Patient Experience and Financial Performance
Impact of Patient Experience on Hospital Financial Metrics
Abstract
Patient Experience has long had a direct impact on healthcare organization’s financial performance by way of the Hospital Value Based Purchasing Program. Prior published analysis of patient experience has pointed to a correlation with financial performance. New research indicates that Hospital Value Based Purchasing represents a small portion of the financial incentives a healthcare organization sees when they have a high patient experience rating. Strong patient experience can lead to strong increases in hospital financial performance in and of itself. HydraCor’s analysis of the Centers for Medicare & Medicaid Services published data across the U.S. shows a causation of patient experience and improved financial performance. Hospital profitability as seen by examining Return on Assets, Net Margin and Operating Margin were used to gauge direct causation with patient experience. Our findings conclude that even accounting for factors like Hospital Value Based Purchasing and hospital characteristics, patient experience, as measured by the percentage of patients that rated a hospital a nine or ten on a ten-point scale (Rate Hospital Top-Box) on the Hospital Consumer Assessment of Healthcare Providers and Systems survey, has a significant impact on each financial metric studied. We found there was an approximately 1.04-point increase in ROA, 1.13-point increase in NM and 1.13-point increase in OM on average for each ten-percentage point increase in Rate Hospital Top-Box.
In Addition, sites that utilized the AdvoCor platform had improved financial performance for all metrics as compared to their counterparts that did not (Figure 5.). As evidenced by the data hospitals that utilized the AdvoCor platform had an overall average ROA of approximately 7.83, NM of 7.06 and OM of 6.00 while sites that did not had an average ROA of 4.56, NM of 4.47 and OM of 4.00.
Introduction
Financial incentives for strong patient experience are well known and easily quantifiable in the form of HVBP (Hospital Value Based Purchasing) program incentives where patient experience, represented by the Person and Community Engagement Domain, accounts for 25% of the Total Performance Score2. However there has been little study on the impact of patient experience on hospital financial performance not tied to HVBP. AdvoCor believes, with the recent growth of healthcare consumerism, patient experience may have an impact on hospital financial metrics well beyond the effects from Medicare reimbursement payments.
In 2018 the Beryl Institute conducted a study of consumer perspectives on patient experience through a survey with 2000 respondents. They found that “on a four-point response scale of ‘not at all important’ to ‘extremely important,’ less than 10% suggested experience was only somewhat important or less, 32% of respondents believed patient experience was very important and overwhelmingly 59% of respondents believed patient experience to be extremely important5.” For a vast majority of patients, their experience is a major area of concern when making healthcare decisions. This same study also found that for patients who reported they had a negative experience, 76% will tell another person about the experience. Of those who will tell others about the negative experience, 43% would not return, and 37% will find a new organization and doctor for their care. (Figure 1.)5. Moreover, the study found that for patients who reported a positive experience, 70% will tell another person about the experience and 73% will continue to use the same doctor or organization.
Figure 1. Beryl Institute Patient Experience Survey Findings
These findings suggest that a positive patient experience instills a sense of loyalty to the care provider from current patients, along with attracting new patients through referrals that may have otherwise chosen a different provider. Furthermore, a negative patient experience not only has a strong possibility of losing the current patient but also a greater loss of potential new patients through negative referrals, the largest probability out of any of the categories measured. With this evidence of the importance of strong patient experience in the eyes of the healthcare consumer, AdvoCor has set out to demonstrate the modern impact of strong patient experience on hospital profitability in a quantifiable way.
In 2014 Deloitte studied the impact of patient experience on hospital profitability by examining its effect on Return on Assets (ROA), Net Margin (NM) and Operating Margin (OM) from fiscal year (FY) 2008-2014. They found that strong patient experience led to increased results in all three categories. In a similar vein AdvoCor has decided to create an updated, large scale, and transparent study of these metrics to emphasize the importance of strong patient experience to hospital financial health.
Our findings conclude that patient experience, as measured by the percentage of patients that rated a hospital a nine or ten on a ten point scale (Rate Hospital Top-Box) on the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey, has a significant impact on each financial metric studied. It was found that there was an approximate 1.04 point increase in ROA, 1.13 point increase in NM and 1.13 point increase in OM on average for each ten percentage point increase in Rate Hospital Top-Box. From our findings, there is strong evidence of the positive financial impact of a strong patient experience highlighting the importance of improving and maintaining patient perceptions of care.
Data
The main financial dataset containing information on various hospital financial metrics and characteristics was acquired through publicly available Hospital Provider Cost Reports for FY 2014-2018 from the Centers for Medicare & Medicaid Services (CMS). This dataset contained over 31,000 individual data points for 6,416 unique CMS Provider IDs. From this financial data, various metrics were calculated and extracted such as ROA, NM, OM, percentage of gross revenue from Medicare/Medicaid, location data, urban/rural designation, provider and control types, number of total beds for provider and all sub-providers, and average number of FTE employees. The equations used to calculate ROA, NM, and OM are shown below:
1. ROA = (Net Income)/(Total Assets)
2. Net Sales = Gross Revenue - Discounts & Allowances
3. NM = (Net Income)/(Net Sales) 4. OM = (Net Sales - Total Operating Expenses)/(Net Sales)
An additional dataset of various publicly provided HCAHPS scores from 2013-2018 from 4,993 unique CMS Provider IDs was acquired to create patient experience measures. From this dataset, all scores for Rate Hospital Top-Box were selected and matched to the financial dataset by CMS ID, measure start year/month, and measure end year/month; discarding all data which either did not have associated financial metrics or HCAHPS scores. A further two datasets were collected and matched to the existing data, one that contained publicly reported HVBP adjustment factors tied to CMS IDs to measure the effect of stronger or weaker HVBP incentives, and one that contained Hospital Referral Regions (HRR) representing 306 unique regional health care markets for tertiary medical care from the Dartmouth Atlas.
As a secondary objective, a final dataset of 199 unique CMS IDs that utilized the AdvoCor rounding platform for the period of 2016-2019 matched to the final dataset by CMS ID, start year and end year was collected utilizing internal AdvoCor platform data to explore the impact of patient experience rounding. For the AdvoCor dataset start year was defined as the first year in which a site had performed patient experience rounding on more than 5% of their patients and the end year was defined as the last year a site had performed patient experience rounding on more than 5% of their patients. For sites which had duplicate CMS IDs due to joint reporting, their rounding metrics were calculated using a weighted mean by CMS ID. In the final dataset CMS ID data points where the year studied was less than or equal to the end year and greater than or equal to the start year, were considered as utilizing AdvoCor at the time. After all data was matched and collected the final data set contained 4,119 unique Provider IDs with 15,348 total data points and 2,829 unique Provider IDs matched to HVBP data with 10,442 total data points. The HVBP modifier ranged from 0.982513 to 1.040313 so it was transformed by multiplying by 100 and then subtracting 100 from the new value. This allowed HydraCor to change the scale and had negative values representing a loss in revenue from HVBP and positive values representing a gain. Due to the large amount of data available, three distinct datasets were created for each financial metric studied where the bottom and top 2.5% scores were removed creating a 95% interval to diminish the potential for outlier values to affect the analysis.
Analysis and Discussion:
Exploratory Analysis
To explore the link between patient experience and hospital profitability the total dataset was divided into 3 distinct groups based on their Rate Hospital Top-Box performance: “Excellent”, “Moderate”, and “Low” performers. To create these groups, a percentile ranking of Rate Hospital Top-Box for each FY was calculated. Those sites that had a percentile rank >= 66 for the given FY were assigned the “Excellent” label, those with a rank < 66 but >= 33 were assigned to the “Moderate” group, and the rest were assigned to the “Low” group. Unlike the Deloitte study classified hospitals with above the median ‘Middle Box’ scores (the percentage of respondents who gave the hospital a rating of 7 or 8 out of 10) as “Moderate”, the groupings in this study were chosen to ensure similar sample sizes for each grouping1. The data showed that hospitals in the excellent group for a given year averaged an approximately 6.14 ROA, 6.14 NM, and 5.63 OM, while hospitals in the low performance group for a given year averaged a 3.30 ROA, 2.95 NM, and 2.50 OM.
Figure 2. Average Financial Metrics by Score Label and FY
The moderate group averaged a 4.65 ROA, 4.83 NM, and 4.49 OM sitting firmly between the excellent and low performing groups in financial performance. While there is some expected variation, this pattern continues to hold for each individual year studied, implying a connection between patient experience and financial performance (Figure 2). Though profitability metrics and patient experience scores vary across different hospital characteristics such as Urban vs Rural designation, HHR, Ownership Type, and Teaching Status, the order of profitability by patient experience grouping generally holds. For instance, Governmental hospitals tend to have significantly lower average OM (Average Operating Margin = 2.10, n = 841 unique CMS IDs) than Voluntary Non-Profit Hospitals (Average Operating Margin = 4.52, n = 2515 unique CMS IDs) or Proprietary Hospitals (Average Operating Margin = 4.89, n = 715 unique CMS IDs), however within each control type the stronger patient experience groupings still tend to have greater OMs (Figure 3.) As shown in Figure 3., there are variations in how much each patient experience grouping differentiates in OM from others and the overall average values in OM between different types of control, yet only a single year in each type has a different ordering between patient experience groupings.
Figure 3. Average Net Margin by Score Label, FY, and Control Type
Regression Analysis:
The previous data from exploratory analysis suggested a link between patient experience and financial profitability, so regression models were employed to further study and quantify this link. To study each of the three financial profitability metrics, three main models utilizing ROA, Net Margin, and Operating Margin as outcome variables were created. Each model was then fitted using the Rate Hospital TB and along with other characteristic variables to control for factors that may influence hospital profitability. The factors used to control for possible influences can be broken down into time characteristics, hospital characteristics, and location characteristics. Since each financial metric was measured based on the fiscal year, the fiscal year was utilized as a factor to control for changes in baseline hospital profitability over time. Hospital characteristics utilized included the percentage of revenue coming from Medicare, the HVBP adjustment factor the hospital received for a given FY, the type of control, the provider type, the number of beds at the hospital and all sub providers. Location characteristics included Rural/Urban markers and HRR number. Finally a variable to mark whether the site utilized a patient experience rounding service (AdvoCor) over the time period was included in order to study whether such a service had an effect on financial metrics. To account for possible differences between hospitals that could affect financial outcomes that may not have been accounted for or are potentially unobservable (e.g. hospital culture, leadership quality, management practices, etc.) this analysis utilized a mixed-effect regression model with each CMS ID serving as the random effect. A mixed effects model includes both fixed effects, which are model components used to define systematic relationships, and random effects, which account for variability among subjects around the systematic relationships captured by the fixed effects. Mixed effects models are a common method of analyzing longitudinal data with large inter-individual differences by accounting for variation in the fixed effects along different values of the random effect. Utilizing the CMS ID as the random effect allows the removal of the effect of underlying differences between hospitals that may have not been accounted for from the analysis on the fixed effects. After creating the full model, an Patient Experience and Financial Performance 5 Figure 3. Average Net Margin by Score Label, FY, and Control Type ANOVA was utilized to determine which effects are truly significant to our model and remove the insignificant ones to create a final model. The final models predicting ROA, NM, and OP each found Rate Hospital TB to be a strongly significant predictor, achieving a total R2 of approximately 0.72, 0.72, and 0.80 respectively. For each model the Rate Hospital TB score was found to be extremely significant (p < 0.000001), with regression coefficients of 0.10370, 0.11250 and 0.11254 for the ROA, NM, and OM models respectively. In fact, for all models Rate Hospital TB was the most important factor in explaining model variance as measured by Mean Squares (Figure 4.).
Figure 4. Top Variables per Model as Measured by Mean Squares
As measured by the model coefficients, even accounting for extraneous factors and variation between hospitals, there is a 10 percentage point increase in Rate Hosp TB results in an average increase of 1.04 points in ROA, 1.13 points in NM, and 1.13 points in OM. As the overall mean values for ROA, NM, and OM were 4.62, 4.55, and 4.11 respectively, this increase could be incredibly significant resulting in a positive increase of approximately 22.46% in ROA, 24.72% in NM and 27.37% in OM for the ‘average’ hospital. While HVBP did have an impact on financial metrics, it accounted for less than half of the variation that Rate Hospital TB in all models and only 2.75% of variation in the OM model, only approaching significance being above the 0.05 cutoff in p-value (0.05 < p < 0.1). HVBP, as measured by mean squares, never reaches a proportion over 15% of total mean squares in any model. This implies that HVBP is a much less important factor than the overall patient experience satisfaction and not enough to explain the increase in profitability due to increased patient experience scores alone.
Figure 5. Financial Metrics by AdvoCor Utilization and Year
Patient Experience Rounding Analysis:
As a secondary objective, AdvoCor wanted to explore whether hospitals that utilize a patient experience rounding tool experienced improved financial outcomes. To this end, hospitals which actively utilized the AdvoCor patient experience rounding platform throughout the period for each FY were marked. While the dataset for this analysis was limited, consisting of only 121 unique CMS IDs and 190 data points throughout the time period of FY 2016-2018, the results were intriguing enough to merit further study. When examining the data by FY it can be seen that sites that utilized the AdvoCor platform had improved financial performance for all metrics as compared to their counterparts that did not (Figure 5.). As evidenced by the data hospitals that utilized the AdvoCor platform had an overall average ROA of approximately 7.83, NM of 7.06 and OM of 6.00 while sites that did not had an average ROA of 4.56, NM of 4.47 and OM of 4.00.
Conclusion:
Through exploratory and quantifiable regression analysis AdvoCor found strong evidence that patient experience scores do affect hospital financial outcomes, even accounting for direct value gained through CMS HVBP incentives, confirming the conclusions reached by Deloitte1. Further research should continue to analyze additional factors, such as clinical outcomes, along with what specific factors may affect patient loyalty and experience, such as communication with staff or satisfaction with hospital food quality. This study provides strong evidence for the importance of patient experience on hospital financial performance that should be used to drive further research in the realm of patient experience and patient consumerism.
References:
1. Betts, D., Balan-Cohen, A., Shukla, M., & Kumar, N. (2016). The Value of Patient Experience. Deloitte Center for Health Solutions, 1-22. https://www2.deloitte.com/content/dam/Deloitte/us/Documents/life-sciences-health-care/us-dchs-the-value-of-patient-experience.pdf
2. The Centers for Medicare & Medicaid Services. (2021, December 1). Hospital Value-Based Purchasing Program. CMS. Retrieved August 3, 2022, from https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Hospital-Value-Based-Purchasing-
3. The Centers for Medicare & Medicaid Services. (2022, June 15). Hospital Provider Cost Report - Centers for Medicare & Medicaid Services Data. CMS Data. Retrieved August 3, 2022, from https://data.cms.gov/provider-compliance/cost-report/hospital-provider-cost-report
4. The Centers for Medicare & Medicaid Services Data. (2015-2019). Survey of patients’ experiences (HCAHPS) | Provider Data Catalog. CMS Data. Retrieved August 3, 2022, from https://data.cms.gov/provider-data/topics/hospitals/hcahps
5. Wolf, J. A., & CPXP, P. (2018). Consumer Perspectives on Patient Experience 2018. Beryl Institute. Accessed August 1, 2022.
6. Dartmouth Atlas Project. (2022). Supplemental Data - Dartmouth Atlas DATA. - Dartmouth Atlas DATA. Retrieved August 3, 2022, from https://data.dartmouthatlas.org/supplemental/#crosswalks
Your contact information is in safe hands.
We will only contact to communicate with your organization. Your information will not be distributed or made public. We promise not to spam you!
Share This News, Choose Your Platform!
