Objective It is difficult to make the differential diagnosis between tuberculous meningitis (TBM) and cryptococcal meningitis (CM) when the smear is negative. The objective of this study was to create a diagnostic rule for differentiating TBM from CM in adult HIV-negative patients based on clinical and laboratory features. Methods The clinical and laboratory data of 219 adult HIV-negative patients satisfying the diagnostic criteria for tuberculous (n=100) and cryptococcal (n=119) meningitis hospitalized at the Third Affiliated Hospital of Sun Yat-Sen University during the period 2000-2009 were retrospectively analyzed. Features found to be independently predictive of tuberculous meningitis were modeled using a multivariate logistic regression to create a diagnostic rule. The performance of the diagnostic rule was assessed using a prospective test data method. Results Six factors were found to be predictive of a diagnosis of tuberculous meningitis: gender, mental disorders, vision and/or hearing damage, proteins in the cerebrospinal fluid, the total cerebrospinal fluid white cell count and the coexistence of tuberculosis in peripheral organs. The diagnostic rule developed using these features exhibited 78.0% sensitivity, 95.2% specificity, 92.9% positive predictive value and 84.4% negative predictive value. The corresponding values for the diagnostic rule were 70.0% and 88.0% using prospective test data. Conclusion Clinical and laboratory features can be helpful in the differential diagnosis of tuberculous meningitis and cryptococcal meningitis in adult HIV-negative patients.