What is the relationship between mortality alerts and other indicators of quality of care? A national cross-sectional study

Abstract

Objectives To assess whether mortality alerts, triggered by sustained higher than expected hospital mortality, are associated with other potential indicators of hospital quality relating to factors of hospital structure, clinical process and patient outcomes. Methods Cross-sectional study of National Health Service hospital trusts in England (2011–2013) using publicly available hospital measures reflecting organizational structure (mean acute bed occupancy, nurse/bed ratio, training satisfaction and proportion of trusts with low National Health Service Litigation Authority risk assessment or in financial deficit); process (mean proportion of eligible patients who receive percutaneous coronary intervention within 90 minutes) and outcomes (mean patient satisfaction scores, summary measures of hospital mortality and proportion of patients harmed). Mortality alerts were based on hospital administrative data. Results Mortality alerts were associated with structural indicators and outcome indicators of quality. There was insufficient data to detect an association between mortality alerts and the process indicator. Conclusions Mortality alerts appear to reflect aspects of quality within an English hospital setting, suggesting that there may be value in a mortality alerting system in highlighting poor hospital quality.