Estimating prognosis in sickle cell anemia (SCA) assumes greater importance as intensive treatments, such as hematopoietic SCT (HSCT), are being tested. Here we estimate the mortality risk from the walk-PHaSST (Sildenafil Therapy for Pulmonary Hypertension and Sickle Cell Disease) trial of homozygous SCA patients with suspected pulmonary hypertension (19/468 deaths; 10 centers in the US and UK). Parallel investigations were also undertaken in the Cooperative Study of Sickle Cell Disease (CSCCD) and a contemporary urban sickle cell disease population (Case Western Reserve University-University Hospitals (CWRU-UH), Cleveland, OH, USA). One- and two-value positive predictive values for 2-year mortality (from study entry) are calculated using factors that include demographics, laboratory values and clinical evaluations. We define high-, intermediate-, and low-risk SCA as > 15%, 10-15% and < 10% 2-year mortality. In walk-PHaSST, no single factor qualifies as high-risk SCA, although several combinations of two factors (that is, both age > 35 years and history of chronic transfusion) do. Either elevated white blood cell count (> 13.5 × 10(3) cells/mcL, 7/70 deaths) or elevated Tricuspid Regurgitant Jet Velocity (⩾ 3.0 m/s, 8/67 deaths) was individually associated with intermediate-risk disease, as were many two-factor combinations. N-terminal pro-brain natriuretic peptide > 160 ng/L, lactate dehydrogenase > 600 IU/L, history of chronic transfusion, sepsis or age > 35 years are individually associated with low-risk SCA, as are many two-factor combinations. SCA risk was integrated with estimated donor type-associated risk from HSCT to form 'Traffic Light' eligibility criteria for clinical trials of HSCT. This method is adaptable to evolutions in clinical care.