Wolosin, Robert PhD; Ayala, Louis PhD; Fulton, Bradley R. PhD
Medicare’s hospital inpatient value-based purchasing (VBP) program links a portion of hospital reimbursement to scores on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey.1 Consequently, there is increasing interest in the drivers of patient satisfaction, particularly global satisfaction with hospital care. The importance of the issue has been magnified by the fact that VBP creates winners and losers in a zero-sum game. Unlike the clinical process measures portion of VBP that hospitals control directly, patients themselves are the final arbiters of satisfaction. How patients respond to the HCAHPS overall rating question, in which they choose a number from 0 to 10 to “rate this hospital during your stay,” has financial implications far beyond the innocuous appearance of the item, which is just 1 of 27 items on the survey.
Patient satisfaction and its determinants are important in their own right. Studies have shown that a patient’s satisfaction with his/her healthcare experience is positively related to the patient’s compliance with treatment and healthcare outcomes.2-8 Others have demonstrated that patient evaluations of hospital experiences have been successfully used to improve the quality of delivered care.5,9-11 At the same time, higher levels of patient satisfaction can aid in the retention of existing patients (ie, increased loyalty) and lower the cost associated with new patient acquisition (eg, positive word of mouth as a form of advertisement).12-14
When it comes to discerning the impact of the various components of a healthcare visit on a patient’s overall level of satisfaction with a particular healthcare experience, “there is consistent evidence across settings that the most important health service factor affecting satisfaction is the patient-practitioner relationship.”3(pii)
The inpatient experience is unique in that it exposes patients and their families to components of a healthcare organization, including practitioners, in ways that are quite disparate from experiences with other services (eg, outpatient, emergency rooms). This greater exposure to various aspects of the healthcare organization means more opportunities to affect perceptions of the hospital. An understanding of the relative importance of these components, including doctor and nurse interactions, for overall patient satisfaction, is vital for healthcare organizations looking to improve quality and maximize Medicare reimbursements.
Evidence suggests that nurses’ interaction and communication with patients are most strongly associated with high levels of patient satisfaction in inpatient settings.2,15-19 Various reasons have been posited for this, most having to do with the amount of time and interaction these care providers have with inpatient clients.
Unfortunately, the applicability of much of the previous work on the determinants of inpatient satisfaction is limited by issues related to either the samples and/or the measurement instruments that served as the basis of the research. For instance, previous studies of inpatient satisfaction have utilized facility-specific case studies or small regional samples. This has hindered the ability of researchers to adequately account for a comprehensive set of patient demographics and characteristics (eg, health status, reason for hospitalization) that have been identified as impacting patients’ overall satisfaction with their inpatient experience, potentially limiting the applicability of findings to other hospitals. At the same time, Otani and Kurz16 noted that many previous studies of inpatient satisfaction lack measures for a comprehensive set of visit attributes that cover the full inpatient experience, making it difficult to infer that nursing care is more important than all other attributes of the typical inpatient healthcare visit (ie, facilities, physicians, etc). Still other researchers have warned about the failure in much of this literature to use psychometrically validated surveys/measurement scales20,21; failure to do so reduces the usefulness of any result and the potential to generalize the derived information to the wider patient public.22
This study compares the relative impact of nursing care with other components of hospitalization on overall satisfaction, using an analytic approach addressing the deficiencies of previous research on the determinants of inpatient satisfaction, namely, small sample sizes and poorly validated instruments. In addition to addressing those methodological issues, the specific aim of the study was to identify the effect of various aspects of the inpatient experience on the HCAHPS measure of overall satisfaction to allow hospital administrators to focus resources where they may have the greatest impact.
Survey and Data
The patient-related measures are drawn from a patient satisfaction instrument that incorporates HCAHPS questions with items used for quality improvement. This instrument consists of several demographic questions (eg, age, sex of patient, etc), 22 HCAHPS experience-related questions, and 38 standard satisfaction questions developed and used by Press Ganey Associates, a healthcare consulting firm. The latter set of items is divided into sections labeled admission, room, meals, nurses, tests and treatments, visitors and family, physician, personal issues, discharge, and overall assessment (See Figure, Supplemental Digital Content 1, http://links.lww.com/JONA/A72). Some example items are “speed of admission process” (from the admission section), “how well the nurses kept you informed” (from the nurses section), and “staff concern for your privacy” (from the personal issues section). The non-HCAHPS items had been selected and validated prior to this study; the source survey was revalidated in 2007. As part of the revalidation, a principal components factor analysis (with Promax Oblique Rotation) of test data identified 9 factors that accounted for 73% of the total variance in patient responses. The resulting factors paralleled the structure of subscales on the questionnaire. Reliability estimates (Cronbach α’s) of the subscales range from .77 (admission section) to .95 (physician section). The reliability for the entire questionnaire is α = .97, attesting to the instrument’s high internal consistency.
The wording of the survey questions is such that the patient provides a numeric rating of his/her satisfaction with an aspect of care rather than reporting whether or how often an event occurred. This method is based on the premise that each patient’s evaluation reflects his/her unique reality, making this survey methodology patient-centered rather than provider-centered.23(p156) In other words, patient satisfaction is not constrained to merely a visit-specific frequency assessment. These standard questions are Likert-type items rated on a balanced 5-point response format, with anchor points ranging from 1 (very poor) to 5 (very good). Responses are converted to a 100-point scale by a linear transformation for analysis and reporting purposes.
The data for this analysis represent more than 136,000 randomly sampled respondents who were inpatients at 1 of 302 US hospitals that were randomly selected from the Press Ganey 2008 inpatient database. Table 1 displays salient characteristics of these facilities including their size, geographical distribution, and ownership. The modal hospital was of medium size, located in the Midwest, and organized as non–faith-based, not-for-profit. In order to be eligible to be in the study, a survey must have been received in 2008 from an adult patient (≥18 years old) discharged from a general acute-care hospital after an overnight stay who (1) did not receive psychiatric or rehabilitative services (2) was not a prisoner, (3) did not have an international address, (4) was not discharged to hospice, and (5) whose inpatient stay was less than 90 days. A dual-wave, mail-out, mail-in method was used to reduce the tendency to acquiesce to an interviewer’s presumed preferences or to present oneself in a positive light.24 Surveys were mailed to patients within 1 to 2 days of their records being uploaded to the proprietary database by the hospital; patients were surveyed no more frequently than once every 3 months. The response rate averaged approximately 34%, slightly greater than the 33.17% national response rate given by the Centers for Medicare & Medicaid Services for the 2008 Hospital Compare database. The Hospital Compare database includes facility-level data for all eligible acute care facilities in the United States. Survey data were stripped of identifying information prior to analysis. Research involving the study of existing data, records, and so on, is exempt from institutional review board review if the information is recorded in such a manner that the subjects cannot be identified directly or indirectly.
The major dependent variable of interest is the percent top box on the HCAHPS overall satisfaction item, whose exact wording is: “Using any number from 0 to 10, where 0 is the worst hospital possible and 10 is the best hospital possible, what number would you use to rate this hospital during your stay?” The respondent marks a response by choosing 1 alternative from a vertical array of numbers, from 0 to 10. Only responses of 9 and 10 are considered top-box responses for purposes of this analysis.
Patient satisfaction with each component of the inpatient visit is represented by a composite score derived from those related items drawn from the previously described inpatient survey. Each respondent has a corresponding measure for their satisfaction with the admissions experience, their room, meals, nursing care, tests and treatments, visitors and family, physician, personal issues, and discharge experience.
Underlying differences among patients are accounted for in a number of ways. First, the researchers included controls for patient demographics and characteristics found to affect respondents’ satisfaction with their hospitalization3,25,26 including age, gender, race, education, whether English is the primary language at home, self-reported general health status, and the number of days they were hospitalized (see Table 2). Furthermore, previous research demonstrated that differences in the conditions and/or treatments for which patients are hospitalized (which affect their interaction with staff and hospital resources) can also affect their inpatient satisfaction.15 Therefore, we included a control for patients’ underlying diagnosis in our analysis. Specifically, we defined 55 possible reasons for hospitalization using a combination of the standard 25 major diagnostic codes (MDCs) and the 2 service types (medical or surgical). We split each standard MDC by service type; for those hospitalization conditions that were not assigned to a numeric MDC (eg, transplants, prostate surgical procedure, etc) in version 26 of the Medicare severity–diagnostic related group (MS-DRG) codes, we grouped similar DRGs together in custom MDCs. MDCs for obstetric DRGs and newborns were split out for purposes of this analysis. To account for important differences between the hospitals where the patients stayed and were treated, we included controls for the number of licensed beds at each facility as well as their American Hospital Association–designated region (location), as these nuisance variables are known to influence patient satisfaction.27 To estimate the relative impact of individual patients’ satisfaction with the various components of their inpatient experience on their overall satisfaction with that healthcare organization, Table 3 presents the results of a patient-level logistic regression. A statistical methodology for determining whether a change in the independent variable will increase (or decrease) the odds of an occurrence of a dichotomous dependent variable, logistic regression, was used to determine whether a higher score in a predictor variable (nurses, room, meals, etc) would also increase the probability that a respondent would give the facility a top-box response to the overall rating of the hospital HCAHPS question.
Table 3 reports the results of such a patient-level logistic regression using the dichotomized version of the HCAHPS overall rating of the hospital question as the dependent variable. For each 1-point increase in nursing section score, there is a corresponding increase in the odds of achieving a top box of 4.9%. Therefore, a 5-point increase will yield an increase of exp (b)5 [=1.0495] = 1.27, or a 27% increase, contrasting this with physicians where a 5-point increase will yield an increased probability of 1.0115 [=1.056] or a 5.6% chance of increasing top box.
Discussion and Implications
Our analysis has addressed the deficiencies of previous research on the determinants of inpatient satisfaction, while facilitating the comparison of the relative impact of patients’ satisfaction with their nursing care on their overall satisfaction. Utilizing a comprehensive, psychometrically validated survey conducted on inpatients sampled from facilities nationwide, we identified nursing care as having the most substantial impact on the HCAHPS overall rating top-box percent—a measure of patient satisfaction with implications for hospital reimbursements. Patient satisfaction with rooms, physician care, and meals were also significant predictors of the HCAHPS overall rating top-box percent, but nursing care was clearly the most important. These findings show that hospitals focusing on HCAHPS overall satisfaction, and thus their VBP score and Medicare reimbursement, would likely see the greatest impact by engaging in improvements to nursing care.
Implications for Practice, Health Policy, and Leadership
The implications of the findings are relatively straightforward. For practice, we suggest that nurse workloads should be managed so as to afford nurses the time required to provide personalized patient care. For health policy, we suggest that steps be taken to recognize the critical role of nurses in enhancing the long-term financial health of their employers under VBP. Finally, for leadership, we suggest that nurse leaders institute policies that encourage nurses to enact the behaviors measured via a patient experience inpatient survey, such as responding promptly to the patient’s call button, keeping patients informed of their treatment progress, and paying attention to patients’ special or personal needs. Such policies should include choosing candidates whose skill sets include good interpersonal skills, training with periodic reinforcement, and incentives that reward good performance, such as evidence through patient feedback.
1. Centers for Medicare and Medicaid Studies. Open Door Forum: Hospital Value-Based Purchasing. Fiscal Year 2013 Overview for Beneficiaries, Providers, and Stakeholders. July 27, 2011. Available at www.cms.gov/OpenDoorForums
. Accessed September 19, 2011.
2. Glickman SW, Boulding W, Manary M, et al.. Patient satisfaction and its relationship with clinical quality and inpatient mortality in acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2010; 3 (2): 188–195.
3. Crow R, Gage H, Hampson S, et al.. The measurement of satisfaction with healthcare: implications for practice from a systematic review of the literature. Health Technol Assess. 2002; 6 (32): 1–244.
4. Eisenberg B. Customer service in healthcare: a new era. Hosp Health Serv Adm. 1997; 42 (1): 17–31.
5. Williams B. Patient satisfaction: a valid concept? Soc Sci Med. 1994; 38 (4): 500–516.
6. Aharony L, Strasser S. Patient satisfaction: what we know about and what we still need to explore. Med Care Rev. 1993; 50 (1): 40–70.
7. Cleary PD, Leroy L, Karapanos G, McMullen W. Patient assessments of hospital care. QRB Qual Rev Bull. 1989; 15: 172–179.
8. Stewart M. Which facets of communication have strong effects on outcome—a meta-analysis. In: Stewart M, Roter D, eds. Communicating With Medical Patients. Newbury Park, CA: Sage; 1989: 183–196.
9. Calnan M, Katsouyinannopoulos V, Ovcharov VK, Prokhorskas R, Ramic H, Williams S. Major determinants of consumer satisfaction with primary care in different health systems. Fam Pract. 1994; 11: 468–478.
10. Epstein KR, Laine C, Farber NJ, Nelson EC, Davidoff F. Patients’ perceptions of office medical practice: judging quality through the patients’ eyes. Am J Med Qual. 1996; 11 (2): 73–80.
11. Hays RD, Ware JE Jr. My medical care is better than yours: social desirability and patient satisfaction ratings. Med Care. 1986; 24 (6): 519–524.
12. Friesner D, Neufelder D, Raisor J, Bozman CS. How to improve patient satisfaction when patients are already satisfied: a continuous process-improvement approach. Hosp Top Res Perspect Healthc. 2009; 87 (1): 24–40.
13. Press I. Patient Satisfaction: Defining, Measuring, and Improving the Experience of Care. Chicago, IL: Health Administration Press; 2002.
14. Burroughs TE, Davies AR, Cira J, Dunagan WC. Understanding patient willingness to recommend and return: a strategy for prioritizing improvement opportunities. Jt Comm J Qual Improv. 1999; 25 (6): 271–287.
15. Elliott MN, Kanouse DE, Edwards CA, Hilborne LH. Components of care vary in importance for overall patient-reported experience by type of hospitalization. Med Care. 2009; 47 (8): 842–849.
16. Otani K, Kurz RS. The impact of nursing care and other health care attributes on hospitalized patient satisfaction and behavioral intentions. J Healthc Manag. 2004; 49 (3): 181–196.
17. Oswald SL, Turner DE, Snipes RL, Butler D. Quality determinants and hospital satisfaction: perceptions of the facility and staff might be key influencing factors. Mark Health Serv. 1998; 18 (1): 18–22.
18. Beck KL, Larrabee JH. Measuring patients’ perceptions of nursing care. Nurs Manage. 1996; 27 (9): 32B–32D.
19. Hays RD, Nelson EC, Rubin HR, Ware JE, Meterko M. Future evaluations of the PJHQ Scales. Med Care. 1990; 28 (9): S29–S39.
20. Moret L, Nguyen JM, Pillet N, Falissard B, Lombrail P, Gasquet I. Improvement of psychometric properties of a scale measuring inpatient satisfaction with care: a better response rate and a reduction of the ceiling effect. BMC Health Serv Res. 2007; 7: 197–206.
21. Stizia J. How valid and reliable are patient satisfaction data? An analysis of 195 studies. Int J Qual Health Care. 1999; 11 (4): 319–328.
22. Gonzalez N, Quintana JM, Bilbao A, et al.. Development and validation of an in-patient satisfaction questionnaire. Int J Qual Health Care. 2005; 17 (6): 465–472.
23. Wolosin RJ. The voice of the patient: a national, representative study of satisfaction with family physicians. Qual Manag Health Care. 2005; 14 (3): 155–164.
24. Hall MF. Patient satisfaction or acquiescence? comparing mail and telephone survey results. J Health Care Market. 1995; 15 (1): 54–61.
25. Rahmqvist M. Patient satisfaction in relation to age, health status and other background factors: a model for comparisons of care units. Int J Qual Health Care. 2001; 13 (5): 385–390.
26. Cohen G. Age and health status in a patient satisfaction survey. Soc Sci Med. 1996; 42: 1085–1093.
27. Press Ganey Associates, Inc. Hospital Pulse Report 2010
. South Bend, IN; Press Ganey Associates, Inc; 2010.
© 2012 Lippincott Williams & Wilkins, Inc.