Portal de Periódicos da CAPES

Examining the Effect of EVS Spending on HCAHPS Scores: A Value Optimization Matrix for Expense Management

2013; Lippincott Williams & Wilkins; Volume: 58; Issue: 5 Linguagem: Inglês

10.1097/00115514-201309000-00005

1944-7396

Deirdre McCaughey, Samantha Stalley, Eric Williams,

Economic and Environmental Valuation

EXECUTIVE SUMMARY Using the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey, the Centers for Medicare & Medicaid Services' Value-Based Purchasing program has now linked patient care experience rating to hospital revenue reimbursement, thereby establishing a key relationship between revenue cycle management and the patient experience. However, little data exist on the effect of hospital resource spending on patient HCAHPS ratings. This article examines environmental services (EVS) expenses and HCAHPS ratings on hospital cleanliness and overall patient experience ratings to determine how these variables are related. No linear relationship between EVS expense spending and HCAHPS ratings was found, but post hoc analysis identified a matrix that differentiated on hospital cleanliness ratings and overall EVS spending. A value score was calculated for each quadrant of the matrix, and it was determined that organizational value derives from management of expense spending rather than pursuit of high HCAHPS scores. A value optimization matrix is introduced, and its four quadrants are described. With increased emphasis on subjective patient experience measures attached to financial consequences, leaders in the healthcare industry must understand the link between expense management and HCAHPS performance. This study has shown that effective operations are derived from the efficient use of resources and are supported by strong leadership, strategic management, and a culture of patient-centered achievement. The capacity of healthcare organizations to identify their unique coststo-outcomes balance through the value optimization matrix will help provide them with a means to ensure that optimal value is extracted from all expense spending. INTRODUCTION In October 2012, the Centers for Medicare & Medicaid Services (CMS) began to reimburse inpatient hospitals for Medicare patient services on the basis of the value of services provided rather than purely on service quantity. Through its Value-Based Purchasing (VBP) program, CMS awards hospitals points; those with more points receive higher reimbursement (Office of the Federal Register, 2011a). Points are earned for either high performance or large improvement in publicly reported quality measures. In the first year of the program, 30% of these points are earned on the basis of patients' experiences as measured by the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey. While links between patient outcomes and reimbursement are not new, VBP incorporates a growing trend in revenue cycle management: the patient experience. The VBP patient experience score emphasizes patients' perceptions of their hospital stay, a subjective measure of hospital performance. This change in reimbursement forces hospitals to rethink strategies to increase value and maximize reimbursement. While hospitals have historically focused on clinical outcomes to measure quality and optimize reimbursements (Carling et al., 2008; Lehrman et al., 2010; Porter & Teisberg, 2006), VBP requires a broader strategy beyond objective clinical process measures. Yet, the majority of literature focuses on clinical outcomes to measure value (Carling et al., 2008; Lehrman et al., 2010; Porter & Teisberg, 2006). As such, the purpose of our study is to determine if a relationship exists between hospital spending on patient experience-enhancing services (cleanliness and environmental services) and patient experience scores as measured by HCAHPS. PATIENT PERCEPTIONS AND HOSPITAL PERFORMANCE Recent studies suggest that expenses associated with improvements in the patient experience domain will positively affect hospitals and increase focus on the experience-outcome relationship (Balik, Conway, Zipperer, & Watson, 2011; Beryl Institute, 2010; Center for Health Care Quality, 2010). The relationship between patient experience and hospital performance is emphasized by publicly reported measures on forums such as Hospital Compare, which encourage consumers to actively choose among healthcare facilities. Since the beginning of public reporting, many HCAHPS measures have, on average, seen improvement nationally, including the cleanliness domain (Elliott et al., 2010a). In addition, the Robert Wood Johnson Foundation-funded program Aligning Forces for Quality found that patient perceptions of quality underlie nearly 30% of variations in hospital financial performance (Center for Health Care Quality, 2010). Given the robustness of the patient experience-hospital performance relationship, patient experience should be included in the healthcare value equation. Studies linking value to patient experience and HCAHPS indicators are scarce, and research is necessary to define metrics that quantify how patient experience adds value to healthcare services. One HCAHPS patient experience domain includes hospital cleanliness; through VBP, hospitals directly attribute value to patient perception of room and bathroom cleanliness (Office of the Federal Register, 2011a). Investments in services that improve patient perception of hospital cleanliness should positively affect HCAHPS scores, providing value from these investments. Elliott, Kanouse, Edwards, & Hilborne (2009a) found that a hospital's physical environment, including cleanliness, was the third highest contributor to a hospital's overall HCAHPS rating. Moreover, patients identified cleanliness as an indicator of hospital attention to details and of infection risk level (Elliott et al., 2009a; Sofaer, Crofton, Goldstein, & Crabb, 2005). The importance of the hospital's physical environment was also found during focus groups (Sofaer et al., 2005), in which 34% of patients included hospital room and bathroom cleanliness in their top two HCAHPS measures. Similarly, hospital executives acknowledge the importance of cleanliness and consistently place it among their top three priorities (Beryl Institute, 2010). Evidence supporting these relationships was found in studies that show hospital physical environments have a significant positive impact on revenue enhancement in addition to overall patient experience (Balik et al., 2011; Sadler, Joseph, Keller, & Rostenberg, 2009). Furthermore, better hospital performance on the HCAHPS cleanliness measure is correlated with improvements in the patient safety measures developed by the Agency for Healthcare Research and Quality (AHRQ) (Isaac, Zaslavsky, Cleary, & Landon, 2010) and decreased in-hospital fall rates (Tzeng, Hu, Yin, & Johnson, 2011). The AHRQ measures and a similar hospitalacquired condition measure will also be included in the VBP payment calculation in future years (Office of the Federal Register, 2011b). Additional financial impact is partly derived from a relationship between overall patient satisfaction and hospital cleanliness. The literature suggests that investing in environmental services (EVS) can positively affect patients' perceptions of cleanliness and their overall impression of the care they receive, which may subsequently result in higher HCAHPS scores and higher Medicare reimbursements. On the basis of these findings, we hypothesize the following: H1.A positive relationship exists between 1a.EVS expenses (costs) and HCAHPS patient experience scores (room cleanliness) 1b.EVS expenses (labor hours) and HCAHPS patient experience scores (room cleanliness) 1c.EVS expenses (costs) and HCAHPS patient experience scores (overall rating) 1d.EVS expenses (labor hours) and HCAHPS patient experience scores (overall rating) METHOD Design and Sample The data were obtained from ARA-MARK Healthcare, a large, international company that provides contracted support services to hospitals, healthcare facilities, and senior living communities. ARAMARK provided EVS expense data for a representative sample of 75 hospital accounts in the United States. HCAHPS data for the sample were supplied by ARAMARK and collected from the Hospital Compare database (HHS, 2012). Three hospitals were dropped from the data, as specific links to HCAHPS could not be ascertained. The remaining hospitals range in size from 30 to 1,025 beds, the majority (94.4%) are located in urban centers, and 66.2% are not-for-profit organizations (NFPs). Measures HCAHPS scores. HCAHPS is a National Quality Forum-endorsed survey of patient outcomes and perceptions regarding hospital care. The environmental cleanliness score was derived from the percentage of patients who responded “always” on HCAHPS Question 8, “How often your room and bathroom were clean” (range 50% to 84%). The overall rating score was derived from the percentage of patients who rated the hospital a 9 or 10 on HCAHPS Question 21, “[R]ate the hospital overall” (range 69% to 90%). All scores reflect averages across the period of April 2009 through March 2010; at the request of ARAMARK, the HCAHPS variables have been converted to z-scores to maintain proprietary data confidentiality. Environmental expense. Environmental expenses were measured using two variables. The first is the average total EVS cost per 1,000 square feet per month, and the second is average labor hour usage per 1,000 square feet per month. Both variables were calculated from 2010 fourth-quarter data. The expense variables have also been converted to z-scores to maintain the confidentiality of ARAMARK's proprietary data. Control variables. Bed size categories were derived from the American Hospital Association (2011) and have been collapsed to four categories ( 499 beds) to reflect the sample size. Hospital location was coded as either urban (0) or rural (1), and ownership status was coded as NFP (0), religious NFP (1), government (2), or for-profit (3). These variables are examined as standard control variables in other HCAHPS analyses (Jha, Orav, Zheng, & Epstein, 2008; Lehrman et al., 2010). Analysis Data analyses were performed using SPSS version 20.0 (SPSS, 2011). Due to sample size (n = 71), ordinary least squares (OLS) regression was used to examine the data relationships. While the sample size is small for the number of variables in the regression, our equation falls within acceptable cases-tovariables ratio guidelines. Urdan (2010) recommends 30 cases plus 10 for each predictor variable (n = 70), while Brace, Kemp, and Snelgar (2009) identify a 10:1 ratio of cases per variable (n = 40) as an acceptable level. FINDINGS Descriptives The intercorrelations of HCAHPS scores (environmental cleanliness rating and overall rating) and EVS expenses (total costs and labor hours) were calculated. Correlation coefficients and descriptive statistics were computed for all variables, and correlations between the variables trended in the expected directions and within acceptable levels with respect to multicollinearity. EVS labor hours were positively correlated with EVS expenses (R = .543, p < .01) and negatively correlated with HCAHPS cleanliness score (R = -.262, p < .05). HCAHPS cleanliness score is positively correlated with HCAHPS overall rating (R = .665, p < .01). Regression Analysis Two sets of OLS regression models were run with two predictor outcomes. Three control variables are included in the analysis: bed size, location (urban, rural), and ownership status (for-profit, NFP, religious NFP, and government). Hypotheses 1a and 1b state that a positive relationship exists between EVS expenses (total dollars and labor hours) and HCAHPS patient experience scores (room cleanliness). To test these hypotheses, a series of multiple regressions were conducted, with HCAHPS room cleanliness scores being regressed on environmental expenses (total costs and labor hours). The first regression model demonstrated that, while the model is significant, no significant relationship was found between environmental expenses (total costs) and HCAHPS room cleanliness scores (adjusted R2 value of .287, F = 7.79, p = .007). Only the control variables, bed size (β = -.309, p = .007) and for-profit ownership status (β = .542, p = .044), were found to be significant (see Table 1).TABLE 1: Regression Analysis Results ( N = 71)The next regression model demonstrated that, again, while the model is significant, no significant relationship exists between environmental expenses (labor hours) and HCAHPS room cleanliness scores (adjusted R2 value of .282, F = 7.21, p = .009). The same two control variables were found to be significant (see Table 1): bed size (β = -.306, p = .009) and for-profit ownership status (β = .560, p = .037). These findings fail to support Hypotheses 1a and 1b. However, smaller hospitals and for-profit hospitals tend to have higher room cleanliness scores. Hypotheses 1c and 1d state that a positive relationship exists between EVS expenses (total costs and labor hours) and HCAHPS patient experience scores (overall hospital rating). To test these hypotheses, a third model was run and found no significant relationship between environmental expenses (total costs) and HCAHPS overall hospital rating score (adjusted R2 value of .069, F = .093, p = N/S), with for-profit status (β = .667, p = .031) being significant. The results from the final model also were not significant, indicating no relationship between EVS expenses (in labor hours) and HCAHPS overall hospital rating score (adjusted R2 value of .045, F = 1.82, p = N/S), yet for-profit status (β = .646, p = .031) was found to be significant, indicating that for-profit hospitals tend to have higher overall rating scores. These findings fail to support Hypotheses 1c and 1d. Overall, these findings fail to support the study's hypotheses, as a positive correlation does exist between EVS expenses (in either total costs or labor hours) and HCAHPS patient experience scores (room cleanliness and overall hospital rating). Upon reaching these nonsignificant findings, we turned to additional analysis examining value (a key component of the VBP program) to better understand the relationship between spending and scores. According to Porter and Teisberg (2006), value is defined as being equal to patient outcome per dollar of cost. Post Hoc Analysis Value Using Porter and Teisberg's (2006) value definition, a formula for an environmental services value equation was calculated by setting value as the outcome of HCAHPS scores divided by expenses. The environmental services value equation was calculated for each hospital by dividing HCAHPS score (room cleanliness) by EVS expenses (in dollars). It was found that the average value of the sample is .21, with a range of .11 to .44. The higher the value, the better the patient outcome per dollar of expense and the better the cost-effectiveness of EVS in relation to patient experience. Using a scatter plot displaying hospital value scores, the graph was divided into quadrants. It was found that data points distinctly clustered into four quadrants. An analysis of variance (ANOVA) tested for significant differences in the EVS value averages between the four quadrants to assess potential differences in costeffectiveness (patient outcome per dollar of cost). Quadrants Quadrant 1. The 12 hospitals in this quadrant have lower EVS expenses and higher HCAHPS scores. The average bed size in this quadrant is 282 beds. Quadrant 1 has the highest average value, .303, making it the most cost-effective in providing environmental services (see Figure 1).FIGURE 1 Value Optimization Matrix Matrix Analysis of HCAHPS Scores/EVS ExpensesQuadrant 2. Hospitals in this quadrant (n = 11) have higher EVS expenses and higher HCAHPS scores. The average hospital size in Quadrant 2 is 183 beds, the smallest of the quadrants. Quadrant 2's average value is .180. This measure places Quadrant 2 third in terms of the environmental services value equation. Quadrant 3. The hospitals in Quadrant 3 (n = 25) have lower HCAHPS scores and lower environmental expenses. This quadrant had the second highest average bed size, with 362 beds. The average environmental services value equation for these hospitals is .243, the second highest of all quadrants. Quadrant 4. Hospitals in this quadrant (n = 23) have higher expenses and lower HCAHPS scores. These hospitals had the largest average bed size at 417. Quadrant 4 is the lowest value quadrant at .152, indicating the least cost-effective environmental services. The ANOVA test found EVS expenses to be significantly correlated with HCAHPS cleanliness scores [F (3, 67) = 33.75, p < .001]. Follow-up tests were conducted to evaluate pairwise differences across the means; the Levene statistic was found to be significant (p = .003). Thus, equal variances were assumed and a Scheffe test was applied. The results of the post hoc test are shown in Table 2.TABLE 2: ANOVA Post Hoc Analysis ( N = 71)Significant differences were found in the means between Quadrant 1 (high cleanliness and low expenses, mean = .303, p < .05) and the other three quadrants. Quadrant 1 had the highest value. Quadrant 2 (high cleanliness and high expenses, mean = .180, p < .05) was found to have the second lowest value score and was significantly lower than Quadrants 1 and 3, while the results were nonsignificant compared to Quadrant 4. Quadrant 3 (low cleanliness & low expenses, mean = .243, p < .01) was found to have the second highest value score, which was significantly higher than both Quadrants 2 and 4 yet lower than Quadrant 1. Finally, Quadrant 4 (low cleanliness and high expenses, mean = .152, p < .05) was found to have the lowest value score of all the quadrants. The post hoc analysis offers evidence that the value scores within the four quadrants are significantly different than each other. DISCUSSION The principal finding of this study is the lack of relationship between EVS spending and HCAHPS scores. However, post hoc analysis revealed the value optimization matrix, which identifies four distinct value quadrants (see Figure 1). The value matrix shows that hospitals that spend less than average on environmental services (Quadrants 1 and 3) had higher value scores (were more cost efficient) than hospitals spending more than average on environmental services (Quadrants 2 and 4). That is, more spending does not guarantee a higher HCAHPS environmental score per dollar spent; instead, HCAHPS value is derived from how those expense dollars are managed. Understanding the implication of HCAHPS on hospital performance is challenging given the dearth of literature examining HCAHPS and hospital performance. While evidence has shown that hospital performance is improving on overall HCAHPS measures (Elliott et al., 2010a), other studies have found that hospital HCAHPS performance varies as a result of diverse factors, including patient hospitalization reason, patient ethnicity, and patient mix (Elliott et al., 2009a, 2009b, 2010b). With this variety of influencing factors, few studies appear in the HCAHPS literature with which to interpret our findings. As HCAHPS surveys are intended to provide publicly available data with which healthcare consumers can identify high-performing care providers and help healthcare organizations identify areas for quality improvement (Elliott et al., 2010a), we turn to the patient quality literature to interpret these findings. In an attempt to understand these value findings, we use Donabedian's (1966, 1980) structure-process-outcomes (SPO) framework. The SPO framework is useful because it provides a theoretical basis by which to interpret our findings linking HCAHPS performance and hospital quality performance improvements, which is a primary purpose of the HCAHPS program (CMS, 2012). Originally proposed as a mechanism for evaluating quality of patient care (Donabedian, 1966), the SPO framework has been extensively used to understand quality of care, patient safety, adverse event frequency, and patient experiences (Hearld, Alexander, Fraser, & Jiang, 2008; Kobayashi, Takemura, & Kanda, 2011; Pronovost, Miller, & Wachter, 2006). Its strength is that it provides analytical methodology that allows one to define, categorize, and measure structures, processes, and outcomes to evaluate their effectiveness and value. Structure is defined as the stable prerequisites that facilitate the provision of services within an organization (Donabedian, 1980). Examples include organizational size, volume, staffing, and ownership (Hearld et al., 2008; Kunkel, Rosenqvist, & Westerling, 2007). Hospitals vary greatly on many of the structure factors that are important to achieving success in terms of patients' perceptions and evaluation of cleanliness (Bush, 2011). A hospital's size, patient volumes, and status as either an acute, long-term care, or outpatient facility contribute to its overall throughput and ability to maintain the cleanliness of the facility (Bush, 2011). A process captures the ways in which the structural items are used or converted into actions. Examples include patient care processes, EVS expenses, quality improvement efforts, leadership initiatives, and staffing ratios (Hearld et al., 2008; Kunkel et al., 2007). These process variables represent the basis from which outcome variables are derived. Outcomes are the measureable results of the processes. Examples of outcomes include event frequency, quality metrics, and HCAHPS results. Value Optimization Matrix Using Donabedian's (1966, 1980) framework to interpret our findings, we introduce the value optimization matrix, which features the four quadrants formed by spending and HCAHPS scores (see Figure 2).FIGURE 2 Value Optimization MatrixQuadrant 1: The Investors Quadrant 1, which we refer to as the Investors, has above-average HCAHPS scores and below-average expenses, suggesting that the hospital strategically invests its finite resources. This finding begs the question of how that feat was accomplished. The SPO framework suggests that both structural and process variables affect the outcomes accomplished. Research on quality outcomes (Hearld et al., 2008; Kobayashi et al., 2011) identifies a variety of structural and process variables that affect quality outcomes. These reviews found that the structural variables of staffing (higher) and professionalization/autonomy (higher) were both consistently related to high-quality outcomes. They also identified leadership as the key process driver of quality of care. Thus, it is possible that the key to efficiently achieving high HCAHPS scores is having a strong organization at both the strategic and operational levels. We believe that Quadrant 1 hospitals, the Investors, reflect this strategic positioning; hospital leadership articulates cleanliness (e.g., hand-washing campaigns) as a strategic goal and aims to have employees buy into these goals as a reflection of the hospital's culture of quality. On the financial side, tight budgetary control may be observed for EVS expenditures coupled with rigorous return on investment (ROI). The reward system for employees may also be aligned with the organizational strategy around cleanliness. At the operational or EVS level, within the environmental team, training and leadership seem to be well integrated into daily operations. Appropriate staffing, professional motivation, and education are key features of these institutions, characterized by pride of institution and job-well-done thinking. Teams in Quadrant 1 hospitals often feature good communication, empowerment, trust (autonomy), and clear performance expectations. The physical environment for these teams should provide a safe working environment, functioning equipment, and easily accessible resources for job performance. Quadrant 2: The High-Rollers Returning to the matrix, Quadrant 2 hospitals, the High-Rollers, make significant investments in cleaning and achieve high HCAHPS scores but gain less value than Quadrant 1 hospitals do as a result of their spending. These organizations are likely to feature many of the best practices demonstrated in Quadrant 1 hospitals, as they generate high HCAPHS scores, but the key difference is that spending by Quadrant 2 hospitals is not strategically applied. That is, resources are spent without a clear understanding of which investments garner the best ROI. The High-Rollers invest resources to generate high HCAHPS scores without value consideration, rather than spending strategically, as those in Quadrant 1 do. High-Roller spending is expended on a variety of items that may or may not have adequate ROIs relative to HCAHPS. Quadrant 3: The Savers Quadrant 3 hospitals, the Savers, spend less than average on environmental services and achieve lower-than-average HCAHPS scores. Interestingly, they have the second highest value average next to Quadrant 1. However, where hospitals in Quadrant 1 may be well-run organizations that yield high HCAHPS scores for their strategic investment, Saver hospitals lack both the financial investment and the organizational resources to yield high HCAHPS scores. In spite of this lack of resources, they yield the second highest value for their efforts. This finding suggests that their organizational efforts in EVS management trump (in value terms) the large investment of resources by Quadrant 2 hospitals, the High-Rollers. Quadrant 4: The Spenders Quadrant 4 hospitals, the Spenders, expend more than average on environmental services but achieve lower-thanaverage HCAHPS scores. In the value analysis, they have the worst return on investment. Quadrant 4 hospitals have similar spending patterns to hospitals in Quadrant 2, but unlike hospitals in Quadrants 1 and 2, Spender hospitals do not seem to have as effective an organization in place to turn these investments into high HCAHPS scores. Implications Given the industry's shifting focus toward healthcare value, these results suggest that high value can be achieved in two ways. First, high-value environmental services can be derived from low expenses and high HCAHPS scores, indicating highly efficient investments and processes in EVS (the Investors, Quadrant 1). Second, high-value environmental services can be derived from low expenses and lower HCAHPS scores than those achieved by Investor hospitals (the Savers, Quadrant 3). Theoretically, hospitals can choose either of these strategies to increase the value of environmental services because, under the current fee-for-service reimbursement model, HCAHPS scores do not contribute to the financial viability and sustainability of hospitals. However, under CMS's reimbursement model, hospitals in Quadrant 1, the Investors, will likely see higher reimbursement due to higher HCAHPS scores in conjunction with higher quality measures. Higher reimbursement combined with these hospitals' strategic and efficient use of resources will result in even higher value and ROI. Although high value can currently be achieved through low expenses regardless of HCAHPS scores, emphasis on consumer choice and VBP reimbursement will shift the high-value model toward Quadrant 1. Focusing on efficient use of resources, both expenses and labor, will provide the highest value and highest ROI. As noted earlier, one purpose of publicly reporting HCAHPS data is to provide hospitals with information to help direct performance and quality improvements (Elliott et al., 2010a). In support of that goal, the value optimization matrix offers managers a tool with which to balance resource management and fiscal prudence with HCAHPS performance improvement, thereby contributing to better overall quality outcomes for both the hospital and its patients. Limitations The first limitation pertains to the sample. The sample size is small (n = 71) and is not representative of the broader population of U.S. hospitals. Second, our data do not include adjustments for differences among hospitals that may influence HCAHPS ratings, such as age of facility and status as an academic medical center. As organizations become more familiar with HCAHPS, senior leadership in some healthcare systems argue that factors such as these negatively influence HCAHPS performance (Bush, 2011). Third, our study is limited by the assumption of linearity inherent in our regression analysis. A levelingoff effect between service expenses and patient satisfaction is possible, indicating a curvilinear relationship. At successively higher levels of service expenditures, the impact on patient satisfaction diminishes. Directions for Future Research To determine the validity and reliability of the value matrix, additional studies are needed that examine the value equation using other expense and outcome measures (e.g., expense spending on patient safety and quality). We encourage researchers to investigate what other variables contribute to or are a component of value creation. Our findings need to be replicated with a larger, more diverse sample and include other types of healthcare organizations, such as those that provide rehabilitative or long-term care. Longitudinal studies would help ascertain if seasonal patterns emerge from the data, if value equations are influenced by annual costs, and if EVS resource spending at Time 1 is consistent with outcomes at Time 2 to determine whether the value equation holds over time. CONCLUSION The healthcare industry is developing an understanding of the link between subjective patient experience measures and financial consequences. As this relationship is formalized through VBP, emphasis on patients' experiences will grow. The literature suggests that expenses associated with improvements in the patient's experience positively affect hospitals (Kobayashi et al., 2011). However, organizations cannot just invest resources to pursue performance and operational excellence without understanding whether and what value can be derived from those expenditures. In other words, effective operations are derived from the efficient use of resources and supported by strong leadership, strategic management, and a culture of patient-centered achievement. Identifying an organization's unique costs-to-outcomes balance through the value optimization matrix will help provide its leaders with a means to ensure that optimal value is extracted from expense spending.