An Approach to Assessing Patient Satisfaction in HCAHPS Databases by Analyses Using Distances.
Abstract Background: Patient satisfaction studies have explored domains of patient satisfaction, the determinants of domains, and score differences of domains by patient/hospital characteristics but reports on the structure of patient satisfaction with respect to similarities among domains are scarce.Objective: To explore whether similarities among patient-satisfaction domains are influenced by hospital characteristics by analyses using distances, and to design a model evaluating relationships between the structure of patient satisfaction and hospital characteristics.Methods: Hospital Consumer Assessment of Healthcare Providers and Systems 2012 survey scores and their structural measures from the Hospital Compare website reported the adjusted percentages of scale for each hospital. Contingency tables of nine measures and their ratings were designed based on hospital characteristics, followed by three different analyses using distances - clustering, correspondence analysis and ordinal multidimensional scaling – for robustness to identify homogenous groups with respect to similarities.Results: Of 4,677 hospitals, 3,711 (79.3%) met the inclusion criteria and were analyzed. The measures were divided into three groups plus cleanliness. Certain combinations of these groups were shown to be dependent on hospital characteristics. High value ratings for communication and low value ratings for medication explanation, quietness and staff responsiveness were not influenced by hospital characteristics, but the varied-ratings domain group similarities, including items such as global evaluation and pain management, were affected by hospital characteristics.Conclusions: Analyses using distances can reveal the hidden structure of patient satisfaction. This study suggests that hospital characteristics including hospital size, the ability to provide acute surgical treatment, and hospital interest in improving medical care quality are factors which may influence the structure of patient satisfaction.