Introduction:Alcohol withdrawal delirium (AWD) or delirium tremens (DT) is associated with severe complications and high mortality. Prospectively identifying patients with increased risk of developing DT would have important preventive and therapeutic implications. Thus, the present study aimed to identify clinical risk factors predicting the development of DT.Materials and Methods:The study was a cross-sectional quasi-experimental one with equivalent control group, conducted at a tertiary hospital from August 2014 to May 2015. Forty adult male inpatients, diagnosed with DT, were compared with forty age- and sex-matched inpatients in alcohol withdrawal state without delirium. Assessments were done using confusion assessment method, Clinical Institute Withdrawal Assessment of Alcohol Scale, and Mini–Mental Status Examination. For group comparisons, Pearson's Chi-square test and independent sample t-test were used; logistic regression was applied to identify predictors followed by receiver operating characteristic curve analysis.Results:Heavy drinking (P = 0.005; odds ratio [OR]: 1.17, confidence interval [CI]: 1.05–1.31), continuous pattern of drinking (P = 0.027; OR: 4.67, CI: 1.19–18.33), past history of delirium (P = 0.009; OR: 552.8, CI: 4.88–625.7), alcohol-induced psychosis (P = 0.002; OR: 74.6, CI: 4.68–1190), and presence of cognitive deficits (P = 0.044; OR: 12.5, CI: 1.07–147.3) emerged as strong predictors of AWD.Conclusion:The risk factors found can be easily evaluated in a clinical setting for physicians to readily identify patients at risk for developing DT and plan intensive therapies for them. At a neurobiological level, patients with preexisting brain neurotransmitter disturbances are at greater risk for developing DT.