Do Hospital and Physician Volume Thresholds for the Volume-Outcome Relationship in Heart Failure Exist?

Background: Although volume-outcome relationships have been explored for various procedures and interventions, limited information is available concerning the effect of hospital and physician volume on heart failure mortality. Most importantly, little is known about whether there are optimal hospital and physician volume thresholds to reduce heart failure mortality.

Objectives: We used nationwide population-based data to identify the optimal hospital and physician volume thresholds to achieve optimum mortality and to examine the relative and combined effects of the volume thresholds on heart failure mortality.

Methods: We analyzed all 20,178 heart failure patients admitted in 2012 through Taiwan's National Health Insurance Research Database. Restricted cubic splines and multilevel logistic regression were used to identify whether there are optimal hospital and physician volume thresholds and to assess the relative and combined relationships of the volume thresholds to 30-day mortality, adjusted for patient, physician, and hospital characteristics.

Results: Hospital and physician volume thresholds of 40 cases and 15 cases a year, respectively, were identified, under which there was an increased risk of 30-day mortality. Patients treated by physicians with previous annual volumes <15 cases had higher 30-day mortality compared with those with previous annual volumes ≥15 cases, and the relationship was stronger in hospitals with previous annual volumes <40 cases.

Conclusions: This is the first study to identify both the hospital and physician volume thresholds that lead to decreases in heart failure mortality. Identifying the hospital and physician volume thresholds could be applied to quality improvement and physician training.