10 Things We Know About HCAHPS and Value Based Purchasing
(1) The Clock Has Started
Although most hospitals have been participating in the HCAHPS program since late 2006, it has not been clear until now what historic data, if any, would actually be used in the calculations that will determine hospital reimbursement as part of the Value Based Purchasing (VBP) program. In January of 2011, CMS released a document for public comment that provided detailed information about VBP. In this document, CMS suggests using HCAHPS data from the following dates to calculate a hospital’s reimbursement in the new pay for performance environment.
HCAHPS results for patients discharged from July 1, 2009 – March 31, 2010 will be used as the baseline year from which to gauge improvement over time.
HCAHPS results for patients discharged from July 1, 2011 – March 31, 2012 will be used as the performance period from which to gauge a hospital’s current performance.
CMS will use data from these two 9-month periods to calculate a hospital’s reimbursement for FY2013. CMS has yet to provide the dates that will be used to calculate reimbursement in subsequent years but has indicated that they will use 12-month rather than 9-month periods of time beginning with FY2014.
(2) Hospitals Nationally Are Achieving High Scores
The results to the HCAHPS survey nationally show a negative skew—meaning that scores are grouped closely together at the positive end of the distribution. In lay terms, this means that patients responding to the HCAHPS survey are much more likely to give a “top box” response (i.e., the most positive response possible) to most questions than any other response. The table below shows the percent of patients nationally who gave a “top box” response in the December 2010 public release. For example, this table indicates that 71% of patients said that the area around their room and bathroom was ALWAYS kept clean.
Based on results from 3,798 U.S. hospitals
It is important to realize that the scores above represent roughly the 50th percentile for each HCAHPS theme. Hospitals will need to be consistently performing at this level or higher on each survey theme in order to at least break even financially under the Value Based Purchasing program.
(3) Hospitals Nationally Are Improving
The CMS national database has improved in each of the past three years. The table below shows a 1-point overall improvement in both 2008 and 2009 and a 2-point overall improvement in 2010.
The implication of this improvement is that hospitals that are not improving are actually losing ground. The whole boat is rising and failure to at least keep pace means that your hospital’s ranking in the national database is declining. We have seen numerous hospitals whose 10-theme aggregate score on HCAHPS has remained stable over the past 3 years, but their percentile score has steadily eroded because the performance of hospitals nationally has improved.
(4) Not All HCAHPS Themes Will Count Toward CMS Reimbursement
For the past several years, we have all been using an HCAHPS survey that measures patient perceptions of care along 10 dimensions, ranging from nurse communication to quietness. In calculating hospital scores for reimbursement purposes under the Value Based Purchasing program, CMS intends to use only 8 dimensions. CMS plans to omit the “would recommend” question entirely and combine the “cleanliness” and “quietness” dimensions into one theme.
Source: CMS Report entitled “Medicare Program; Hospital Inpatient Value-Based Purchasing Program” released for public comment on January 7, 2011
(5) It’s Important to Know Who Says “Always” and Who Doesn’t
The HCAHPS survey clearly shows that most patients are having a positive experience of care at their local hospital. This is evidenced by the fact that the majority of patients are giving either the highest or 2nd highest possible rating to each survey question. Since most questions in the HCAHPS survey are on an always, usually, sometimes, or never scale, this means that most patients are giving a rating of “always” or “usually” to the various aspects of their care.
Given this, the focus of HCAHPS for most hospitals is to figure out how to make more patients give a rating of “always” when they are currently inclined to give a rating of “usually.” At HealthStream, we encourage our hospitals to dig into their survey results to determine which patients are giving the “usually” ratings vs. those who are rating “always.” We think this profile information can be very helpful in making improvements in your HCAHPS scores. For example, in one hospital we discovered that it was the OB patients staying less than 24 hours who were least likely to give an “always” rating to cleanliness. At another, we found that it was the patients on the surgical services unit who were least likely to give an “always” rating to staff responsiveness in answering the call button. These drill downs into the data can be invaluable in determining where to start your improvement efforts.
(6) Small Improvements Can Make A Big Difference
We find very few hospital leaders who are satisfied with their HCAHPS performance. Even our highest performers are worried that their scores are “not good enough,” given the implications of both public transparency and reimbursement risk. We also find that many of the lower performers—those with a high number of scores falling below the 50th percentile—are discouraged and worried they cannot improve “enough to catch up.” For all of you who are serious about wanting to improve your scores, there is some good news.
In the national HCAHPS database, the survey results are tightly distributed, meaning that the difference between the highest scoring hospitals and the lowest scoring hospitals in the country are not that great. In fact, we have discovered that, depending on the HCAHPS question, it is only 17 – 34 points that separate the top performers from the lowest performers. For example, on the highly important “nursing communication” theme, there is only 20 points separating the top and bottom performers.
Source: 4th Quarter 2010 CMS Public Reporting Period based on patients discharged April 2009 – March 2010. Based on results from 3,798 U.S. hospitals.
The good news in all this is that small improvements (often only a couple of points) can make a big difference in your percentile score. In one recent example, one of our clients only had to improve its “discharge information” score by one point—from 41% to 42%--to improve its percentile score from the 39th to the 50th percentile nationally.
(7) We Know What a Top Performer Looks Like
The chart below shows the Top Box Scores from the HCAHPS survey that equate to being at the 75th and 95th Percentiles nationally. For example, a Top Box Score of 85 (or 85% of patients saying “always”) is the score that was equal to the 95th Percentile in the 4th Quarter Public Release of data by CMS. Similarly, a Top Box Score of 70 was equal to the 75th Percentile nationally. Hospitals performing in this range on the HCAHPS survey are among the best hospitals in the country, from the patient’s point of view.
Source: 4th Quarter 2010 CMS Public Reporting Period based on patients discharged April 2009 – March 2010. Based on results from 3,798 U.S. hospitals.
(8) CMS Values High and Consistent Performance
In calculating a VBP score for each hospital, CMS places a high premium on high and consistent performance. Hospitals will be able to earn up to 100 points based on their HCAHPS survey results—up to 10 points on each of the 8 HCAHPS themes (80 points total) and an additional 20 points if all 8 themes are at or above the 50th percentile. Given this, one of the first objectives for every hospital should be to make sure they are at or above the 50th percentile on all 8 themes so that they are assured of receiving all 20 “consistency” points from CMS.
To earn the remaining 80 points, CMS will look at each theme individually and assign it either “attainment” or “improvement” points—whichever is greater. For attainment points, hospitals can earn up to 10 points, depending on where there score falls between the 50th percentile and the percentile at the mean of the top decile (typically the 95th percentile). Hospitals whose theme score is below the 50th percentile earn no attainment points. Hospitals will earn 1 attainment point for being at the 50th percentile and up to 10 points for being at the 95th percentile.
CMS will also compare a hospital’s current performance (based on patients discharged from July 1, 2011 – March 31, 2012) to their baseline performance two years ago (based on patients discharged from July 1, 2009 – March 31, 2010). For improvement points, hospitals can earn up to 9 points, depending on how much their score has improved between their baseline period and the current performance period. Hospitals who have not shown any improvement or who have lost ground would receive no improvement points. Hospitals will earn between 1 and 9 improvement points based on the amount of improvement toward the 95th percentile. Hospitals who have improved all the way to the 95th percentile will earn the maximum of the 9 possible points. CMS will then assign an attainment or an improvement score to every HCAHPS theme, whichever score is greater.
(9) Hospitals with VBP Scores Above the 50th Percentile Are Likely to Make Money
For FY2013, CMS has proposed to hold back 1% of hospital payments (representing approx. $850m). Hospitals can then “earn back” their payment at an incentive rate determined by the percentile ranking of their total VBP score. The total VBP score will be based on a hospital’s scores from the HCAHPS survey (which will account for 30% of the VBP score) as well as its scores from the 17 clinical process of care measures that have been identified for inclusion in the Value Based Purchasing program (which will account for 70% of the VBP score).
CMS has indicated that hospitals scoring at or above the 50th percentile in terms of their overall VBP score are likely to break even or benefit from the Value Based Purchasing program. These hospitals will be positively rewarded based on the strength of their performance relative to other hospitals.
CMS Estimated Hospital Incentive Rates
Source: CMS Report entitled “Medicare Program; Hospital Inpatient Value-Based Purchasing Program” released for public comment on January 7, 2011
(10) Hospitals with VBP Scores Below the 50th Percentile Are Likely to Lose Money
Conversely, hospitals that have a total VBP score that is below the 50th percentile will experience a drop in their CMS reimbursement rate. It is the intent of CMS to make Value Based Purchasing a budget neutral government program—i.e., CMS will not pay out more money to hospitals than it would normally pay. Rather, the money will be redistributed from lower performers to those who achieve higher scores. As the CMS chart below indicates, lower performing hospitals are likely to see their incentive payment from CMS cut in half.
Source: CMS Report entitled “Medicare Program; Hospital Inpatient Value-Based Purchasing Program” released for public comment on January 7, 2011