February 19, 2019
How to Build Inpatient Satisfaction with HCAHPS Data: A Blueprint
Share This Post
A landmark event occurred in the realm of hospital care in March of 2008. At that time the U.S. Department of Health and Human Services launched its Hospital Compare web site where inpatient satisfaction survey data were reported on a voluntary basis by 2,600 hospitals. This large-scale national survey was called HCAHPS, the Hospital Consumer Assessment of Healthcare Providers and Systems. It was launched on the dual premise that: consumers seek evaluations from peers about inpatient hospital experiences; and the quality of hospital care improves with comparative inpatient satisfaction data. The concept of HCAHPS was embraced, and even became tied to the Annual Payment Update (APU) for Inpatient Prospective Payment System (IPPS) hospitals. This linkage spurred participation among even more U.S. hospitals.
A simple but comprehensive evaluation tool, the HCAHPS instrument has 16 items that assess the inpatient care around thematic constructs or domains that are vital to the patient hospital experience:
With the exception of the two Discharge questions, questions are rated on a four-point rating scale of Always (4) Usually (3) Sometimes (2) or Never (1). The two Discharge questions are on a dichotomous rating scale, Yes (1) or No (2). In addition, there are two global-focused items. The first is an eleven-point rating scale (0-10) to measure Overall Rating of the hospital, and the second, a Willingness to Recommend the hospital.
On the Hospital Compare website, individual data for each hospital are presented along with aggregated data for hospitals by state and for all hospitals in the entire U.S. It is easy to interpret the data when the goal is to compare one hospital against another hospital in the same zipcode, city, county, state, or the U.S. Yet, the HCAHPS survey offers hospitals an additional opportunity for those hospital administrators and patient care providers willing to roll up their sleeves.
The HCAHPS data at the local level can serve as a lever for improving inpatient care, for promoting patient care staff development, for boosting employee morale, and for tracking structural innovations and their subsequent impact. However, this requires the need to analyze HCAHPS data for each of the 16 evaluation items. Such a practice focuses assessment directly on aspects of patient satisfaction. No doubt, this type of work is detailed and time-consuming. But neglecting the analysis of each item masks vital insights that can impact your hospital inpatient care. Here is an easy three step blueprint for how you do it.
The first step is to develop a simple chart by individual HCAHPS items with each of the four responses and their respective percentages. This should be done for each quarter of the year for five quarters, yielding a full year of data for comparison. At a glance, the most current quarter can be compared with the previous quarter or quarters, as well as the quarter a year ago. Here is some hypothetical information you can readily infer from charting.
Charting gives each item stem the attention it deserves in the big picture of inpatient improvement. Since percentages are easy to interpret, there is little confusion or misinterpretation. Essentially, the chart provides trend analysis about whether an area of patient care is improving, declining or unchanged. Also, the trend data can track an outlier (like the 3rd Quarter ) to determine if it is a real problem or just a blip in the radar. For hospitals short on resources, this determination prevents unnecessary time, money and manpower from being deployed for a problem that really does not exist.
Some hospitals have relegated authentic inpatient improvement to capturing the gold ring. The sifting of precious data is traded for playing the numbers game of "our score is better than theirs" or "our scores increased from last week" even though it was one-tenth of a percentage point. That is why taking the time to analyze in detail what the data are saying about your hospital is vital.
The second step is drilling down your data. The raw data set from HCAHPS is a gold mine. One of the most important questions it can answer for you is what are the major predictors of inpatient satisfaction at your hospital. Overall Patient Satisfaction is the HCAHPS global rating where patients rate your hospital between 0 and 10 to indicate their overall satisfaction. To determine what the major drivers are, you need to determine which two or three of the 16 HCAHPS items correlate the highest with the Overall Patient Satisfaction rating. The statistic is a simple Pearson Product Moment Correlation coefficient.
The HCAHPS also allows the data to be drilled down by patient segments. Are the drivers of Overall patient satisfaction the same for males as well as for females? Are they the same for the three specialty groups of inpatients: Medical, Surgical and Obstetrics? These findings have direct implications for inpatient care providers on the floors. The different models can be expanded to ethnic groups, race, admission status (ED/non-ED), age category and others to see if there are differences that have actionable patient care delivery implications.
The third step is to use the data to test for the impact of your interventions. Your HCAHPS data set can be used to test whether a response to an inpatient problem (that you may have identified in Step 1 of this blueprint), worked. Essentially, the question is whether human, physical and fiscal resources were worth the investment to solve the problem. For example, most hospitals are challenged by the inpatients' call for more Quiet at Night around their Room. If a hospital decided to implement a strategic initiative to mitigate nighttime noise on patient floors, the HCAHPS data would be an excellent source to quantify impact.
This can be accomplished quite simply. With the data that are in your HCAHPS database, a simple statistical procedure can be undertaken to see if the pretest data significantly increased at posttesting. This might mean that the aggregate data from January through June (mean) is compared to the aggregate data from July though December (mean). A correlated t-test would be an appropriate way to measure significant gains over time. There are multivariate procedures as well that are more complex and have more precision but this statistic is simple and would serve to document a concrete gain from before your intervention and after.
These are three steps that you can implement to make your HCHAPS data a workhorse. It is a reservoir that is rich in the information needed to make patient satisfaction authentic and that benefits the patient, the family, your staff and the community your serve. And best of all it is free for the taking.
Share This Post