Patients are hospitalized for disparate conditions and procedures. Patient experiences with care may depend on hospitalization type (HT).
Determine whether the contributions of patient experience composite measures to overall hospital ratings on the Hospital Consumer Assessment of Healthcare Providers and Systems Survey vary by HT.
In cross-sectional observational data, we defined 24 HTs using major diagnostic category and service line (medical, surgical, or obstetrical). To assess the importance of each composite for each HT, we calculated the simultaneous partial correlations of 7 composite scores with an overall hospital rating, controlling for patient demographics.
Nineteen thousand seven hundred twenty English- or Spanish-speaking adults with nonpsychiatric primary diagnoses discharged home 12/02-1/03 after an overnight inpatient stay in any of 132 general acute care hospitals in 3 states.
Patient-reported doctor communication, nurse communication, staff responsiveness, physical environment, new medicines explained, pain control, and postdischarge information; overall 0 to 10 rating of care.
Nurse communication was most important overall, with a 0.34 average partial correlation (range: 0.17-0.49; P < 0.05 and among the 3 most important composites for all HTs). Discharge information was least important (0.05 average partial correlation; P < 0.05 for 10 of 24 HTs). Interactions demonstrated significant (P < 0.05) variation in partial correlations by HT for 5 of 7 composites (all but responsiveness and environment), with nurse communication, doctor communication, and pain control showing the most variation (F > 2, P < 0.05).
The importance of patient experience dimensions differs substantially and varies by HT. Quality improvement efforts should target those aspects of patient experience that matter most for each HT.
From the RAND Corporation, Santa Monica, California.
Supported by AHRQ contract 5U18 HS09204; and in part by the Centers for Disease Control and Prevention (CDC U48/DP000056) (to M.N.E.).
A preliminary version of this work was presented at the Academy Health Annual Research Meeting, Orlando, FL, June, 2007.
The contents of the publication are solely the responsibility of the authors and do not necessarily reflect the official views of the AHRQ or the CDC.
Reprints: Marc N. Elliott, PhD, 1776 Main Street, Santa Monica, CA 90407. E-mail: firstname.lastname@example.org.