How to develop guidelines that work
Conduct a cost analysis
The key to successfully implementing clinical guidelines lies in truly understanding the costs and the benefits, experienced guideline developers tell Hospital Peer Review.
Janet B. Mitchell, PhD, uses a guideline for carotid endarterectomy as an example of how to conduct a cost analysis, and follow through with your analysis once the guideline has been implemented.1 Mitchell is president of the Center for Health Economics Research in Waltham, MA, a private, nonprofit company that does public-policy research for federal and state governments and private foundations.
Some clinicians use noninvasive diagnostic techniques (Doppler and duplex scans, for example) before the carotid endarterectomy, rather than cerebral angioplasty (CA), which is costly and presents higher risks. Some patients may eventually receive CA after the noninvasive tests prove inconclusive.
Using Medicare Part B claims data, Mitchell describes the current national patterns of diagnostic testing in the 90 days prior to carotid endarterectomy. In that period, only 10.3% of patients receive noninvasive testing, while 83.7% receive CA alone or CA in addition to noninvasive tests. Mitchell calculates the impact of a guideline recommending surgical candidates receive only noninvasive testing, as follows:
* Based on current practice patterns, preoperative diagnostic testing averages $500 per patient, or $24.9 million annually (based on Medicare volume of 49,700 carotid endarterectomies per year).
* Since cerebral angiography averages $418 per case in physician fees, eliminating this test would save Medicare about $17.4 million annually.
* Some patients either received no cerebrovascular tests prior to surgery or only received angiography. Providing noninvasive tests to these patients would cost $1.6 million, or $201 per patient.
* The net impact would be a savings of almost $16 million.
The Center for Health Economics Research does all the cost analyses for clinical practice guidelines published by the federal Agency for Health Care Policy and Research. "We do that on a national basis using national data," Mitchell says. "But you could do it for any sub-area."
One hindrance, Mitchell says, is obtaining comprehensive data for a smaller area. "The kind of data we typically use is claims data," she says. "So, you'd need a data source that would include all the claims in an area."
Use Medicare and Medicaid data, as well as your own internal data or local information from private payers. Use expected results from implementing the practice guideline as your point of comparison, Mitchell says.
You can simulate the expected cost impact of the guidelines using baseline data on current practice patterns, as in the carotid endarterectomy example; then evaluate the actual impact of implementation, she says.
(See Mitchell's list of key cost analysis data, below.)
Evaluate your results
Evaluating the effectiveness of a guideline requires measuring the extent to which providers adopt it, how much they change their practice habits to reflect criteria spelled out in the guideline, and what other unexpected reactions result.
"You need to know whether providers are adopting the guidelines," Mitchell says. "You also have to know whether patients are going along with it, as well. There are some guidelines where patient compliance is an important part."
For example, the success of using a guideline on cardiac rehabilitation depends on the patient's ability to stick with the program. "You would have to follow them prospectively after they were discharged from the hospital," Mitchell says.
If a hospital cannot or will not find out how many people are using the new guideline, it would not be able to determine whether the guideline has any effect at all. "You could see whether you had any more claims for rehabilitation, but the guidelines include services that are not covered by insurance," Mitchell says. "You would not be able to tell if people are adhering to the guidelines from the claims data."
Make sure you survey physicians on their use of the guidelines. In fact, Mitchell has noticed that some medical procedures and guidelines do not get used because providers are not aware of them. "Make sure it gets a lot of publicity," she says. "Many physicians are not using the most up-to-date treatments, presumably because they do not know about them. Dissemination is critical."
Reference
1. Mitchell J. Cost analysis of clinical guidelines: Which data to use and how to find them. Presented at the Cost Analysis Methodology for Clinical Practice Guidelines conference, Baltimore; November 1993. *
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content