A Nationwide Population-based Study From Taiwan Assessing the Influence of Preventable Hospitalization Rate on Mortality After Oncologic Surgery.

Objective: To determine the effect of a previously unassessed measure of quality-preventable hospitalization rate-on mortality after oncologic surgery for 4 procedures with established volume-outcome relationships. We hypothesize that hospitals with higher preventable hospitalization rates (indicating poor quality of primary care) have increased hospital mortality. Additionally, patients having surgery at hospitals with higher preventable hospitalization rates have increased mortality.

Summary Background Data: Although different factors have been used to measure healthcare quality, most have not resulted in long-term hospital-based improvements in patient outcomes.

Methods: We retrieved data from Taiwan's National Health Insurance database for patients who underwent surgery during 2001 to 2014 for esophagectomy, pancreatectomy, lung resection, or cystectomy. Preventable hospitalization rates assess hospitalizations for 11 chronic conditions that are deemed to be preventable with effective primary care. The outcome was 30-day surgical mortality. Identifiable factors potentially related to surgical mortality, including surgeon and hospital volume, were controlled for in the models.

Results: Our dataset contained 35,081 patients who had surgery for one of the procedures. For all procedures, hospitals with high preventable hospitalization rates were associated with higher mortality rates (all P < 0.01). For esophagectomy, lung resection, and cystectomy, the adjusted odds of individual mortality increased by 8% to 10% (P < 0.01) for every 1% increase in the preventable hospitalization rate. For pancreatectomy, the adjusted odds of individual mortality increased by 21% for every 1% increase in preventable hospitalization rate when the rate was ≥8% (P < 0.01).

Conclusions: Preventable hospitalization rates could serve as warning signs of low quality of care and be a publically-reported quality measure.