A cost-consequence analysis of a nudge intervention to improve hospital care of older people at the end of life: results from a stepped-wedge cluster randomised trial

Objectives The ‘Intervention for Appropriate Care and Treatment’ (InterACT) was a nudge intervention to identify hospital patients at risk of imminent death or deterioration and communicate this information to treating clinical teams. The aim was to improve the quality of care delivered. This paper reports a cost-consequence analysis of the InterACT intervention. Methods A stepped-wedge cluster randomised trial was conducted across three large tertiary hospitals in Australia between May 2020 and June 2021. The cost of implementing the intervention was determined using prospectively collected staff time sheets, study documentation and field notes. Changes to hospital admission costs and health service outcomes between the trial’s intervention and control phases are also reported. Hospital admissions costs and other health service outcomes were obtained from hospital databases and patient chart reviews. Results The mean intervention cost was $A 72 per at-risk patient admission identified. Additional site-level implementation costs ranged between $21 373 to $34 867 per hospital site, translating to $23 per at-risk admission. The intervention did not reduce the cost of intensive care unit admission, length of stay, medical emergency calls or in-hospital deaths. Wide confidence intervals around at-risk admission cost differences (95% CI: –$2264 to $3312) indicated there was large uncertainty. Conclusions This cost-consequence analysis found that the intervention was not effective in reducing the cost of non-beneficial treatment, which is consistent with the broader InterACT results. This simple nudge-intervention alone may not be sufficient to impact health service resource use and costs in the complex end-of-life setting.