Interventions:
The interventions were adequately described. The usual practice seems to have been included, which was good for local decision-makers.
Effectiveness/benefits:
The effectiveness data were from a retrospective cohort study. All children had a final diagnosis of global developmental delay by the end of the study, based on their performance on standard developmental tests. The analysis was limited to the diagnostic performance of the tests, so any treatment utility or disutility related to true and false positives was not analysed. It was not clear what downstream benefits might occur from the different testing pathways.
Costs:
No perspective was stated, but a hospital perspective appears to have been taken. The resource use categories and their sources were reported, but it was not clear where the unit costs were from. The total costs from each source were not clearly reported. The analysis was limited to the diagnostic performance of the tests, so any treatment costs for true and false positives were excluded. The authors stated that the costs of additional diagnostic tests that could be avoided with aCGH first were not included, but it was unclear to what extent these would increase the costs for karyotyping; a sensitivity analysis defining and including these costs would have been useful. The authors indicated that the turnaround time for aCGH (10 days) was much quicker than for karyotyping (three months), which could prevent parental anxiety and eliminate additional testing.
Analysis and results:
There was no evaluation of uncertainty around the cost-effectiveness estimate, but 98% confidence intervals were reported for the cost differences per child in the cohort. The authors reported a saving of CAD 106 for aCGH for all children in the cohort, but an incremental cost of CAD 1,379 per additional diagnosis. It was not clear which patients were included in these figures, and if these statements were consistent with each other. The authors discussed most of the limitations stated above, and reported useful comparisons with the results from other studies.
Concluding remarks:
There were a few significant limitations to the study methods and it was unclear how the cost-effectiveness ratios were calculated; the validity of the authors' conclusions is unclear.