Analytical approach:
A decision tree was used to combine the cost and outcome data from the published literature. The time horizon was lifetime. The authors stated that the UK NHS and Personal Social Services perspectives were adopted.
Effectiveness data:
A systematic review was used to identify the relevant clinical data. The review was conducted as part of the development of the National Institute for Health and Clinical Excellence (NICE) guidelines on delirium. One of the identified studies was a non-randomised study including 852 patients (aged 70 years or older) at intermediate and high risk of delirium (Inouye, et al. 1999 see 'Other Publications of Related Interest' below for bibliographic details). The main measure of effectiveness was the expected number of delirium cases. Survival in the first three years after discharge was based on a calculated adjusted hazard ratio of occurrence of death of 1.71 (the increased risk of death associated with delirium versus no delirium) from one study (Rockwood, et al. 1999 see 'Other Publications of Related Interest' below for bibliographic details). The same general population rates (from UK Life Tables) were applied to both groups.
Monetary benefit and utility valuations:
The utility values came from published studies that were identified during the systematic review.
Measure of benefit:
Quality-adjusted life-years (QALYs) and the net monetary benefit were the summary benefit measures. Future benefits were discounted at an annual rate of 3.5%.
Cost data:
The costs were those associated with the intervention and adverse consequences (such as new admission to institution, extended in-hospital stay, long-term care, pressure ulcers). The costs of adverse events were based on published estimates of the unit cost and the duration of time spent with each adverse consequence. The costs associated with the intervention were from the study protocol (Inouye, et al. 1999), supplemented by data from the Personal and Social Services Research Unit. Future costs were discounted at an annual rate of 3.5%.
Analysis of uncertainty:
Uncertainty was investigated using deterministic and probabilistic sensitivity analyses (using 5,000 iterations). The deterministic analysis varied a number of key inputs including the baseline risk of delirium and various cost estimates. Cost-effectiveness acceptability curves were generated in the probabilistic sensitivity analysis.