The case fatality rate is 12% to 53%.  In settings with facilities for mechanical ventilation, the most common causes of death are autonomic dysfunction and hospital-acquired pneumonia. Where facilities do not allow for mechanical ventilation, the most common cause of death is asphyxia, resulting from laryngeal spasm, respiratory muscle spasm, or extreme fatigue. 
Prompt diagnosis and prediction of severity are vital in order to determine timely management, including transfer to an intensive care unit and early airway protection. This pre-empts the life-threatening complications of severe disease.
Predictors of severe disease and therefore worse outcome are as follows:
Incubation period (time from injury to first symptom) is inversely related to disease severity. An incubation period of less than 7 days is said to predict grade 3b disease. 
Onset period (time from first symptom to first spasm) is also inversely related to disease severity.
Site of infection: umbilical, uterine, head, and neck predict more severe disease.
Intramuscular quinine injections are reported to carry a mortality of 96%.  Heroin is often "cut" or diluted with quinine, and this may contribute to the high mortality described in drug addicts with tetanus. 
Extremes of age. 
Lack of immunity. Previous immunization, even if incomplete, is associated with milder disease. 
Various groups have devised scoring systems to determine prognosis in tetanus. The Phillips score provides a severity index based on the incubation period, site of infection, state of immunity, and complicating factors.  The Dakar score assesses incubation period, onset period, site of injury, and presence of spasms, fever, and tachycardia on admission.  In a study of 500 consecutive patients (non-neonates) admitted to the Tetanus Unit at the Centre for Tropical Diseases in Ho Chi Minh City, Vietnam, between May 1997 and February 1999, a Dakar score of 3 or greater was associated with a 59% mortality compared with 14% mortality for patients with a Dakar score less than 3. Those with a Phillips score of 17 or greater had a mortality of 34% compared with 11% in the group with a Phillips score less than 17.
A new tetanus severity score has been devised using prospectively acquired data from consecutive patients admitted to the Hospital for Tropical Diseases in Ho Chi Minh City with multivariate logistic regression.  The authors compared their new score with the Phillips and Dakar scores, which were published more than 30 years ago without validation data. Their tetanus severity score had a sensitivity of 77% and a specificity of 82% for a fatal outcome when tested with resubstituted data and showed significantly better discrimination between survivors and nonsurvivors than the Dakar or Phillips scores.
In neonatal tetanus, age less than 10 days, symptoms for less than 5 days and presence of risus sardonicus, opisthotonus, and fever indicate a poor prognosis. Mortality is significantly associated with an incubation period of 6 days or less and weight less than 2.5 kg.   
Tetanus severity score. The final score is calculated from the sum of the scores for each section. A total of ≥8 indicates predicted death; <8 indicates predicted survival
From Thwaites CL, Yen LM, Glover C, et al. Predicting the clinical outcome of tetanus: the tetanus severity score. Trop Med Int Health. 2006;11:279-287
Patients should be instructed to ensure they follow the prescribed immunization schedule and as adults ensure they have a tetanus booster every 10 years.