Objective: To clarify the background difference between drug-induced photosensitivity and ultraviolet-visible absorption spectrum or structure and to construct useful information for prevention and prediction of drug-induced photosensitivity. Methods: We investigated whether, for 457 drugs for which the ultraviolet-visible absorption spectrum is listed in the Japanese Pharmacopoeia, there were absorption maxima in the UVA (320 nm or more and less than 400 nm), UVB (280 nm or more and less than 320 nm), or UVA and UVB (280 nm or more and less than 400 nm). Structure was investigated for the existence of “conjugated”, carbonyl, sulfone, nitro and fluorine. The case drug group was taken to be those drugs for which photosensitivity was listed as a side effect on the medical drug package insert. Using statistical software, SPSS statistics ® 24 (IBM), we performed univariate logistic regression analysis, and multivariate logistic regression analysis with a stepwise increment method (likelihood ratio) combining items with p＜0.2, and calculated the odds ratio (hereinafter: aOR). The significance level was taken as 0.05. Results: There were 85 drugs in the case drug group, and 372 drugs in the control drug group. As a result of multiple logistic regression analysis, in Model 1, we placed sulfone (aOR: 4.55, 95% C.I.: 2.22-9.35), fluorine (aOR: 3.66, 95% C.I.: 1.82-7.39) and nitro (aOR: 4.46, 95 % C.I.: 1.73-11.48) in this order. In Model 2, we placed sulfone (aOR: 4, 40, 95% C.I.: 2.12-9.15), fluorine (aOR: 3.81, 95% C.I.: 1.87-7.76), UVA (aOR: 2.40, 95% C.I.: 1.37-4.18) and nitro (aOR: 3.61, 95% C.I.: 1.39-9.40) in this order. Conclusion: When a drug is developed, its ultraviolet-visible absorption spectra and structure become clear, and from this information,measures can be taken which bear the potential risk of photosensitivity in mind.