Determining physician compensation -- Part II. (Nuts and Bolts of Business).
For the base pay with incentives model, the first step in using this model is to determine the source of funds for the bonus pool. Groups may:
* Hold back a portion of salaries to create the incentive pool
* Budget a separate pool of funds
* Rely solely on profits generated by the practice
The second, more difficult, component of this model is to determine how to distribute the bonus pool among physicians. The group must decide if the distribution will be based purely on clinical productivity or whether it will include other factors.
For academic practices, factors such as teaching, research and administrative responsibilities may come into play when determining the distribution of these funds. Yet, how do you place a value on these criteria when determining incentive fund distribution?
One method that allows you to place a value on clinical as well as non-clinical factors is analytic hierarchy processing (AHP), developed by Thomas L. Saaty of the Wharton School of Business. There are five steps in the AHP process.
1. The problem to be solved is structured as a hierarchy.
2. Judgments are elicited regarding the components of the hierarchy.
3. Pair-wise comparisons are performed to represent those judgments with meaningful numbers.
4. The results for various alternatives are synthesized.
5. The sensitivity to changes in judgment are analyzed.
Let's look at a simple example to see how AHP can solve the problem of placing a number value on non-clinical and clinical criteria when determining incentive compensation.
Begin by structuring a hierarchy. The goal in this example is to determine the distribution of an incentive pool. This represents the top level of our hierarchy.
Next, we decide on the criteria we use to make judgments for determining the distribution of funds. Let's say the criteria will be:
* Clinical productivity
* Administrative duties
* Research productivity
* Teaching ability
Now, we determine alternatives for the goal. This is the third level of the hierarchy. In this example, the alternatives will be Doctor A and Doctor B and the distribution of the funds will be to these two alternatives.
Figure 1 illustrates at the structure of this simple hierarchy.
The next step is to elicit judgments regarding the criteria by developing a matrix and making pair-wise comparisons between the alternatives.
Figure 2 is a matrix that allows us to compare the four criteria of clinical productivity. We then judge each of the criteria against one another, determining the level of importance of each when determining our goal.
Let's say the group decides that clinical productivity is twice the value of administrative duties, three times the value of research, and four times the value of teaching.
This is what appears across the first row of the matrix. In the first column, we see the reciprocals of the first row. Likewise, if we value administrative duties to be twice the value of research and three times the value of teaching, we get the results in the second row, with the reciprocals in the second column.
Finally, if we value research to be twice the value of teaching, we see the results by completing both rows and columns three and four. We then add the values of each column.
Weighting the criteria
The next step is determining the weights of each of the criteria. We figure the weights by dividing the individual values of each criterion by the total value of the column.
For example, the value of clinical productivity is one. The total value of the column is 2.08, so the weight is one divided by 2.08 or 0.5. We repeat this process for each value in the matrix in Figure 3.
We then average the results across each row to normalize the weights to equal one. In the same way, we then make pair-wise comparisons and determine weights regarding the alternatives for each of the four criteria.
As you can see, Doctor "A" is more clinically productive, while Doctor "B" is stronger in each of the other three criteria. The results appear in the following figures.
Next, we synthesize the results for each of the alternatives. This is accomplished by determining the sum of the products of the weight of each criterion by the weight for each alternative.
For example, the synthesized results for Doctor A are:
(individual weight for clinical productivity) (weight for clinical productivity) + (individual weight for administrative duties) (weight for administrative duties) + (individual weight for research productivity) (weight for research productivity) + (individual weight for teaching ability)(weight for teaching ability), or
(0.8)(0.45) + (0.3)(0.25) + (0.3)(0.2) + (0.2)(0.1) = 0.52.
The results for Doctor B are:
(0.2)(0.45) + (0.7)(0.25) + (0.7)(0.2) + (0.8)(0.1) = 0.48.
So, based on the judgments made, Doctor A would receive 52 percent of the incentive pool and Doctor B would receive 48 percent. The final step is to alter the judgments and perform a sensitivity analysis to see how results might change.
While AHP is a very useful tool, the computations can become quite tedious, especially for large hierarchies with many alternatives.
Fortunately, there is software available that allows you to quickly build hierarchies make comparisons, compute weights and perform the synthesis process. In addition, it allows you to quickly conduct a sensitivity analysis. The software, Expert Choice(r), was developed by Dr. Ernest Foreman and is commercially available.
Analytic hierarchy processing is a powerful tool that can help place meaningful numbers on criteria to determine incentive compensation, even criteria that are not often given values.
[FIGURE 1 OMITTED]
FIGURE 2 Matrix for pair-wise comparisons C=Clinical productivity A=Administrative duties R=Research productivity T=Teaching ability C A R T C 1 2 3 4 A 1/2 1 2 3 R 1/3 1/2 1 2 T 1/4 1/3 1/2 1 2.08 3.83 6.5 9.0 FIGURE 3 Matrix for determining weights of the criteria C A R T Avg. C 0.5 0.5 0.4 0.4 0.45 A 0.2 0.2 0.3 0.3 0.25 R 0.2 0.1 0.2 0.2 0.20 T 0.1 0.1 0.1 0.1 0.10 Pair-wise comparisons and average weights of the alternatives for Clinical Productivity A B Avg. A 1 4 0.8 B 1/4 1 0.2 1.25 5 1.0 Pair-wise comparisons and average weights of the alternatives for Administrative Duties A B Avg. A 1 1/2 0.3 B 2 1 0.7 3 1.5 1.0 Pair-wise comparisons and average weights of the alternatives for Teaching Abilities A B Avg. A 1 1/4 0.2 B 4 1 0.8 5 1.25 1.0 Pair-wise comparisons and average weights of the alternatives for Research Productivity A B Avg. A 1 1/3 0.3 B 3 1 0.7 4 1.33 1.0
David P. Tarantino, MD, MBA, is the executive medical director of Shock Trauma Associates, P.A., a 50+ physician, multispecialty practice associated with the University of Maryland School of Medicine. In addition, he is the chief executive officer of The MD Consulting Group, LLC, a health care management consulting firm in Baltimore, Md. Tarantino can he reached by phone at 410/328-3198 or by e-mail at email@example.com.
|Printer friendly Cite/link Email Feedback|
|Author:||Tarantino, David P.|
|Date:||May 1, 2002|
|Previous Article:||Divesting losers: chipping away at integrated delivery systems. (Management).|
|Next Article:||Ask the coach: expert opinion to answer your medical management questions. (New Column).|