1) Write a paper of 900 words regarding the statistical significance of outcomes as presented in Messina’s, et al. article “The Relationship between Patient Satisfaction and Inpatient Admissions Across Teaching and Nonteaching Hospitals.”
2) Assess the appropriateness of the statistics used by referring to the chart presented in the Module 4 lecture and the resource “Statistical Assessment.”
3) Discuss the value of statistical significance vs. pragmatic usefulness.
4) Prepare this assignment according to the APA guidelines found in the APA Style Guide located in the Student Success Center. An abstract is not required.
Full textTranslateFull text HeadnoteEXECUTIVE SUMMARY
The need for healthcare executives to better understand the relationship between patient satisfaction and admission volume takes on greater importance in this age of rising patient expectations and declining reimbursement. Management of patient satisfaction has become a critical element in the day-to-day operations of healthcare organizations pursuing high performance.
This study is guided by two principal research questions. First, what is the nature of the relationship between patient satisfaction (as measured by scored instruments) and inpatient admissions in acute care hospitals? Second, does the relationship between patient satisfaction (as measured by scored instruments) and inpatient admissions differ between teaching hospitals and nonteaching hospitals? Although not suggestive of direct causation, the study findings revealed a statistically significant and positive correlation between patient satisfaction and admission volume in teaching hospitals only. In contrast, a nonsignificant, negative correlation was seen between patient satisfaction and admission in nonteaching hospitals. In the combined teaching and nonteaching sample, a statistically significant, negative correlation was found between patient satisfaction scores and admission volume.
With financial performance being driven in part by admission volume and with patient satisfaction affecting hospital patronage, the business case for a strategic focus on patient satisfaction in teaching hospitals is clearly evident. The article concludes with a set of recommendations for strengthening patient satisfaction and organizational performance.
In today’s healthcare marketplace, providers increasingly compete against one another for business. In the late 1980s, healthcare executives were confronted with the realization that they could not just increase charges to generate revenue, but rather they had to contain costs as well. Providers now compete on business factors other than price, such as quality, service, reputation, and other nonmonetary attributes. Ettinger (1998) stressed that successful competition relies on the provider retaining awareness of who it wants to serve, what value it creates for the customer, and how it will create that value operationally. In the end, the provider needs to be strategic rather than tactical and proactive rather than reactive. Providers must shift their focus externally to the consumers’ requirements rather than their own.
The need for research regarding patient satisfaction and market share is evident in this age of declining reimbursement and rising patient expectations. Monitoring patient satisfaction has become a standard operating procedure in most healthcare organizations, especially with new Medicare reporting requirements under the HCAHPS program. While patient satisfaction has been widely studied, a gap exists between the impact of customer satisfaction and organizational performance (Kovner and Neuhauser 2004).
The purpose of this research is to study the relationship between patient satisfaction and inpatient admissions among teaching and nonteaching hospitals. The use of inpatient admissions in this study functions as an indicator of volume rather than as a surrogate measure of hospital size. According to Simone (1999), academic healthcare institutions represent an eclectic mix of traditional academia, hospital operations, multiple academic layers, and patients. Today’s teaching hospitals, compared with the nonteaching hospitals, are complex organizations trying to perform an often conflicting array of responsibilities. This complex environment can be organizationally and politically challenging to individuals working in such an environment and, as this study begins to explore, may affect patient satisfaction. Furthermore, a teaching hospital’s central mission is to provide specialized tertiary care that supports its central objective of training new physicians. In contrast, nonteaching hospitals are organizations that provide general medical-surgical care in an environment that is not focused on training and educating physicians.
Two principal research questions frame this study. First, what is the nature of the relationship between patient satisfaction (as measured by scored instruments) and inpatient admissions in acute care hospitals? Second, does the relationship between patient satisfaction (as measured by scored instruments) and inpatient admissions differ between teaching hospitals and nonteaching hospitals?
Although teaching and nonteaching hospitals alike continue their struggle to capture admissions and, ultimately, market share, research on the relationship between patient satisfaction and volume of admissions has been somewhat limited. A prominent aspect of the relatively sparse body of literature on patient satisfaction as a driver of performance is the difficulty in quantifying customer satisfaction’s direct impact on financial indicator outcomes. Accordingly, substitute measures, such as market share or service volume, are often employed as surrogate indicators of organizational performance.
Woodside, Frey, and Daly (1989) provided early evidence to support the premise that patient satisfaction may directly affect volume. The authors conducted an exhaustive literature review of service quality and satisfaction measurement. Based on this review, they developed a framework of relationships among service quality, customer satisfaction, and behavioral intention for service purchases. Service quality, customer satisfaction, and behavioral intention data were collected from patients discharged from two hospitals. Overall customer satisfaction was associated (r = 0.85, p = 0.05) with behavioral intention to return to both hospitals. Despite some question of the generalizability of a two-hospital study, the research does provide substantial evidence for a meaningful relationship between overall customer satisfaction and behavioral intention for buying a major service. A further, recent example of the link between patient satisfaction and service volume can be found at the University of Colorado Hospital (UCH), which launched an online system designed to streamline the arrival process by allowing patients to complete insurance paperwork, patient consent forms, and Health Insurance Portability and Accountability Act notification acknowledgments before visiting UCH. Patient satisfaction scores increased, helping boost outpatient visits in one year from 608,689 to 631,332 (Burt 2006).
Valuable contributions to expanding our understanding of the connection between patient satisfaction and organizational performance outcomes can also be found in groundwork laid in earlier research conducted by Rust and Zahorik (1993). The researchers identified elements of service satisfaction that may significantly affect customer loyalty and market share; however, the focus of their research was on retention of existing business versus new customer development. While retention of patients for future business purposes is important, attraction of new customers for outpatient services, surgical services, and obstetrics clearly translates into increased volume through ancillary referrals.
Finally, research performed by Andoleeb (1998) stressed how the public is inclined to pay more for care from quality institutions with which they were satisfied. Andoleeb’s study identified several variables that shape patient satisfaction with health services, including quality of communication, perceived competence of service provider, quality of facility, demeanor of hospital staff, and perception of cost and patient satisfaction. The explanatory power of these variables underscores that hospital marketing professionals need to be cognizant of these areas. Andoleeb’s argument postulates that a positive association exists between patient satisfaction and patronage (i.e., volume). Accordingly, strategy formulation should focus on gaining a competitive advantage through delivering high levels of service quality, especially in an age of consumerism where perceived service quality is linked to patient satisfaction, which in turn may result in improved patronage (Scotti, Harmon, and Behson 2007).
The study sample consisted of seven teaching hospitals and seven nonteaching hospitals examined over the five-year period from 1999 to 2003 in response to an invitation extended to all Press Ganey client hospitals in New Jersey. Data for all admitted patients who completed the satisfaction survey were included in me study. Press Ganey Associates functioned as the clearinghouse for data to maximize confidentiality of participating hospitals. The sample included seven hospitals in the north region, five hospitals in the central region, and two hospitals in the soum region of New Jersey. The questionnaire mailing yielded study participants from geographic regions exhibiting demographic diversity with respect to income levels, insurance coverage, average age, emnicity, and other characteristics. The patient satisfaction data were collected for each hospital using the complete data sets collected through discharge surveys conducted at the respective institutions. The geographic distribution of New Jersey hospitals statewide is shown in Table 1 .
Teaching hospitals, by their very mission, participate in the education of physicians through formal residency training programs. Depending on the type and number of residency programs offered, a hospital is generally designated either a major teaching or minor teaching institution. To be a major teaching hospital, the facility typically offers residencies in medicine, surgery, obstetrics/gynecology, and pediatrics. Many major teaching hospitals also offer residencies in several subspecialties, such as pathology, anesthesiology, and family practice. A minor teaching hospital typically has only two or three residencies, which may include surgery, geriatrics, or obstetrics/gynecology. Depending on the involvement and politics of an academic university, teaching hospitals are often university hospitals, university affiliated, or independent (Swayne, Duncan, and Ginter 2006).
The study included adults who voluntarily completed surveys mailed to dieir households immediately following discharge. To encourage patients to respond to the survey, a solicitation letter was sent by a representative of the hospital to the patient. In the cover letter attached to the survey, an explanation of the study and me purpose of the survey questionnaire were provided. Consent is implied when patients voluntarily complete the hard-copy survey; enclose it in a sealed, addressed envelope; and return it either to their respective hospital or directly to die clearinghouse. Confidentiality and otiier rights of patients consenting to participate were protected in accordance witii IRB (institutional review board) requirements. Because of restrictions imposed by the survey management process, a formal assessment of nonresponse bias was difficult to ascertain.
The overall scores for this study were obtained from hospitals that had consented to participate after receiving a letter crafted by members of the research team. Hospitals interested in participating in the study submitted their inpatient admission data for the years 1999, 2000, 2001, 2002, and 2003. After coding the patient survey data and pairing them with respective hospital volume data, a database was compiled to facilitate statistical analysis. To preserve confidentiality, the identities of the participating hospitals were blinded, and the database provided only general information about the geographic location of the responding institutions. The names of participating hospitals were shared with the New fersey Hospital Association, which in turn provided an overview of state demographics using the most current data available. Participating sites were offered a copy of the study results. These data were strictly informational and not used in the statistical analysis.
The questionnaire used to measure inpatient satisfaction in this study was first developed in the late 1980s. In 1997, it was modified to maintain its validity in tracking patients’ preferences and experiences. The instrument was developed after conducting customer/ patient focus groups, reviewing the current customer satisfaction literature, reviewing survey instruments from across the United States, and using the latest tools and techniques on survey design from healthcare organizations across the United States. In 2002, a validation study of the Inpatient Satisfaction Survey was conducted to ensure internal consistency and reliability (Press Ganey Associates 2002). The survey instrument was found to be psychometrically stable across a wide spectrum of tests of validity and reliability. The overall Cronbach’s alpha reliability score is 0.94.
The survey included queries related to background, admission process, room, meals, nurses, tests and treatments, visitors and family, physician, discharge, personal issues, and overall assessment. Patient satisfaction was measured using a five-point Likert-type scale labeled as follows: 1 = very poor, 2 = poor, 3 = fair, 4 = good, and 5 = very good. The data were then converted to a 0 to 100 scale, (entering the 1-5 scores and averaging them), with 0 being the low end of very poor and 100 being the high end of very good.
Descriptive statistics in the form of frequencies, means, medians, and standard deviations were computed and used to examine the specific characteristics of the hospitals with respect to their (1) patient satisfaction mean scores and (2) inpatient volume data as measured by admissions. Statistical measures of skewness and kurtosis were also performed to permit scrutiny of the shape and distribution of the survey response data.
Examination of the data revealed that, because of the small sample size, whether the data were normally distributed could not be conclusively determined; therefore, nonparametric statistical testing was chosen to further analyze the data. The Spearman coefficient of rank-order correlation was used to analyze relationships between the independent variable (patient satisfaction mean score) and the dependent variable (volume as measured by admissions). Correlation analyses were performed on a pooled sample of seven teaching and seven nonteaching hospitals. Then an analysis of the differences between the teaching subsample and nonteaching subsample were performed using a Mann- Whitney U-test. Following this test, separate analysis was performed on the seven teaching and seven nonteaching hospitals. The two variables were used not to discover whether a causal relationship existed but to discern whether an association existed between satisfaction mean score and admissions.
The following descriptive statistics are for the aggregate set of teaching and nonteaching facilities. The mean number of admissions across all hospitals was 19,1 1 1 over the five-year period from 1999 to 2003, with a range of 4,513 to 70,465. The aggregate satisfaction mean score was 82.57 for the fiveyear period, with a minimum of 79.05 and a maximum of 86.18. The descriptive statistics for skewness and kurtosis indicate that the admission volumes were not normally distributed. Patient satisfaction mean scores were approximately normally distributed. A summary of descriptive statistics is presented in Table 2.
Spearman rank-order correlation analysis revealed a significant negative correlation (r = -0.287, p = 0.018) between patient satisfaction and admission in the combined sample, suggesting that higher patient satisfaction mean scores are associated with lower inpatient volumes.
A comparative analysis of patient satisfaction that examines differences between teaching and nonteaching hospitals was performed using a MannWhitney U-test. This test is a nonparametric analog of the independent group’s t-test. It was used to determine if differences existed between the two independent groups-teaching and non-teaching-based on rank-ordered scores. Mean rank for teaching hospital patient satisfaction was 25.76; mean patient satisfaction rank for nonteaching hospitals was 45.24. Mann-Whitney U tables can be used to determine significance when there are 20 or fewer cases. At n > 20, the value of U approaches a normal distribution. This study involved five years of data from seven hospitals (n = 35). Therefore, U is transformed to a z-statistic, and the value of ±z can be compared on a table of critical values for a normal distribution. The z-statistic (z = -4.064, p < 0.001) was significant, indicating that the null hypothesis that the two groups-teaching versus nonteaching-are identical must be rejected and the alternate hypothesis that the two groups are significantly different is supported. Here, the p score results were approximately four standard deviations away from the mean.
The mean admission volume of teaching hospitals in our sample (19992003) was 27,745 (median = 22,820; mode = 14,244, multiple modes exist, the smallest is shown) with a range of 14,244 to 70,465. The mean for nonteaching admission volume was 10,722 (median = 12,314) with a range of 4,513 to 16,067. The data indicate that teaching hospitals had higher admission volume than nonteaching hospitals in the years spanning 1999 through 2003. The admission volume data did exhibit deviations from normality in the teaching institutions, but not in the nonteaching institutions.
The mean patient satisfaction score for teaching hospitals was 81.54 (median = 81.78), with a minimum of 79.05 and a maximum of 84.12. The mean nonteaching satisfaction score was 83.58 (median = 83.58), with a range of 80.61 to 86. 18. Analysis of skewness and kurtosis statistics did not suggest a significant departure from normality in the distribution of satisfaction scores (refer to Table 3).
Analysis of the individual Spearman rank-order correlation coefficients revealed a statistically significant and positive correlation (r = 0.581, p < 0.001) between patient satisfaction and admission volume in teaching hospitals. In contrast, a nonsignificant and negative correlation (r = -0.097, p = 0.579) was seen between patient satisfaction and admissions in nonteaching hospitals. The results are reported in Table 4.
The purpose of this study was to examine the relationship between patient satisfaction and volume of inpatient admissions in teaching versus nonteaching hospitals. In the aggregate analysis, the results of the study show a significant, but negative, relationship between patient satisfaction and inpatient volume. Further study revealed that differences exist between mean scores for patient satisfaction in teaching and nonteaching hospitals. When disaggregated into subsamples, a significant, positive relationship is found between patient satisfaction and inpatient volume in teaching hospitals, and a nonsignificant, negative relationship is seen between these variables in nonteaching hospitals. These findings suggest that in the combined teaching and sample, as satisfaction drops, volume increases; however, this counterintuitive conclusion is partially offset by the emergence of a significant positive lation between satisfaction and admission volume in teaching hospitals alone. The later finding may be attributed to the fact that patient satisfaction mean scores of teaching hospitals are statistically lower, thus exhibiting a significant, positive correlation between satisfaction and admission volume.
Another possible explanation for the statistical differences in teaching and nonteaching hospitals’ patient satisfaction is the size and complexity of teaching organizations in contrast to nonteaching facilities with multiple caregivers and contact points with a given patient. This organizational difference in part may explain the lower patient satisfaction scores. Patients in teaching hospitals tend to be immersed in a very complex environment mat, at times, may challenge service providers, as well as the organization at large, to provide patient-focused care (Press 2002). The mission in a tertiary teaching environment contrasts sharply with the mission of a nonteaching community hospital, whose mission is to provide personal healthcare in a manner that uses the available resources most effectively for the community’s benefit (Griffidi and White 2006).
The provision of graduate medical education vastly complicates the process of rendering care in a teaching hospital. Therefore, it is no surprise that statistically significant differences in teaching versus nonteaching mean patient satisfaction scores were found. It is possible that the focus on medi- cal education and the technical aspect of care emphasizes elements of service quality mat patients are not qualified to judge or do not find intrinsically satisfy- ing. Accordingly, less concentration may be placed on the provision of healthcare in a personal manner, to which patients are likely to respond more favorably. It is mis more personal provision of care that is part of the mission in the non- teaching community hospital. The focus in mission may, in part, explain me lower scores in patient satisfaction in teaching versus nonteaching hospitals. Further, length of stay tends to be higher in teaching hospitals and may also affect patient satisfaction scores. Additionally, the typically higher admission rate in teaching hospitals than in nonteaching hospitals may explain the negative correlation associated with lower patient satisfaction mean scores.
The findings in this study suggest that patient satisfaction may be a factor driving volume in teaching hospitals. The study group revealed that for nonteaching hospitals, patient satisfaction and volume growth were not strongly correlated, which may suggest the opportunity for teaching hospitals to capture additional patient volume by studying and revitalizing their approach to and emphasis on patient satisfaction. As such, organizations should consider refocusing their service delivery systems from provider-centric models to patientoriented models.
All patients come to the hospital with their own set of expectations of service and care. Early identification and recognition of these expectations are critical, as true patient satisfaction is derived from the balance of patient culture and clinical culture (Press 2002). Healthcare today is provided by a complex and diverse array of professionals, and patient satisfaction is accomplished through a complicated set of exchanges that translate into a healthcare experience (Sturm 2005). Business success cannot be built on a series of one-time visits, but it can be seriously damaged by a series of one-time experiences. True patient satisfaction means a total, positive healthcare experience.
Results from this study suggest that in some cases improving patient satisfaction pays. Studies confirm the link between patient perceptions of quality and financial measures, particularly profit margins (Press 2002). In a study of 82 hospitals conducted by Harkey and Vraciu (1992), a one standard deviation change in the quality score resulted in a 2 percent increase in operating margin. Garman, Garcia, and Hargreaves (2004) estimated that increasing average patient satisfaction scores from the 3-4 range to the 4-5 range translated into a $2.3 million boost in incremental annual revenue. While further research is clearly needed in the field, healthcare executives might consider practices that include, but are not limited to, enhancing patient satisfaction as a core strategic goal; expanding ongoing satisfaction measurement systems; maximizing steering committee performance results; and implementing sound patient satisfaction training modules to employees, medical staff, and student interns, just to name a few.
Patient-focused care does not mean just listening to the customer but rather making the customer the pivot point of all initiatives to evaluate or redesign care, including (1) customizing service based on patient needs and values, (2) empowering the patient to take control by participating in the care process, (3) sharing knowledge and information, and (4) practicing evidence-based decision making (Bisognano, Lloyd, and Schummers 2007). Consumer-driven healthcare is no longer a speculative trend; it is an emerging reality.
Study Limitations and Future Research
Our study represents only one of a few evaluating the relationship between patient satisfaction and inpatient volume. This study supports the need for further research with larger diverse populations mat might ensure an even higher degree of generalizability. It also raises questions about the need for further research in both the teaching and nonteaching environment.
The competitive nature of the industry and the desire to protect and grow provider markets produced great concern about release of data from the participating hospitals. While geographic representation on the surface appears to have been accomplished, many other influencing variables such as definitive locations, payer mix, bed capacity, marketing budgets, physicians on staff, and major program development were unknown. The influence of these variables could only be evaluated in the broader context of the state. However, release of any one or a series of these demographic data points could have breached the confidentiality of the participating hospitals’ data. Throughout the study, confidentiality was critical to sustained participation of the subjects. The inability to control these influencing variables hence greatly affects the generalizability of these findings.
Methodological limitations, such as the small sample size, may also have affected the findings. Future studies should increase the sample size of teaching and nonteaching hospitals, expand the number of years over which data are collected, and expand into other geographic areas nationwide to maximize the generalizability of the findings. In addition, controlling for age, managed care penetration, payer mix, bed capacity, marketing budgets, clinical program offerings, and medical staff size and satisfaction levels would also be needed to conduct a study to determine the cause of the relationships between patient satisfaction and inpatient volume growth found in this study.
Another methodological adjustment might be to use percentile scoring rather than mean scoring. Using the percentile score places the data in a broader context of where patient satisfaction with the organization falls in relationship to its competitors. An increase in the mean score may not adequately reflect a significant improvement in patient satisfaction overall if the organization’s competitors have increased their mean scores to a greater degree, thereby dropping the measuring organization’s percentile rank. Therefore, patient satisfaction might be more accurately represented by percentile rank than by satisfaction mean scores.
Because this is a cross-sectional study, the findings do not suggest causation but simply establish a correlation between teaching hospital satisfaction and inpatient volume. The study attempts to provide a starting point for further analysis of this relationship. Further research might include adjustments for severity, hospital size, Medicare case-mix index, service line analysis, or length of stay. This last factor may reveal some interesting findings regarding the relationship between satisfaction scores and length of inpatient stay.
This research effort is not intended to be a prescription for increasing volume. Clearly, further study is required to strengthen the conclusion mat patient satisfaction drives volume. Until subsequent research addresses these many unanswered questions through the longitudinal study of larger data sets, making a definitive business decision for the allocation of resources to improve patient satisfaction is difficult. However, the present research supports the conclusion that a positive correlation exists between patient satisfaction scores and volume growth among teaching hospitals, providing one more instrument in the practicing healthcare executive’s tool kit.
Special thanks to Margaret Vargo for her administrative support; Janice Breen, PhD candidate, RN, PHCNS-BC, for her statistical assistance in the development of this research effort; Robin Siegel, MLS, AHIP, for her research assistance; and Jane DeTullio, MA, for her editorial reviews.
Andoleeb, S. 1998. “Determinants of Customer Satisfaction with Hospitals: A Managerial Model.” International Journal of Health Care Quality 11 (6-7): 181-87.
Bisognano, M., R. Lloyd, and D. Schummers. 2007. 10 Powerful Ideas for Improving Patient Care, 52. Chicago: Health Administration Press.
Burt, T. 2006. “Reinventing the Patient Experience.” Healthcare Executive 21 (3): 8-14.
Ettinger, W. H. 1998. “Consumer-Perceived Value: The Key to a Successful Business Strategy in the Healthcare Marketplace.” Journal of the American Geriatric Society 46 (1): 111-13.
Garman, A. N., J. Garcia, and M. Hargreaves. 2004. “Patient Satisfaction as a Predictor of Return-to-Provider Behavior: Analysis and Assessment of Financial Implications.” Quality Management in Health Care 13 (1): 75-80.
Griffith, J. R., and K. R. White. 2006. The WellManaged Healthcare Organization, 6th ed. Chicago: Health Administration Press.
Harkey, J., and R. Vraciu. 1992. “Quality of Health Care and Financial Performance: Is There a Link?” Health Care Management Review 17 (4): 55-63.
Kovner, A. R, and D. Neuhauser. 2004. Health Services Management: Readings, Cases, and Commentary, 8th ed. Chicago: Health Administration Press.
Press, I. 2002. Patient Satisfaction: Defining, Measuring, and Improving the Experience of Care. Chicago: Health Administration Press.
Press Ganey Associates, Inc. 2002. Inpatient Survey Psychometrics Report, pp. 1-7. South Bend, IN: Press Ganey Associates.
Rust, R., and A. Zahorik. 1993. “Customer Satisfaction, Customer Retention, and Market Share.” Journal of Retailing 69 (2): 193-212.
Scotti, D. J., J. Harmon, and S. J. Behson. 2007. “Links Among High-Performance Work Environment, Service Quality, and Customer Satisfaction: An Extension to the Healthcare Sector.” Journal of Healthcare Management52 (3): 109-25.
Simone, J. V. 1999. “Understanding Academic Medical Centers: Simone’s Maxims.” Clinical Cancer Research 5 (9): 2281-85.
Sturm, Jr., A. C. 2005. “Dolls and Dollars: New Insight for Revenue.” Healthcare Financial Management 59 (11): 76-80.
Swayne, L. E., W. J. Duncan, and P. M. Ginter. 2006. Strategic Management of Health Care Organizations, 5th ed. Maiden, MA: Blackwell.
Woodside, A. G., L. L. Frey, and R. T. Daly. 1989. “Linking Service Quality, Customer Satisfaction, and Behavioral Intention.” Journal of Health Care Marketing 9 (4): 5-17.