Consecutive cycles of hospital accreditation: Persistent low compliance associated with higher mortality and longer length of stay.

Objective: To examine the association between compliance with consecutive cycles of accreditation and patient-related outcomes.

Design: A Danish nationwide population-based study from 2012 to 2015.

Setting: In-patients admitted with one of the 80 diagnoses at public, non-psychiatric hospitals.

Participants: In-patients admitted with one of 80 primary diagnoses which accounted for 80% of all deaths occuring within 30 dyas after admission.

Intervention: Admission to a hospital with high (n = 125 485 in-patients) or low compliance (n = 152 074 in-patients) in both cycles of accreditation by the Danish Healthcare Quality Programme.

Main outcome measures: A 30-day mortality, length of stay (LOS) and all-cause acute readmission. We computed adjusted odds ratios (OR) and hazard ratios (HR) using logistic and Cox Proportional Hazard regression including adjustment for six potential patient-related confounders.

Results: The 30-day mortality risk for in-patients admitted at high compliant hospitals was 3.95% (95% confidence interval (CI): 3.84-4.06) and 4.39% (95% CI: 4.29-4.49) at low compliant hospitals. In-patients admitted at low compliant hospitals had a substantially higher risk of dying within 30-day after admission (adjusted OR: 1.26 (95% CI: 1.11-1.43) and a longer LOS (adjusted HR of discharge: 0.89 (95% CI: 0.82-0.95) than in-patients at high compliant hospitals. No difference was seen for acute readmission (adjusted HR: 0.98 (95% CI: 0.90-1.06)). Focusing on the second cycle alone, in-patients at partially accredited hospitals had a higher 30-day mortality risk and longer LOS than admissions at fully accredited hospitals (30-day: adjusted OR: 1.12 (95% CI: 1.02-1.24) and LOS: adjusted HR: 0.91 (95% CI: 0.84-0.98)).

Conclusion: Persistent low compliance with the DDKM (in Danish: Den Danske Kvalitetsmodel) accreditation was associated with higher 30-day mortality and longer LOS.