HOSPITAL Score for Readmissions

Why use it

Unplanned 30-day readmissions affect approximately 15–20% of hospitalized patients, carry significant morbidity, and are a major quality and financial burden. The HOSPITAL Score identifies the highest-risk patients at discharge who may benefit from enhanced transitional care, allowing allocation of limited post-discharge resources to those most likely to benefit. It outperforms the LACE Index and LACE+ in several validation studies and is simpler to calculate.

About

The HOSPITAL Score is a validated, point-based risk stratification tool developed by Donzé and colleagues to predict potentially avoidable 30-day hospital readmissions at the time of discharge. Published in 2013 using data from Partners HealthCare System (Boston, USA), the score was derived from 7 easily available clinical variables obtainable at discharge: Hemoglobin level, discharge from an Oncology service, Sodium level at discharge, Procedure performed during the index hospitalization, Index admission type (emergent/urgent), number of hospital admissions in the prior year, and Length of stay. The HOSPITAL Score is validated in international cohorts (Switzerland, France, Australia, Israel) and has been endorsed for use in hospital quality improvement programs. It identifies high-risk patients (score ≥7) who may benefit from targeted post-discharge interventions such as early follow-up, care coordination, and transitional care services.

When to use

• At the time of hospital discharge for all adult inpatients to risk-stratify readmission probability
• In discharge planning meetings to identify patients who need post-discharge follow-up within 7 days
• To guide allocation of care coordination and transitional care resources (social work, home health, SNF referral)
• As part of hospital quality improvement programs targeting readmission reduction
• For electronic health record integration to flag high-risk patients automatically at discharge

Formula

Pearls & pitfalls

• The HOSPITAL Score is most useful at DISCHARGE — all variables are documented or calculable at the time of the discharge note.
• The Oncology service variable (O) carries double the weight (2 points) — cancer patients have dramatically higher readmission rates due to disease complexity, immunosuppression, and treatment toxicities.
• The number of prior-year hospital admissions variable (T) also carries 2 points for ≥5 admissions — high utilizers are the highest readmission risk group and often have complex psychosocial needs.
• The score is validated for "potentially avoidable" readmissions specifically — not all readmissions. Planned readmissions (elective chemo, scheduled procedures) may inflate apparent readmission rates without reflecting care quality.
• Hemoglobin <8 g/dL at discharge may indicate inadequately treated anaemia, active blood loss, or severe systemic disease — address underlying cause before discharge if possible.
• Hyponatraemia (<135 mEq/L at discharge) is a marker of severity and is independently associated with longer LOS, readmission, and mortality across many diagnoses.
• The score has been validated internationally (Switzerland, France, Australia, Israel) with consistent performance — area under the ROC curve 0.70–0.76 across cohorts.
• High HOSPITAL Score does NOT mean readmission is inevitable — it identifies patients who benefit most from targeted transitional care investment.

Next steps

Low Risk (0–4): Standard discharge instructions and routine follow-up. No specific readmission intervention required. Intermediate Risk (5–6): Enhanced discharge counseling. Consider early follow-up appointment within 7–14 days. Targeted patient education. Social work screening. High Risk (≥7): Intensive transitional care interventions: follow-up appointment within 7 days, post-discharge phone call within 72h, care coordinator assignment, home health evaluation if appropriate, medication reconciliation. Consider 30-day care management program enrollment.

Evidence

Donzé J et al. Potentially avoidable 30-day hospital readmissions in medical patients: derivation and validation of a prediction model. JAMA Intern Med. 2013;173(8):632-638. PMID:23529115. Derivation cohort n=9,036; validation cohort n=6,213. AUROC 0.72 (derivation), 0.71 (validation). 30-day potentially avoidable readmission rates: low risk (score 0–4) 5.8%, intermediate (5–6) 14.4%, high (≥7) 28.3%. International validation: Donzé J et al. J Gen Intern Med. 2016. Prospective validation in multiple centres across 4 countries.

References

1. Donzé J et al. Potentially avoidable 30-day hospital readmissions in medical patients: derivation and validation of a prediction model. JAMA Intern Med. 2013;173(8):632-638. — link
2. Donzé JD et al. International validity of the HOSPITAL Score to predict 30-day potentially avoidable hospital readmissions. JAMA Intern Med. 2016;176(4):496-502. — link
3. Graham KL et al. A validated disease-specific 30-day readmission risk score: the HOSPITAL score. Eur Heart J. 2018. — link