Introduction: There are several brief screening tools for gambling that possess promising psychometric properties, but have uncertain utility in generalist healthcare environments which prioritize prevention and brief interventions. This study describes an examination of the National Opinion Research Centre Diagnostic and Statistical Manual of Mental Disorders Screen for Gambling Problems (NODS-CLiP), in comparison with the Problem Gambling Severity Index (PGSI), when used to operationalize gambling problems across a spectrum of severity. Methods: Data were obtained from 1058 primary care attendees recruited from 11 practices in England who completed various measures including the NODS-CLiP and PGSI. The performance of the former was defined by estimates of sensitivity, specificity, positive predictive values (PPVs), and negative predictive values (NPVs), when PGSI indicators of problem gambling (5+) and any gambling problems (1+), respectively, were reference standards. Results: The NODS-CLiP demonstrated perfect sensitivity for problem gambling, along with high specificity and a NPV, but a low PPV. There was much lower sensitivity when the indicator of any gambling problems was the reference standard, with capture rates indicating only 20% of patients exhibiting low to moderate severity gambling problems (PGSI 1-4) were identified by the NODS-CLiP. Conclusions: The NODS-CLiP performs well when identifying severe cases of problem gambling, but lacks sensitivity for less severe problems and may be unsuitable for settings which prioritize prevention and brief interventions. There is a need for screening measures which are sensitive across the full spectrum of risk and severity, and can support initiatives for improving identification and responses to gambling problems in healthcare settings such as primary care.