COST-EFFECTIVENESS ANALYSIS OF UNIVERSAL NEWBORN SCREENING FOR SEVERE COMBINED IMMUNE DEFICIENCY IN WASHINGTON STATE

Purpose: To evaluate the expected cost-effectiveness of the addition of severe combined immunodeficiency (SCID) to the Washington State newborn screening (NBS) panel in accordance with the US Recommended Uniform Screening Panel.

Method: We constructed a decision analysis model to estimate the direct costs and health outcomes of SCID NBS in an annual birth cohort of 86,600 babies. Point estimates and ranges for parameters, including screening test characteristics, survival rates, costs of screening and diagnosis, costs for treating infants with other forms of T-cell lymphopenia detected by screening for SCID, and costs for definitive treatment (hematopoietic cell transplant [HCT] or enzyme replacement therapy for many infants with adenosine deaminase deficiency) were derived from published estimates, expert opinion, and NBS program records. We estimated treatment costs in 2012 US dollars for infants with early or late identification and for infants who die before definitive treatment. We performed one-way and threshold sensitivity analyses to evaluate the influence of key parameters on the outcome measure (net direct cost per life-year saved).

Result: Of 1.49 annual cases of SCID, 0.30 would have been detected in the absence of screening due to a positive family history, and 0.40 annual deaths would be averted through screening for SCID, or one additional survivor every 2-3 years. The total cost of screening, diagnosis, and follow-up in Washington State, which tests each infant twice, is $8.71 per infant. Costs of treatment per infant with SCID are lower with NBS, $150,666 vs. $363,354, and avoided costs are projected to offset 42% of the costs associated with screening. In the base-case scenario, our model suggests an incremental cost-effectiveness ratio (ICER) of approximately $36,000 per life-year saved. One-way sensitivity analyses found that cost-effectiveness results were robust, with ICER values less than $100,000, for plausible parameter assumptions, including birth prevalence of SCID and survival in infants with late-diagnosed SCID. The NBS strategy could potentially be cost-saving, with a negative ICER, if the treatment cost per late-identified infant with SCID who received HCT as first-line therapy were to exceed $1 million. 

Conclusion: Our model suggests that NBS for SCID is likely to be cost-effective relative to benchmarks commonly used in the United States.