Do it yourself: MD creates his own practice guideline

Protocol saves $1,045 to $1,425 per patient

Maybe you have an HMO anxious to push you to accept a certain clinical guideline. Or maybe you just have a hunch that if you looked at a high-volume procedure in your practice, you might find ways of performing it more cost-effectively with equal or better clinical results.

Whatever your motivation, physician-created clinical guidelines - or those created with the assistance of an insurer, rather than by the dictate of an insurer - can vastly lower your risk in capitation contracts and give you confidence that you're doing the right thing, says Neil S. Baum, MD, a urologist in private practice in New Orleans. Baum and a professor colleague, Dan Murtagh, MD, at the Medical College of Ohio in Toledo, followed these steps in assessing a high-volume procedure in Baum's five- physician practice:

1. Identify an area where there is significant variation in practice patterns among physicians and a high-volume service within your practice. The area Baum and Murtagh examined was the metastatic workup for prostate cancer. The evaluation included the use of acid phosphatase measurement, a bone scan, and/or a CT scan of the pelvis and abdomen. The physicians determined there was ample practice variation and volume to make this procedure suitable for further research.

Five of the 20 highest-volume Medicare services provided in physicians' offices are office visit codes, for which HCFA has released practice guidelines. (See chart showing the 20 highest-volume services, p. 118.) Due to the storm of controversy over HCFA's guidelines, in which some physicians charged they were "accounting" guidelines instead of clinical guidelines, HCFA postponed the guidelines' implementation and is revisiting the content.

2. Research the literature. Baum and Murtagh researched what experts describe as scientifically acceptable approaches to the evaluation of patients with prostate cancer. In reading the peer-reviewed literature, Baum and Murtagh found that patients with prostate-specific antigen (PSA) levels lower than 10 and Gleason scores lower than seven did not require additional expensive testing such as bone scans and CT scans before proceeding with radical proctectomy or radiation therapy.

3. Make recommendations for a new protocol, based on current research findings, to physician colleagues and allow freedom of choice. It will be difficult to change a protocol without physician buy-in. To achieve that, make sure colleagues know where the recommendations came from. Then, stipulate that the protocol is completely voluntary and that physicians can deviate from it at any time. In Baum's practice, his colleagues also agreed that Baum and Murtagh would track deviations from protocol to determine the results.

4. Lay out the specifics of the proposed protocol or guideline. In this example, the protocol stratified patients into four groups: group one had 30 patients with PSAs of less than 4ng/ml, regardless of their Gleason scores; group two had 65 patients with PSAs between four and 10 and Gleason scores of less than seven; group three had 30 patients with PSAs between four and 10 and Gleason scores of less than seven; and group four consisted of 60 patients with PSAs under 10 without consideration of their Gleason scores.

Patients in groups one and two were to have only an acid phosphatase test; group three and four were to have an acid phosphatase test and a bone scan. Other tests could be ordered outside the protocol if the physician felt it was in the patient's best clinical interest.

5. Collect data on a form that serves as an order sheet and that contains progress notes. In Baum's example, forms were color-coded so they could be easily located in the patient's chart. The five participating physicians only had to check a box for the appropriate group and sign the bottom of the form. If the physician wanted to deviate from the protocol, he or she was asked to list which tests were to be added with a brief note of explanation on the form.

6. Collect data for three months, and then analyze them and share your findings with colleagues. This was how Baum and Murtagh tested the protocols in Baum's practice. Then, after the first three months, they repeated the process for another three-month monitoring period. At the end of the each period, findings were shared with physicians. Each physician received a code that corresponded to his or her data so that the physicians' identities were protected while allowing them to compare their results with each other.

7. Calculate cost savings. After the results were determined, cost savings were calculated based on the number of studies that were not required for metastatic work-up. These were compared with costs of evaluating the previous 100 patients before the protocol process was implemented.

8. Assess physician behavior change. Compare physician practice patterns for this treatment from the beginning of the protocol or guideline testing to the end of the six-month period, and note any changes in behavior for ordering tests.