The selection of the intervention was appropriate as the proposed intervention was compared against the usual care for this patient population.
The clinical data came from a published RCT, which is usually considered to be a valid source of evidence given the strengths of its design. The author did not report the details of the patient sample, the randomisation procedure, nor other methodological aspects, such as the use of a power calculation and follow-up. In general, the reporting of the clinical analysis was limited. The author stated that the intermediate outcome of the number of falls avoided was preferred over more general measures such as life-years, as these would have been more difficult to understand from a health service perspective.
The categories of costs were consistent with the viewpoint, and a breakdown of cost items was given. The unit costs and quantities of resources used were reported for all items. The author provided a justification for the exclusion of some cost categories, such as out-of-pocket expenses, which were likely to vary widely in different health systems and services. The sources of data were appropriately reported for all items. The price year was not reported, which limits the possibility of making reflation exercises in other time periods. The economic analysis was satisfactorily reported.
Analysis and results:
The costs and benefits of the two interventions were reported and incremental cost-effectiveness ratios were calculated. The issue of uncertainty was only partially addressed, as the deterministic analysis investigated only variations in the resource use and unit costs. Other aspects of uncertainty were not considered. The author noted that the main limitation of this analysis was the use of self-reported data for both clinical and economic estimates. It is well known that self-reported estimates may not be accurate due to recall bias, but traumatic episodes such as falls might be easier to remember.
Although the economic data was well reported, there were some methodological limitations, which should be considered when judging the validity of the author’s conclusions.