/mL (P = .003). Conversely, in the UVA1 therapy group, 25(OH)D serum levels decreased significantly from 21.9 to 19.0 ng/mL (P < .001). LIMITATIONS: The study design was open trial without randomization. An influence of a precise skin disease cannot be excluded because of the heterogeneous diagnoses. Bias may have arisen from patient preference for treatment at our center, referral, unrecognized differences in underlying skin disease, and other factors.