The physical constraints once identified are still difficult to manage due to conflicts in the mental models of the stakeholders. The core conflicts come between cost containment (trimming personnel until everyone is always busy, for example) versus increasing Throughput (by having protective capacity at all support functions). The other related core conflict is between local optimization (measures that focus on individual performance) versus global results (measures that focus on organization performance). These core conflicts and other conflicts are studied using the Evaporating Cloud (EC) technique. In Fig. 31-10, the core conflict of increasing revenues versus controlling OE is portrayed as an EC with its assumptions and injections. Many of these injections15 (actions) were used in the doctor's application.

FIGURE 31-10 EC with assumptions and injections of the core conflict of hiring more staff versus keeping few staff.

Throughput Accounting for Performance Measurement and Decision Making in Health Care16

TA for health care is different from the usual TA17 in that health care service is mostly intangible. In most instances, the doctor should be treated as the constraint and the patient as the consumer. Therefore, for healthcare, Throughput (T) means the rate of cash generation through delivery of high quality, reliable service to patients. T is the payment for services related to a specific patient minus the variable cost of laboratory work, supplies, etc. for that patient. Total T is directly related to Q (quantity of patients treated and paid for in a given time) and dollar value per patient. The quality and reliability of the process directly influence the amount of doctor's time spent managing the care of the patient.

Investment (I) is the total capital invested in designing the physical sites and delivery system of the service for the patients. It includes the cost of physical facilities, equipment, tools, IT systems, HR system, and money spent to obtain market data to develop the services for the target market. This investment is depreciated over time as OE.

Totally Variable Cost (TVC or more frequently VC) is the cost of supplies and laboratory work paid for specific tests. Since this VC varies significantly for each patient or segment of patient population, it is subtracted when the patient treatment is completed. In health care, it is not possible to focus on each patient cost due to the high degree of variability. We usually look at a segment of population served, for example, the patient population by insurance companies versus by age groups or by specific procedures.

In the hospital setting, insurance is paid by International Classification of Diseases (ICD) code and if Medicare population requires a lot of laboratory testing and multiple supplies, and increased length of stay, the VC will increase compared to other populations.

The symbol I18 is used for investment that is depreciated over time, and the symbol i is used for inventory that includes medical, surgical, and office usable supplies. It can include all the unfinished treatment plans or unpaid bills from insurance companies that is similar to inventory waiting to be worked on. OE is all the expenses to deliver services including doctors and staff salaries, benefits, leases, equipment, utilities, insurances, supplies, etc. It also includes selling and general administrative costs. I is depreciated over time as an OE. Little i is the cost of supplies, lab work, and work-in-progress (WIP).

As in TA, one strives to increase Throughput while decreasing Investment and Operating Expenses. The normal relationships among the variables still hold.

NP = T OE

ROI = NP/I.

In health care, T can increase if we increase the velocity of understanding patient expectations, diagnose the problems accurately, create treatment plans (like design engineers), and execute the best treatment option for the patients in the shortest possible time (similar to multiple projects). The quality and reliability of patient care provided by the doctor is important in this concept because we will lose the time of our most valuable resource, the doctor, if we have to readmit the patient or we have to contact the referring doctor again and again to get the test results.

In health care, the management priorities are to: Increase Throughput (T) > Increase in Investment (I) + Increase in Operating Expenses (OE) In decision making and selection of patients in a for-profit health care organization, it is important to understand T/DU.

If the patient care takes too long from start to finish or a third party takes too long to pay, it will increase the OE, increase i, and decrease (T).

Staff turnover cost is viewed in terms of impact on total Throughput that is the effect on doctor time and collection of fees. To be a good decision, the hiring of staff must result in an increase in T (T) that is greater than increase in OE (OE) caused by the hiring of staff.

When referring procedures to other practitioners (similar to outsourcing), the decision must be made based upon overall impact on the NP of the practice at the end of the year. In decision-making, we take into account the cost of taking time off from the practice, tuition paid to develop specialized skills, investment in equipment and inventory, hiring and training of staff, opportunity cost of time allocation to providing care to select patients, marketing and sales to potential patients, and quality of service including readmitting the patients for care. The decision criterion is: The change in Net Profit NP = T > OE Decisions about developing new services in the practice must also be based upon this formula. If total T increases greater than the OE after accounting for all OE and opportunity costs, we will increase NP. Any Investment required can be expensed over time as part of OE in making the decision. If the primary care has a small laboratory to do simple tests including ECG, pulmonary function tests, blood tests, urine analysis, etc. and the increase in OE after all the investment is less that the increase in T, then the investment is a good decision. On the other hand, if the primary care wants to add imaging service in their practice, they will have to include all the I expenses in equipment, additional personnel, and time of their critical resources like the doctors learning to read CT scans and opportunity cost of not seeing regular patients while spending time in reading scans. If all this adds up to more OE than the increase in T, the decision to invest in an in-house imaging center must be abandoned.

Both Throughput Dollar Days (TDD) and Inventory Dollar Days (IDD) are valuable measures in healthcare as well. Decisions about integrating care with select specialists in health care similarly must take into account TDD. TDD is the Throughput you would have had if a certain specialist, laboratory, or imaging center had completed their work on time. It is a penalty for lateness. IDD is made up of open treatment plans sitting queued in front of a specialist in the integrated network of health care providers with incorrect or incomplete information from the primary care provider, the laboratory, or the imaging center. It is a penalty for earliness (or doing something that should not have been done). The providers in the network who jointly treat patients can develop an informal or formal system of accountability based upon TDD or IDD.

Now that we have seen the approaches to improvement and the measurements to account for them, we will move to S&T and the approach to strategy and tactics for a medical practice.

Strategy and Tactic Tree19 to Implement and Achieve the Viable Vision

As noted earlier, the VV is an approach whereby a company maps its strategy of how to achieve NP within four years equivalent to its annual sales today. The strategy and tactics tree (S&T) is fundamental in mapping out a detailed strategy for achieving this outcome. In the TOCICO Dictionary (Sullivan et al., 2007, 43-44) the S&T Tree is defined as: " A logic diagram that includes all the entities and their relationships that are necessary and sufficient to achieve an organization's goal. The purpose of the S&T Tree is to surface and eliminate conflicts that are manifested through the misalignment of activities with organizational goals and objectives.

Usage: Organizational strategy specifies the direction of the activities that purport to address longer range problems and issues. Tactics are the specific activities needed to achieve the strategic objective involved in implementing organizational strategies. Since strategy and tactics exist and must be synchronized within various organizational levels, this logic tree translates high level strategy down to the level of day-to-day operations." ( TOCICO 2007, used by permission, all rights reserved.) Strategy tells us "What to achieve" and tactics tell us "How to achieve" it. Goldratt (1990a, 5051) also points out that the most important part of any system, including a health care system (for-profit or not-for-profit), is the focus on Throughput instead of the traditional focus on cost savings. The VV is achieved by increasing the rate of flow or velocity of patient flow through the system while ensuring a high level of quality/reliability of services as measured by excellence in clinical outcomes, and total end-customer (patient) satisfaction.

The S&T tree for the VV in health care shows a hierarchical logical tree to achieve the goal. It starts with the firm agreeing on a goal. An example of a goal in a for-profit organization is to improve shareholder value. However, shareholder value can only be achieved if S1 the company is making profits over time. The profits are possible only if S2 the company is providing high value at a reasonable price to its customers. In order to develop high value services, S3 the company must develop delivery systems that provide this value and delivery systems that require highly capable people to make it happen. The highly capable people must be hired, trained, and motivated by its leadership to make it possible. Goldratt calls it strategy and tactics. The lower level specific objectives or tactics to achieve higher-level goals is the strategy.

As seen in Fig. 31-11, all of the steps S1/T1 + S2/T2 + S3/T3 are necessary and sufficient-to achieve the strategy at the above level. The tree includes the strategy and tactics with the logical linkages for the parallel assumptions, necessity assumptions, and sufficiency assumptions.

Parallel Assumptions

Parallel assumptions show why tactics are necessary and how they lead to a strategy being met. At each step, we claim that the specific action plan or tactic will achieve the strategic objectives. This claim is subject to the following challenges: 1. There is no need for an action to achieve the strategy.

2. It is not possible to take the action.

3. There is another, better alternative.

4. There is a need for additional action.

FIGURE 31-11 S&T with assumptions relationships.

How to Find Parallel Assumptions

A parallel assumption is constructed to explain the following: 1. What is currently missing that is preventing us from attaining the desired strategy?

2. Why nothing else besides what is written in tactics can achieve the strategy.

3. Disqualification of the selection of less suitable alternatives.

4. In case the tactic is challenged as a flying pig,20 the lower level details substantiate the claim.

It is important to use language as a tool to verbalize these assumptions. For example: In order to achieve the strategy, I must take the action in the tactic, because . . . . The "because" response of the statement is the parallel assumption.

Necessary Assumptions

A step (for example, S1, S2, or S3) is necessary to achieve the corresponding next higher level (for example, from Level 1 to Level 2). It is important to have an explicit explanation (the necessary assumptions) of why a given step (S1, S2 or S3 in Level 2, for example) is necessary to achieve the higher next step (Tactic x in Level 1). There could be several necessary assumptions. It could be an answer to objections raised that this step is not necessary to achieve the next level results.

Here again the assumption should be verbalized. It should be stated as follows: In order that this step is achieved, we must do another step at the next higher level because . . . Again, the "because" response is the necessary assumption.

Sufficiency Assumptions

When we claim that a group of steps (S1, S2, and S3) is sufficient to achieve the next corresponding higher-level step (S X), we must explicitly explain (sufficiency assumptions) why all the corresponding steps of the lower-level group are sufficient to attain this step. We write only the necessary conditions that are sufficient as a group, and an action that is necessary to achieve them. Sufficiency assumptions are expressed as: If Step 1 and Step 2 and Step 3 (S1, S2, and S3) . . ., then the higher-level step can be completed.

In order to build the tree, it is prudent to start at the higher level. Start with an objective. What is the purpose of this system? What is the reason for the system's existence?

What is the action (tactic) necessary to achieve this purpose? We write all actions necessary to achieve this purpose in the present context of knowledge. These actions cumulatively must be sufficient to achieve the objective. Verbalizing the parallel assumption, why did we choose the tactical entity to achieve the corresponding strategic objective?

An Example

Let us apply this template to an example in Fig. 31-12. The S&T Tree is read from the top down and from left to right. The logic forces us to ensure that important things are not ignored or missed. Quality and reliability of service comes first, then comes marketing, and finally the growth strategy. The staff in the organization must carry out all these improvements. The processes or systems do not exist in a vacuum. Staff executes the tasks and if these tasks are tied to the system's goal, the system will succeed.

Level 1

The S&T Tree documentation contains two elements: the tree itself as shown in Fig. 31-12, and an information table as shown in Table 31-2. These two must be related in reading the S&T Tree. In the following example, we see at Level 1 in the strategy we have the "Health Care System Viable Vision." Relating Table 31-2 to it, we have in the upper left corner a reference to Level 1 of the S&T. This reference in the upper left corner of each table links the table to the S&T Tree structure. Here, we find under "Vision" a more explicit explanation of just what the vision is. Then, looking again at the S&T in Fig. 31-12, we see the first level of tactics (2.1 Base Growth and 2.2 Enhanced Growth). Under "Assumptions behind Tactics," a PA, we see the conditions that must be met in order for the strategy to happen: "Revenue must grow (and continue to grow) much faster than Operating Expense." The assumptions also make clear that this must be done without exhausting the people who work within the system. Now, under "Tactic" we see the tactics that must be employed at the next level down (Level 2) of the S&T. Here we call out Tactic 2.1 "Quality/Reliable Patient Service Competitive Edge" and 2.2 calling for a special competitive edge in premium markets (Premium Competitive Edge).

FIGURE 31-12 VV S&T, Level 1 VV, Level 2 and 3 base and enhanced growth, and Level 3.

TABLE 31-2 Strategy, Tactics, and Supporting Assumptions for Level 1 If the goal of the practice is to make money within the context of a disease management model and with full ethical responsibility, then it must increase the Throughput of the system significantly greater than the increase in OE to achieve the health care system VV in 1 (" S1"). In order to do this, the health care practice, hospital, or integrated health systems must show the capability of developing a Decisive Competitive Edge (DCE) over its competitors (Tactic 1, the next level down [Level 2] in the S&T). This means that they achieve breakthrough results in quality and reliability of patient care service to ensure Base Growth as in S 2.1. The system has competence to attract high premium patients for Enhanced Growth as in S 2.2.

The necessary assumption (assumption behind tactics in Table 31-3) is that quality and reliable service will improve the velocity of the flow of services without unnecessary delays and without readmissions. In addition to that, the patients will get highly reliable, quality care especially designed to capture the patients' needs and wants. This will increase customer satisfaction, which will increase the reputation index in the market place and thus result in an increase in referrals to our system.

The sufficiency assumption (Take Note) is that it is not sufficient to have reliable services in order to make breakthrough profits. We must have additional competence to attract premium customers who pay higher than usual fees for our services. Higher premium patients result in a significant increase in Throughput without adding to the OE except for marketing and advertising expenses.

The tactical action plans to achieve S 2.1 are to implement initiatives that help develop DCE over competitors through operational excellence, sales mastery, and capacity expansion.

TABLE 31-3 Strategy, Tactics, and Supporting Assumptions for Level 2 The S&T tree is a company-wide alignment, synchronization, and communication tool. The goal is to have an ever-flourishing practice that continuously and significantly increases value for customers (patients), staff, and stakeholders. In the S&T, we agree that we are going to transform the revenues into net profits in less than four years to achieve the VV.

The tactic is to develop a DCE with recognition as a leader in providing high quality, reliable service to a select group of patients, and develop capabilities to capitalize on it without exhausting our staff capabilities or taking real risks.

The parallel assumptions or assumptions behind the tactics are that in order to realize the vision, our revenues must continue to grow much faster than the OE (hence, Throughput continually grows). However, if the growth is too fast and the staff-applied capability is not able to cope with the growth rate, the systems will collapse or oscillate and the quality and reliability of service will suffer. This will generate a negative feedback loop contrary to the vision.

Sufficiency assumptions (under "Take Note" in Table 31-3) are our reasons to believe that accomplishing a DCE will be at risk without providing another level of detail to our subordinates. The DCE is to satisfy a patient's significant need to an extent that most other competitors cannot and will not do.

Level 2

Here we move to the next level in the tree as we see in Figure 31-13. Necessary assumption or assumption behind this level of strategy is that in healthcare the patients do not visit doctors for fun; every visit is stressful and a somewhat traumatic experience. The longer the patient has to wait to complete the care (the number of visits to the doctors), the greater the stress. Therefore, high quality and reliable treatment service is the patient's significant need.

Strategy at this level (Level 2) is to have DCE by awareness in the market that our practice has a unique, systematic ability to complete all the necessary treatment with the fewest visits, shortest overall duration, and predictable reliable outcomes.