How do I identify high-acuity patients 48 hours before discharge?

The LACE score (Length of stay, Acuity, Comorbidities, ED visits) remains the most validated pre-discharge risk tool. Patients scoring above 10 on LACE have a 30-day readmission probability above 25%. The data needed to calculate this is already in your EHR discharge dataset.

The LACE score provides a validated, data-derived readmission risk estimate from four variables that are universally documented in inpatient EHR systems. Unlike proprietary risk models that require new data inputs, LACE calculation uses data already present in the discharge dataset—which means it can be operationalized without new data collection infrastructure.

LACE scoring methodology: L (Length of Stay, scored 0–7 based on LOS in days: 1 day = 1 pt, 2 days = 2 pts, 3 days = 3 pts, 4–6 days = 4 pts, 7–13 days = 5 pts, 14–27 days = 6 pts, 28+ days = 7 pts); A (Acuity of admission, scored 0 or 3: acute = 3 pts, non-acute = 0 pts); C (Comorbidity burden, scored 0–5 using Charlson Comorbidity Index derived from ICD-10 problem list diagnoses); E (ED utilization, scored 0–4 based on number of ED visits in the 6 months prior to admission: 0 visits = 0 pts, 1 visit = 1 pt, 2 visits = 2 pts, 3 visits = 3 pts, 4+ visits = 4 pts). Maximum total LACE score is 19. Scores above 10 correlate with 25%+ 30-day readmission probability.

The critical timing distinction is where most hospitals miscalculate intervention opportunity. Most hospitals run LACE at the point of discharge documentation, which provides 4–8 hours of intervention runway at best. Running LACE at day 2–3 of admission gives care management teams 48–72 hours to arrange intensive transition support: SNF pre-authorization, home health orders, medication reconciliation review, and guaranteed 7-day follow-up scheduling. This timing difference is the mechanism that converts a risk stratification tool from a reporting exercise into a prevention program.

Standard clinical dashboards show patients at discharge—when the opportunity to intervene is nearly gone. A daily inpatient census view sorted by LACE score would show which current patients have the highest readmission probability, but most hospitals don't run LACE on current admissions—they run it at discharge as part of discharge documentation, if at all.

Aggregate LACE score distributions also hide the specific pathways that generate high scores. A patient with a LACE score of 13 might have scored primarily on Charlson comorbidity (5 pts) and prior ED utilization (4 pts)—indicating a chronically ill, high-utilization patient who needs community health worker support and medication access intervention. A different patient with the same score of 13 might have scored primarily on acute admission acuity (3 pts) and long length of stay (7 pts)—indicating a complex acute illness with functional decline that needs SNF placement and intensive post-acute monitoring. Same LACE score, completely different intervention pathway.

Automate LACE calculation from EHR data at the end of admission day 2 for all inpatients. This requires extracting four data elements: current LOS (from admission date), admission type (acute vs. elective), Charlson Comorbidity Index from the problem list (most EHRs can calculate this from active diagnoses), and ED visit count in the past 6 months from the encounter history. Route all patients scoring 10 or above to the care management team automatically—no manual identification step required.

For high-LACE patients (score 10+), implement a standard intervention protocol that runs in parallel with clinical treatment: pharmacist-led medication reconciliation review by day 3, social work assessment for post-discharge support needs, SNF pre-authorization if functional decline is present, and confirmed 7-day follow-up appointment before any discharge order is entered. The goal is that no patient with LACE above 10 leaves without a confirmed post-discharge appointment already scheduled.

Measure outcomes at the care management intervention level: LACE-flagged patients who received the full protocol versus those where the protocol was partially completed versus those who were missed. Van Walraven's 2010 LACE validation study (CMAJ) showed that high-LACE patients receiving intensive transition support had 30-day readmission rates 25–40% lower than matched controls. Building this comparison internally with your patient population provides the evidence base for program expansion and board-level ROI reporting.