The selection of comparators was appropriate as the proposed programme was compared against the conventional approach before implementation of the new protocol. Key items of the programme were described.
The evidence came from a retrospective comparison of intervention and control groups of patients who were identified in two different periods of time. Study groups were comparable at baseline but statistically significant differences were observed for open operations, percentage of cancer cases and other clinical factors. Multivariate analysis showed that the operative approach had an effect on length of operative stay. The study design aimed to compare different eras and reflect two alternative systems of managing this patient population. The main drawback of this design was the potential for time-related bias. Factors other than the study intervention might have affected the clinical outcomes. These issues should be considered when assessing the validity of the clinical side of the study. Some statistical analyses were conducted to deal with these issues. The large sample size of the study was a strength of the clinical analysis.
The economic analysis was restricted to hospital costs incurred in managing the patient population. Costs were presented as macro-categories and were not broken down into individual items. No clear information on resource quantities was provided, which reduced the transparency of the economic side of the analysis. In particular it was unclear whether the costs of the enhanced programme were included. All economic data were derived from the economic databases of the participating hospitals but no details of the methods of data collection were given. The impact of variations in cost data was not considered. Conventional statistical tests were applied to determine the statistical significance of cost differences. The price year was not reported.
Analysis and results:
The study results were presented clearly. There was no synthesis of costs and benefits because of the cost-consequences design of the analysis. The authors did not investigate the issue of uncertainty by means of sensitivity analysis. Some statistical analyses were conducted to assess the significance of differences in clinical and economic outcomes. The authors acknowledged some limitations of their analysis, mainly related to the clinical design of the study. The study findings should be considered specific to the authors’ setting and the external validity of the analysis is low.
The study had some methodological limitations that might affect the validity of the authors’ conclusions.