Quality indicators
From HMCwiki
Contents |
Quality indicators take on a new significance
In today’s healthcare environment, the practice of measuring hospital care-giving performance along a set of “quality indicators” has become an important part of hospitals’ business and is set to become even more crucial in the near future. Quality indicators allow hospitals to track their performance in a number of important parameters to see whether the care they provide to patients is of the quality that it should be. By comparing their quality measurements to those of their peers as well as to national averages, hospitals can identify points in which the care they provide is problematic and must be improved. Increasingly, quality measures such as Agency for Healthcare Research and Quality (AHRQ) indicator measurements will become public information allowing patients to decide where they would like to be treated. Furthermore, CMS will soon put into action a plan to base the provision of Medicare and Medicaid reimbursements on quality measures. In this “pay-for-performance” plan, hospitals providing better quality care will receive more money from Medicaid and Medicare.
The current situation necessitates that hospitals have a good understanding of the AHRQ indicators, especially, as well as other indicators as they emerge. The following article is a look at what the AHRQ indicators are, what they mean, and how to go about improving them.
The AHRQ Indicators
The AHRQ defines its indicators as “a set of quality indicators organized into four "modules," each of which measures quality associated with processes of care that occurred in an outpatient or an inpatient setting."[1] The four modules measured by the AHRQ indicators are:
- Prevention quality indicators (PQIs) deal with admissions that could be prevented with better outpatient care and/or access – for the purpose of this article we are looking at inpatient indicators.
- Inpatient quality indicators (IQIs) look at mortality rates of certain medical conditions and procedures as well as utilization rates of procedures that have been identified as problematic.
- Patient safety indicators (PSIs) consider rates of avoidable complications caused by poor care.
- Pediatric quality indicators (PDIs) examine the quality of care administered to children. From the perspective of most hospitals, the most crucial modules are the IQIs and PSIs, as they represent inpatient quality that is not specialized to children.
What quality indicators mean to hospitals
These quality indicators offer a challenge to hospitals around the country: how should these indicators be interpreted and how can they be used to help hospitals improve their quality? In examining the quality measures more closely we have come up with some insights on how they should be viewed by hospitals. First it must be noted that the measures are all different and cannot all be interpreted in the same way. We have broken down the AHRQ indicators into two major categories:
- Deficiency Indicators: Some measures deal with high volume procedures in which one or two of the negative outcomes that the indicator is measuring will not have a large impact on a hospital’s reported rate. These we call deficiency indicators as a poor quality measurement suggests a definite deficiency in performance.
- Episodic Indicators: On the other hand, some indicators measure generally low-volume events in which a single mortality or extra utilization can dramatically increase the indicator rate. An example of such an indicator is Incidental Appendectomy in the Elderly. Similarly, there are high volume procedures in which a negative outcome is rare and therefore one or two negative outcomes can make a hospital’s rate relatively poor compared to its peers, as in Post-Operative Hip Fracture. We call these cases episodic indicators because a single episode can heavily influence the quality measure. Episodic indicators pose a problem – a single case with a negative outcome seems to have a certain randomness to it that can dramatically alter a hospital’s quality rating for a given time period, pointing to low quality care until the next round of indicators is posted.
Analysis of a sample set of hospital data
In order to evaluate differences in quality indicators with more depth, we used a sample set of hospital quality data and created a histogram for each indicator to see how the hospitals’ rates of each indicator are distributed. To make the distributions of all the indicators comparable to each other, since they are all on different scales, we graphed the percentage each hospitals’ rate deviated from the mean. These distribution graphs revealed a few major trends in shape related to the two categories of indicators earlier mentioned.
- The first shape is one in which the predominant value is zero (-100% deviation from the mean), and above that the distribution is fairly widely dispersed, with some indicators varying more than others. This type of graph corresponds with the episodic type of indicator. Here is an example of such a graph:
- The other major shape is a fairly normal distribution, with more of a bell shaped curve around the mean (0% deviation). These correspond with the deficiency type of indicators in which one or two negative outcomes will not drastically change a hospital’s rate. Here is example of this sort of graph:
We then ranked each hospital’s performance in each quality indicator by percentile to see if there is a smooth gradient between well-performing hospitals and poorly performing hospitals. To depict this clearly, we found the average indicator percentile for each hospital, and then graphed this against the hospital’s “rank” compared to other hospitals in the group.
The graph shows an s-curve with a steady increase between 45% and 65%, where the majority of the hospitals fall. At the extremes we see the trend become steeper as the best performers are far better than the norm and the worst performers are far worse than the norm.
We also broke down the proportion of indicators of each hospital in each quartile, and graphed these to see if there seemed to be a functional relationship between average percentile and the number of indicators in each quartile.
These graphs provide us some insight on what lays behind a hospital’s average percentile ranking:
- There is a direct relationship between the number of indicators in the first quartile and a hospital’s average percentile ranking – as the average percentile increases, the number of indicators in the first quartile increases
- There is an inverse relationship between the number of indicators in the fourth quartile and a hospital’s average percentile ranking – as the number of indicators in the fourth quadrant increases, the average percentile declines.
- The number of indicators in the second and third quartiles do not seem to predict hospitals’ average ranking, and even seem to be somewhat random.
Episodic vs. deficiency indicators
The split between the episodic indicators and deficiency indicators is an interesting aspect of quality that deserves further investigation. The thought is that maybe the difference between the two types contributes to the difference between a “good” and a “bad” hospital. The following two graphs show the shape of the distribution of average percentiles of episodic and deficiency indicators, respectively. They also show the relative proportion of each hospital’s indicators that are in the first, second, and third quartile.
From these graphs you can see:
- It is easier to perform well in the episodic indicators…because you can be perfect (at 0% rate)
- Episodic indicators’ performance range is larger (the tails are more extreme). This is because most are indicators with small volume – small changes in volumes of outcomes makes big difference in rates
- Deficiency indicators have a tighter performance range
Improving performance
Quality improvement should not be limited to making your quality indicators look good, but with quality measurements soon becoming public knowledge they will be important to a hospital's reputation. Using all of these previous observations as guidelines, we have attempted to come up with some insight as to the best ways to improve a hospital’s performance as measured by these quality indicators:
- Look to improve your episodic indicators – they are the easiest to get into the first quartile as it is entirely possible to get rates in these indicators down to 0%
- The worst-performance episodic indicators should be the first to be improved as they have the most potential for improvement
- The goal is to increase the number of first quartile indicators and decrease the number of fourth quartile indicators, as these are systematically related to overall performance while the second and third quartiles are not
- Deficiency indicators are usually high volume, more difficult to have perfect performance – improving is more of a process as minor reductions in negative outcomes will not result in significant changes in measured rates of quality.
Summary
Competition between hospitals in the United States is being redefined by quality outcomes. From the patient’s perspective, publicly available quality outcome measures allow patients to compare one hospital’s value versus another; that is, patients can now shop for the best care. So to remain competitive, hospitals need to track their performance in a number of important parameters to see whether the care they provide to patients is of competitive value. By comparing their quality measurements to those of their peers, as well as to national averages, hospitals can identify points in which the care they provide is problematic and must be improved. It apears that "episodic indicators" hold the greatest near-term potential.
References and resources
- ^ Agency for Healthcare Research and Quality: General Questions about the AHRQ QIs (accessed 16 August 2006).
 |
If you would like to provide feedback or discuss the content of this article, please visit the discussion page. To be automatically notified by email of any changes or additions to the article or discussions about this article, please click here. |
