Cost-effectiveness of the newborn screening of long-chain acyl-CoA dehydrogenase deficiency

Full title: Cost-effectiveness of the newborn screening of long-chain acyl-CoA dehydrogenase deficiency

Authors: Castilla I, Vallejo-Torres L, Couce ML, Gorostiza I, García-Hernández L, Dulín-Íñiguez E, Espada M, Valcárcel-Nazco C, García-Pérez L, Posada M, Cuéllar-Pompa L, Serrano-Aguilar P

Contact person: Iván Castilla Rodríguez (ivan.castillarodriguez@sescs.es)

Summary

Introduction

The Ministry of Health, Social Services and Equality aims to reduce the inequalities in the newborn screening programmes offered in the different regions in Spain. After the decision to increase to seven the number of conditions included in the early detection programme offered nationally, decisions makers are now considering the inclusion of very long-chain acyl-CoA dehydrogenase deficiency (VLCADD). Newborn screening programmes have to meet a series of requirements, including providing evidence on the cost-effectiveness of the programme, by comparing the costs and health outcomes of early detection to those related to the clinical diagnosis.

Objectives

To evaluate the cost-effectiveness of incorporating VLCADD to the series of conditions being screened in the newborn screening programme in Spain, and to determine the budget impact of its implementation.

Method

We developed a cost-effectiveness model that compared two options: to include VLCADD to the diseases screened in the national screening programme, and to diagnose this condition by clinical diagnosis. The model takes into account the life expectancy of the newborns, capturing the impact on the long term of the early detection. The perspective of the analysis was that of the National Health Service in Spain, expressing the costs in 2014 prices. We estimated the cost related to the screening programme including screening tests, confirmation tests, as well as treatment and follow up costs of those detected by the programme. These costs were compared with those related to the clinical diagnosis of the condition. The differences in the costs were then compared with the difference on the effectiveness. Effectiveness was measured in life years gained. Health outcomes and costs were discounted at a 3% rate. We undertook a probabilistic sensitivity analysis using Monte Carlo simulation. Finally, we analysed the budget impact of including newborn screening for VLCADD using a five years’ time horizon.

Results

According to the results of this study, newborn screening for primary carnitine deficiency is more costly but more effective than clinical detection. For a lifetime horizon and the estimated amount of neonates who born in Spain annually, around 450,000, the incremental costs are 260,936 € (0.58 € per newborn), and the incremental LYs are 24.33 (0.000054 per newborn), giving an incremental cost effectiveness ratio of 10,723.97 €/LY. The results were found to be sensitive to the probability of sudden infant death among those cases that remain undetected, the probability of missing a case with clinical detection, and the proportion of symptomatic cases despite early detection. The probabilistic sensitivity analysis estimated that newborn screening for primary carnitine deficiency is cost-effective with a probability of 63.7% and for a willingness to pay of 30,000 €/LY. The budget impact for the Spanish newborn children was 67,956 € first year and 116,402 € fifth year.

Conclusions

The introduction of neonatal screening for VLCADD is cost effective for a willingness to pay 30,000 €/LY, and the probability of being cost-effective is around 63.7%.