ECONOMIC ANALYSIS OF TREATMENTS FOR CHILDHOOD IDIOPATHIC THROMBOCYTOPENIA

Purpose: To compare cost-effectiveness of alternative treatments for children hospitalized for acute idiopathic thrombocytopenia (ITP).

Methods: A lifetime Markov model was used to estimate the expected costs and quality adjusted life years (QALYs) associated with the following treatments for ITP:  1) Observation (no treatment); 2) IVIG (single dose 0.8g/kg); 3) Anti-D (single dose 75 mg/kg); 4) Prednisone (4mg/kg per day for 4 days); 5) IV Methylprednisolone (30mg/kg for 3 days). The starting cohort was assumed to be made up of 5 year olds weighing 20kg. During the acute inpatient phase of the model, patients were at risk of intracranial hemorrhage (ICH) every day that platelet levels were <20,000/µL. Patients suffering an ICH were at risk of immediate death. During the long term phase of the model, patients accrued QALYs based on Canadian lifetable mortality rates and age specific utility rates. Patients surviving an ICH were assigned additional long term costs, utility decrements and increased mortality risk. The mean days with platelet levels <20,000/µL for each treatment was based on pooling of published clinical data. Parameter uncertainty was measured using PSA. Various one way sensitivity analyses on key model parameters were also done.    

Results: All results are presented in Canadian dollars. The expected costs and QALYs for the basecase analysis were; $2,805 and 17.461 for observation; $2,074 and 17.469 for IVIG ; $1,932 and 17.466 for Anti-D, $1834 and 17.465 for prednisone; $1,959 and 17.465 for methylprednisolone. The efficiency frontier was made up of the prednisone and IVIG treatment strategies. The incremental cost-effectiveness of moving from prednisone to IVIG was estimated to be $59,506 per QALY. Cost-effectiveness varied according to patient weight. For patients weighing 10kg, IVIG dominated all other treatment strategies. If patients weigh 25kg the incremental cost-effectiveness of IVIG becomes $118,175.  Cost-effectiveness was also very sensitive to the assumed daily probability of ICH with platelet levels <20,000/uL.

Conclusions: This study found that prednisone is a cost effective strategy for treating a 20kg child with acute ITP if societal willingness to pay for a QALY is less than $59,506. If willingness to pay for a QALY is above this level, then IVIG is the most cost-effective strategy. Cost-effectiveness is very sensitive to patient weight.