Cost-effectiveness analysis of diagnostic procedures in children with meningeal signs

Oostenbrink R1,3, Oostenbrink JB2, Moons KGM3, Redekop WK2, Derksen-Lubsen G4, Grobbee DE3, Moll HA1.

1Sophia Children’s Hospital, 2IMTA, Erasmus Medical Centre Rotterdam, 3University of Utrecht, 4Juliana Children’s Hospital, The Hague, The Netherlands

Aim Development of a model to assess changes in diagnostic procedures and treatment in patients with meningeal signs in terms of survival, quality-adjusted-life-years (QALYs) and costs.

Methods The model was developed on a large database of 360 children with ‘meningeal signs’ (0.08–15.0 years), visiting the emergency department of the Sophia Children’s Hospital between 1988-1998. Additional data on probabilities of meningitis and adverse outcome, costs and quality weights (utilities) were derived from literature, database and national registries. The time-horizon of the model was 15 years.

Results: 99 children had bacterial meningitis (10 with adverse outcome), 36 other serious bacterial infections and 225 self-limiting diseases. Mean costs per patient were 8393 Euro (65% associated with long-term neurological sequelae). The mean number of QALYs was 11.19. A diagnostic decision rule to minimise lumbar punctures and empirical antibiotic treatment, without missing one case of meningitis, reduced total costs by 4%. Misdiagnosis of bacterial meningitis resulted in a strong increase of costs of neurological sequelae and a strong reduction of QALYs. A vaccination strategy achieving a 50% reduction of bacterial meningitis reduced costs with 40% and increased QALYs with 1.6%.

Conclusion Costs of long-term neurological sequelae predominate the costs of initial diagnosis and treatment. Misdiagnosing of bacterial meningitis leads to a strong increase in costs and a loss of QALYs. The model can contribute to the cost-effectiveness analysis of vaccination strategies, incorporating the long term costs and quality of life.