305 476

Cited 0 times in

Public reporting of hospital-level surgical volume and its influence on patient and hospital behavior

Authors
 한규태 
Issue Date
2017
Description
Dept. of Public Health/박사
Abstract
Background: In Dec 2007, the South Korean government introduced public reporting of hospital-level surgical volume for major surgeries, including gastrectomy in gastric cancer. It aimed to improve options available to patients and the overall quality of hospitals through informing quality indicators and leading benchmarking between hospitals. The purpose of this study was to investigate the changes in patient and hospital behavior after introducing the public reporting of hospital-level surgical volume among gastric cancer patients who underwent gastrectomy.

Materials and Methods: We used the National Health Insurance (NHI) national sampling cohort data on 2,214 patients who were diagnosed with gastric cancer based on the International Classification of Diseases (ICD)-10 (ICD-10: C16) and underwent gastrectomy in 105 hospitals during 2004–2012. We performed an interrupted time series analysis using generalized estimated equation (GEE) models with Poisson distribution and log link function to examine the impact of public reporting on patients’ choice of hospital as an influence on patient behavior. Subsequently, to measure the impact of public reporting on the volume-outcome relationship in length of stay (LOS), inpatient cost, and 1-year mortality as changes in hospital behavior, we performed a linear regression analysis using the GEE model with gamma distribution and log link or survival analysis using Cox proportional hazard model adopting Difference In Difference (DID) methods. Finally, we suggest new optimal surgical volume criteria for achieving a more effective volume-outcome relationship.

Results: In total, 79.27% patients visited a hospital with a higher surgical volume based on first quartile during study period, and this percentage decreased after public reporting, although it was not statistically significant (80.66% to 78.36%, P=0.1909). The averages of LOS and inpatient costs were 15.03 days and 5.51 million KRW, respectively. The LOS decreased after public reporting, but inpatient cost increased, although the difference was not statistically significant (LOS: 16.15 days to 14.31 days, P<0.05; inpatient cost: 4.75 million KRW to 6.02 million KRW, P=0.084). Further, 5.87% of patients died within 1 year after undergoing gastrectomy, and this proportion increased after public reporting, although not significantly (5.17% to 6.34%, P=0.2529). According to the interrupted time series analysis, the time trends after public reporting seemed to be associated with patient’s choice, but this association was not significant considering the policy about reducing copayment in cancer care. According to the results of multiple analyses for LOS, inpatient cost, and 1-year mortality, there were volume-outcome relationships (for LOS, Q2 ratio of LOS [RL]: 0.901, P<0.05; Q3 RL: 0.886, P<0.001; and Q4 RL: 0.785, P<0.001; for inpatient cost, Q2 ratio of cost [RC]: 0.963, P=0.2143; Q3 RC: 0.927, P<0.05; Q4 RC: 0.817, P<0.001; and for 1-year mortality, Q2 hazard ratio [HR]: 0.876, p=0.6013; Q3 HR: 0.937, P=0.7975; Q4 HR: 0.334, P<0.05). Based on the DID methods, such volume-outcome relationships for LOS and inpatient cost were found greater after public reporting. However, for 1-year mortality, this association disappeared after public reporting. On applying the new criteria for surgical volume identified in the present study, the impact of public reporting of the volume-outcome relationships was more significant for LOS and inpatient cost.

Discussion and Conclusion: This study suggested that public reporting about surgical volume in gastric cancer was not associated with patients’ choice of hospital. Patients were more affected by policies on economic support rather than public reporting, and the changes in treatment options may have been affected by an increasing preference for large size hospitals. Thus, public reporting did not operate effectively for improving the options available for patients and their decision making...
Files in This Item:
T014423.pdf Download
Appears in Collections:
1. College of Medicine (의과대학) > Others (기타) > 3. Dissertation
URI
https://ir.ymlib.yonsei.ac.kr/handle/22282913/154960
사서에게 알리기
  feedback

qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse

Links