Health Care in America: The Relationship Between Subjective and Objective Assessments of Hospitals

ABSTRACT:

Measuring public service performance is a central issue for modern governments. Less attention, however, has been paid to the similarities and differences between various performance indicators. In particular, we know less about the views of citizens concerning performance and how they are correlated with administrative performance. Examining the growing area of health care, we examine how patient perceptions on service quality are related to a series of objective hospital performance indicators. Regarding the quality of interpersonal care and the process of care, we find convergent validity with patient satisfaction and this validity is relatively consistent across diagnosis types. Patient assessments, however, do not track clinical outcomes, such as mortality. Given the complex context and high levels of information asymmetry in health care, the evidence for convergent validity contributes to the generalizability of the theory linking citizen satisfaction to the provision of public services.

Notes

1 As an example, patients cannot judge the skills of practitioners or how much health care they need or how much they should pay for health care.

2 Public employees’ perceptions of their own performance is also frequently used.

3 To access more information on Hospital Compare, please visit www.medicare.gov/hospitalcompare.

4 We exclude hospitals that are not subject to VBP program, including psychiatric hospitals, rehabilitation hospitals, long-term care, children's hospitals, cancer hospitals, critical access hospitals, and hospitals located in the U.S. territories.

5 The raw score of patient satisfaction is calculated as ‘percent top-box’, the percent of patients who responded in the highest scale (e.g., ‘very good’ in the 4-point Likert scale) (Giordano et al. 2010). Since HCAHPS survey items have different Likert scales (e.g., yes/no, 4-point, 5-point, and 6-point scales depending on questions), it is hard to standardize the final score, except using ‘percent top-box’.

6 The impact of patient satisfaction and mortality rates is complicated by the fact that patients who die are not included in the satisfaction survey for obvious reasons. This selection bias is quite likely to affect the relationship between mortality rates and patient satisfaction.

Additional information

Notes on contributors

Ohbet Cheon

Ohbet Cheon ([email protected]) is a postdoctoral fellow at the Houston Methodist Research Institute. Her research agenda focuses primarily on public management, performance management, social disparities in health outcomes, and policy instruments in health care services.

Miyeon Song

Miyeon Song ([email protected]) is a Ph.D. candidate in the Department of Political Science at Texas A&M University, and beginning in fall 2019 she will be an assistant professor at the University of South Carolina. Her research focuses on public management, organizational theory and behavior, government performance, citizen satisfaction, and representative bureaucracy.

Austin M. McCrea

Austin M. McCrea ([email protected]) is a doctoral student in the Department of Public Administration and Policy at American University. His work focuses on public management with a theoretical emphasis on how managers interact with their environment to impact the delivery of public services.

Kenneth J. Meier

Kenneth J. Meier ([email protected]) is a Distinguished Scholar in Residence at American University and holds faculty appointments at the Cardiff School of Business (UK) and Leiden University (the Netherlands). His research interests include public management, equity and public policy, methodological issues in the social sciences, and new theories of representative bureaucracy, organizational environments, and managerial decision making.