Hospital Care for Adult Patients with Congenital Heart Diseases

Objective: The ideal type of hospital to care for adult congenital heart disease (ACHD) patients is not well known. Hospital competitiveness, clinical volume and market structure can influence clinical outcomes. We sought to understand how hospital competitiveness affects clinical outcomes in ACHD patients in the era prior to the Adult Congenital Heart Association accreditation program. Methods: Patient discharges with ACHD diagnosis codes were filtered between 2006-2011 from an allpayer inpatient healthcare database. Hospital-level data was linked to market structure patient flow. A common measure of market concentration used to determine market competitiveness-the Herfindahl-Hirschman Index (HHI)-was stratified into: more competitive (HHI <==25th percentile), moderately competitive (HHI 25th to <75th percentile), and less competitive (HHI >= 75th percentile) hospital. Any complication, home discharge and mortality were analyzed with clustered mixed effects logistic regression. The combined impact of HHI and any complication on mortality by interaction was assessed. Results: A total of 67,434 patient discharges were isolated. More competitive hospitals discharged the least number of patients (N = 15,270, 22.6%) versus moderately competitive (N = 36,244, 53.7%) and less competitive (N = 15,920, 23.6%) hospitals. The adjusted odds of any complication or home discharge were not associated with hospital competitiveness strata. Compared to more competitive hospitals, mortality at moderately competitive hospitals (Adjusted Odds Ratio (AOR) 0.79, 95% CI: 0.66-0.94) and less competitive hospitals (AOR 0.79, 95% CI: 0.63-0.98) were lower (p = 0.025). Age, race, elective admission, transfer status, and payer mix were all significantly associated with adjusted odds of any complication, home discharge and mortality (p = 0.05). Having any complication independently increased the adjusted odds of mortality more than 6-fold (p< 0.001), and this trend was independent of HHI strata. Failure to rescue an ACHD patient from mortality after having any complication is highest at less competitive hospitals. Sensitivity analysis which excluded the transfer status variable, showed that any complication (p = 0.047) and mortality (p = 0.01) were independently associated with HHI strata. Conclusions: Whether lower competition allow hospitals to focus more on quality of care is unknown. Hospital competitiveness and outcome seem to have an inverse trend relationship among ACHD patients. Since medical care is frequently provided away from the home area, hospital selection is an important issue for ACHD patients. Further research is needed to determine why competitiveness is linked to surgical outcomes in this population.