Comprehensive studies of fibrinolysis in non-small cell lung carcinoma have been limited, and assignment of patients to high/low prognosis groups based on arbitrary cut-offs utilizing fibrinolytic measurements is unstandardized. This study was performed in 166 patients to examine the effects of cut-off values determined in three ways. Model 1 assigned patients to one of three equal groups (low, medium, high) based on fibrinolytic measurements made at diagnosis, Model 2 divided patients into low/high groups using median values, and Model 3 grouped according to the parameter being above/below normal range. In model 1, raised plasma fibrinogen, D-dimer and soluble fibrin were positively associated with poorer survival. In model 2, tissue plasminogen activator antigen was additionally related to poorer prognosis. Model 3 identified seven parameters as significantly related to survival, two not identified by the other models becoming significant [plasmin–antiplasmin, tissue plasminogen activator inhibitor-1 (PAI-1) antigen]. Using multivariate analysis, plasma fibrinogen level was the most uniformly significant parameter. Relative risk estimates indicated that raised plasma fibrinogen, soluble fibrin and D-dimer were associated with increased risk of death. Use of the normal/above normal cut-off is recommended to provide the maximum number of significant parameters relating to prognosis, and increased plasma D-dimer, PAI-1 antigen and fibrinogen were most closely related to survival/prognosis.