Comparison of algorithm advice for post-acute care referral to usual clinical decision-making: examination of 30-day acute healthcare utilization.

: Compare patient characteristics and acute healthcare utilization between patients identified as in need of post-acute care (PAC) by the clinical decision support (CDS) algorithm yet were discharged home without services, to those where the CDS and hospital clinicians agreed on no referral. : Retrospective analysis of hospital administrative and clinical data for 1,366 patients. : 30-day acute healthcare utilization rates are significantly higher for those patients flagged as in need of PAC referral. There are also significant differences in patient characteristics based on referral risk. : Clinicians were blinded to the algorithm enabling the comparison of usual care to decision support. Future work will examine the effect of sharing algorithm advice with clinicians on PAC referral rates and utilization. : The CDS algorithm clearly identified patients with high-risk characteristics and those who will go on to utilize acute care resources. Providing CDS to discharge planners may improve patient outcomes.