LACE index (readmission / post-discharge risk)

Enter data for the index hospitalization being discharged to the community. Length of stay is converted to whole days (floor). Charlson should be the total index score used in your chart or administrative data.

Length of stay (days) *

<1 d → 0; 1→1; 2→2; 3→3; 4–6→4; 7–13→5; ≥14→7 pts (Table 3, van Walraven et al.)

Emergent (acute) admission? *

Charlson comorbidity index (total score) *

Maps to LACE C: 0→0, 1→1, 2→2, 3→3, ≥4→5 pts

ED visits in prior 6 months *

0→0, 1→1, 2→2, 3→3, ≥4→4 pts (per index admission)

Disclaimer: Educational tool only. LACE quantifies modeled risk from a specific cohort; it does not diagnose readmission or mandate interventions. Use with current evidence, institutional pathways, and patient context.

LACE index — what it is

The LACE index (Length of stay, Acuity of admission, Charlson Comorbidity index, Emergency department use) was derived by van Walraven and colleagues to estimate the risk of death or unplanned readmission within 30 days after discharge from hospital to the community (CMAJ 2010). Scores range from 0 to 19.

Component scoring (Table 3)

Component	Value	Points
L — Length of stay (days)	< 1	0
	1	1
	2	2
	3	3
	4–6	4
	7–13	5
	≥ 14	7
A — Acuity of admission	Yes (emergent / acute)	3
A — Acuity of admission	No (elective / non-emergent)	0
C — Charlson comorbidity index	0	0
	1	1
	2	2
	3	3
	≥ 4	5
E — ED visits in prior 6 months	0	0
	1	1
	2	2
	3	3
	≥ 4	4

Final LACE score = sum of points for L + A + C + E. Charlson weights in the original report followed Schneeweiss updates (see CMAJ footnote): e.g. 1 point each for prior MI, PVD, cerebrovascular disease, uncomplicated diabetes; 2 for CHF, COPD, mild liver disease, cancer; 3 for dementia or connective tissue disease; 4 for moderate–severe liver disease or HIV; 6 for metastatic cancer.

Expected probability (derivation model, Table 4)

The publication provides calibrated expected probabilities from the logistic model for each integer score (0–19), from about 2.0% at score 0 to 43.7% at score 19. The calculator displays the value for your total score. Observed rates in validation subsets track these estimates but vary by sample.

How to use the result

Stratify post-discharge planning, follow-up intensity, and patient education—not as a sole criterion for admission or discharge.
Compare trends and risk adjustment in quality programs only with methods appropriate to your data source (administrative vs primary).
Re-validate performance if applied to surgical-only cohorts, different countries, or modern care pathways.

Reference: van Walraven C, Dhalla IA, Bell C, et al. Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community. CMAJ. 2010;182(6):551-557. doi:10.1503/cmaj.091117

What problem does the LACE index address?

Unplanned return to hospital and early mortality after discharge are common, costly, and sometimes related to gaps in transitions of care. Clinicians and health systems often need a transparent, quick way to stratify risk using variables that are usually available at discharge—either from the chart or from administrative data. The LACE index was designed for that purpose: it converts four routinely available factors into a single score from 0 to 19 that aligns with modeled probabilities of an early adverse post-discharge outcome.

It is important to be precise about the endpoint. In the original prospective cohort, the outcome was death or unplanned (nonelective) readmission within 30 days after the patient was discharged home from a medical or surgical hospitalization. The index therefore estimates risk for this combined endpoint, not readmission alone. Combining death with readmission reduces bias that can occur if deaths are ignored when readmission is studied in isolation.

Origin and cohort context

The index was derived and internally validated using adults discharged to the community from medical and surgical services across multiple hospitals. Patients were required to be living in the community (not nursing-home residents) and able to participate in follow-up. The cohort included a mix of elective and emergent admissions and a wide range of diagnoses and procedures typical of general hospital practice. Because performance metrics depend on the underlying population, scores should be interpreted as risk estimates anchored to the derivation framework, then adjusted with local judgment when applied to highly selected groups (for example, only elective surgery, only obstetrics, or specialized transplant services).

The four components and the LACE mnemonic

LACE is an acronym:

L — Length of stay captures the intensity and complexity of the index hospitalization as reflected by how long the patient remained an inpatient.
A — Acuity of admission distinguishes emergent from non-emergent (elective) hospital entry, proxying for instability at presentation and less planned care pathways.
C — Comorbidity uses the Charlson comorbidity index total score to summarize chronic burden and illness severity carried into the admission.
E — Emergency department use counts ED visits in the six months before the index admission, reflecting healthcare utilization patterns, access issues, disease instability, or fragmented care that often precede hospitalization.

Each domain contributes discrete points, and the final LACE score is the sum of all four components.

How each component is scored

Length of stay (L)

Length of stay is scored in whole-day bands tied to the index admission:

Less than one day: 0 points
1 day: 1 point
2 days: 2 points
3 days: 3 points
4–6 days: 4 points
7–13 days: 5 points
14 days or more: 7 points

In practice, hospitals may record length of stay slightly differently (calendar midnights versus 24-hour blocks). For consistency, teams should agree whether the value used for LACE is the same LOS reported on the discharge abstract and used for quality reporting.

Acuity of admission (A)

Acuity is binary in the published index:

Emergent (acute) admission: 3 points
Elective / non-emergent admission: 0 points

Mapping real-world encounters to this field requires a reliable source (admission status in the chart, triage data, or administrative urgency flags). When documentation is ambiguous, the score’s reliability for that patient decreases; this is a limitation of any rule-based tool rather than a flaw unique to LACE.

Charlson comorbidity index (C)

The Charlson index summarizes selected chronic conditions with weights that reflect their association with one-year mortality in historical cohorts. For LACE, the total Charlson score is not used as a continuous predictor in the bedside form; instead it is mapped to discrete points:

Charlson 0 → 0 points
Charlson 1 → 1 point
Charlson 2 → 2 points
Charlson 3 → 3 points
Charlson 4 or higher → 5 points

Because Charlson implementations differ (ICD code lists, coding version, updated weights, timing of diagnosis capture), two sites can produce different Charlson totals for clinically similar patients. For fair comparisons—especially in research or health-system benchmarking—teams should standardize coding rules and confirm which weighting scheme is applied. The original LACE report referenced updated weights consistent with contemporary comorbidity scoring practices at the time of publication.

Emergency department visits (E)

ED utilization in the six months prior to the index admission is scored as:

0 visits → 0 points
1 visit → 1 point
2 visits → 2 points
3 visits → 3 points
4 or more visits → 4 points

Operational definitions matter. In administrative applications, analysts typically count ED encounters captured in ambulatory care reporting systems across a defined look-back window. In primary-data applications, clinicians may rely on chart review, patient report, or regional health information exchange feeds. The index admission ED encounter should be handled consistently with the rule set you adopt (often, prior visits exclude the index presentation depending on data architecture).

Score range, discrimination, and calibration

The LACE score spans 0 through 19. In the derivation work, the model showed moderate discrimination for the combined endpoint, with a C statistic around the high 0.60s to low 0.70s depending on the analytic slice—typical for parsimonious indices built from a handful of variables. Discrimination describes how well the score ranks patients by risk; it does not by itself guarantee that predicted probabilities will match observed event rates in a new setting.

Calibration is equally important: for each integer score, the publication provides an expected probability of death or unplanned readmission within 30 days after discharge, ranging from roughly 2% at score 0 to roughly 44% at score 19 in the fitted model. These probabilities are useful as anchors for counseling and planning, but they can drift when case mix, coding practices, or post-discharge services differ from the derivation era. External validation studies in other countries and data sources have reported variable calibration, which is expected for any transportable model.

How clinicians and teams often use LACE in practice

Common uses include prioritizing discharge planning intensity (medication reconciliation, teach-back education, early outpatient follow-up, home services), flagging patients for phone follow-up or nurse navigation, and supporting case management workflows. Some organizations embed LACE into electronic tools to generate automated risk flags at discharge, while others apply it selectively for higher-complexity services.

LACE can also support communication across sites when a concise risk summary must travel with the patient (for example, from inpatient teams to primary care). Because the inputs are relatively standard, it can align clinical and analytics discussions—provided everyone agrees on data definitions.

Limitations and cautions

Endpoint scope: LACE targets death or unplanned readmission within 30 days, not longer horizons, elective readmissions, observation stays, or avoidable versus unavoidable returns.
Social determinants and function: Factors such as housing instability, substance use, caregiver availability, health literacy, and frailty are not explicit LACE components, yet they strongly influence post-discharge outcomes.
Interventions change risk: A high score does not mean an adverse outcome is inevitable, and a low score does not remove risk. Effective discharge interventions can mitigate hazards that the score only estimates.
Data integrity: Miscoded admission acuity, incomplete capture of comorbidities, or missing ED history will distort the score.
Not a standalone decision rule: LACE should complement, not replace, bedside judgment, specialty consultation, institutional pathways, and patient preferences.

Practical tips for consistent implementation

Document which LOS definition, Charlson implementation, and ED visit counting rules your program uses, and keep them stable for trending.
Re-evaluate model performance locally if your population or care models shift materially (new care pathways, telehealth expansion, payer-driven home programs).
Pair LACE with actionable workflows so risk stratification leads to specific, measurable supports rather than labels alone.

CalcMD.

LACE Index for Readmission Risk

Learn the LACE index (0–19): length of stay, emergent admission, Charlson score, and prior ED visits for modeled 30-day risk after hospital discharge. Scoring, calibration, limitations, and clinical uses for transitions of care.