Nonsuicidal self-injury (NSSI) involves the deliberate damage of one’s own bodily tissue without suicidal intent. A number of psychological disorders and indicators of distress are correlated with the behavior, including suicidal behaviors (e.g. Whitlock & Knox, 2007), depression (e.g. Ross & Heath, 2002), and anxiety (e.g. Victor & Klonksy, 2014), and yet the research literature has been mixed on whether increased frequency of NSSI is correlated with increased levels of these variables. The present study hypothesized that these relationships are curvilinear. Data from a larger study were analyzed using curvilinear regression analyses, and hypotheses were partially supported. Curvilinear relationships were found between NSSI frequency and both depression and anxiety, such that the relationships were positive until approximately 300 incidents, after which they became negative. The relationship between NSSI and suicide ideation was positive and linear. Among the whole sample, there were curvilinear relationships between both NSSI and suicide attempts as well as NSSI and suicide threats. Among only the portion of the sample who reported a history of these variables, there was no relationship between frequency and suicide attempts, and a curvilinear relationship between frequency and suicide threats that declined after 325 incidents. Results add to the current understanding of NSSI frequency and provide support for evidence that conflict with the proposed frequency criterion for nonsuicidal self-injury disorder.