If the value of a healthcare system is defined as the quality of the system divided by its cost, then the value of care in the United States is very low compared to other developed nations. The United States spent nearly 18% of GDP in 2012 on health care, while the next highest spending among developed nations was 12%.¹ It is estimated that more than $700 billion of healthcare spending each year in the United States is waste.² In the measure of outcomes, the US healthcare system ranks 37th globally.³ Low quality divided by high cost equals a system that has great potential for improvement.

Inpatient hospital care accounts for nearly one quarter of the annual amount spent on child health,⁴ with that figure climbing even higher when emergency department visits are added. Therefore pediatric hospitalists are in a position to exert strong influence and leadership upon improving the value of that care by increasing the quality and decreasing the cost. Examples of quality projects in hospitals that have improved value include reduction of catheter-associated bloodstream infections (CLABSI),⁵ reduction in mortality through rapid response teams,⁶ reduction in codes outside of the ICU,⁷ increases in hand washing,⁸ decreases in contaminated blood cultures,⁹ and decreases in identification band errors.^10,11 Payers have already begun to increase pressure on hospital systems to deliver value by implementing payment penalties for care that is perceived to be of low quality. The Center for Medicare and Medicaid Services (CMS) has issued instructions to stop paying for CLABSIs and pressure ulcers that occur during a hospital stay. Additionally, despite the ambiguity of evidence that unplanned readmissions are either preventable or a marker of poor quality in pediatrics,^12,13 many state Medicaid plans have begun to penalize hospitals for readmission rates that are perceived to be high.

Historically, one of the barriers to improved quality was that the return on investment did not accrue to the hospital or the physicians. For example, decreasing a hospital’s CLABSI rate would lower charges and payments, thus lowering the revenue stream to the hospital and increasing profits for the payer. Over the previous decade, many individual quality measures had payment penalties associated with them as an incentive to the hospitals to improve. However, with the passage of the Patient Protection and Affordable Care Act (PPACA) in 2010, the potential for the incentives around quality improvement to align with both the payer and the hospital increased dramatically. One of the cornerstones of the PPACA was the creation of vehicles known as accountable care organizations (ACOs) that allow institutions to take financial risk for a group of patients.¹⁴ By taking on that risk, the increased value in the system that is created by quality improvement can accrue to the institution bearing the risk, instead of an insurance company.

Using evidence-based medicine and the diffusion of new advances to change physician practice is challenging.¹⁵ Recently, however, it is becoming accepted that evidence-based medicine and care practice guidelines represent an excellent way to increase utilization of proven therapy processes, which commonly increase quality and decrease cost (as discussed further in Chapter 3)

Walter Shewart and W. Edwards Deming are generally thought of as the fathers of the modern era of quality improvement, as they introduced new theories and techniques for process control and quality assurance.¹⁶ Many of their techniques are still in use today. What follows is a brief description of the most valuable of those techniques for improving hospital care.

PLAN-DO-STUDY-ACT

Also known as the PDSA cycle, Plan-Do-Study-Act describes the four phases of a quality improvement project. The “plan” phase is comprised of an analysis of the deficiencies of the system to be improved and the actions that will be taken to improve it. Once the plan is established, the “do” phase begins, and the plan is put into action. While those actions are taking place, predefined measurements are collected, and an analysis is undertaken to “study” how well the original plan is providing a remedy for the deficiencies in the system. After understanding where the original plan needs to be altered, those changes are “acted” upon, and the cycle continues.

STATISTICAL PROCESS CONTROL

Statistical process control (SPC) is a statistical tool that allows for representation of data on a graph that helps users to understand what parts of processes are due to “natural” or “common cause” variation in systems, and what parts are due to changes that were made to a system, or “special cause.” With this method, variation in processes and systems can be understood in significantly less time than it would take to sample every single event, or to establish control groups,¹⁷ which is often an impossibility in a hospital. (Having a process run differently every other day just to study which way is better would be impossible for most processes in a modern hospital.)

The fundamental tool in SPC is the control chart. The control chart takes a run chart (a series of measurements plotted in chronological order around a mean) and adds an upper control limit (UCL) and a lower control limit (LCL). These control limits are calculated by known formulas that rely on the measurement of variation inherent to the system. Points that fall outside of one of the control limits is considered to be a result of a “special cause”; something new in the system caused the change, as opposed to the normal random fluctuations one sees between the control limits.¹⁸ An example of an SPC chart for an improvement project in reducing identification bands defects in a children’s hospital is shown in Figure 4-1.

SPC charts are the simplest and most powerful tools in the quality improvement arsenal to track the effects of a PDSA cycle project. Although it is relatively simple to create an SPC chart using published formulas, many commercial programs are available that will create the appropriate charts when the user simply enters the collected data.

THEORY OF CONSTRAINTS

The Theory of Constraints was introduced by Eli Goldratt in his novel The Goal,¹⁹ an allegory that tells the story of a manager whose plant is about to be closed. Through a series of encounters with a physicist, the manager creates rules that allow the system to reduce bottlenecks (constraints) and improve efficiency. In practical usage, this technique allows quality improvement efforts to be focused where they will return the most improvement.

Consider, for example, a desire to decrease IV drug delivery times during the night shift. A hypothetical pharmacy system might work like this (see Figure 4-2): The pharmacy printer which receives the order can print 10 orders per hour → the tech can mix 6 orders per hour → the labeling machine can label 20 orders per hour → the pharmacist can check 8 orders per hour → the delivery person can distribute 5 orders per hour.

If a hospital were to start a quality improvement project without first knowing all of the steps and constraints, it might add another tech with the expectation that the process rate would move to 8 doses per hour, as that is what the pharmacist can check. However, since the delivery person can only deliver 5 doses per hour, no improvements in IV drug delivery times would actually be seen.

Mr. Goldratt provides five specific steps for solving production problems in his novel, but they essentially boil down to the example above. One must find the tightest bottleneck (constraint) in the system and solve that one before solving any others, as solving any bottleneck other than the tightest one produces no value.

LEAN SIX SIGMA

Lean and Six Sigma began as two separate methodologies that have merged into one way of thinking in its application in healthcare. Originally, Lean was concerned with the reduction of waste in the process steps of a production schema, while Six Sigma was invented by Motorola as a method of describing statistically how many defects per million were produced.²⁰ That is to say, Lean makes a process more efficient, while Six Sigma makes the output of that process have fewer defects.

Clearly, both improved efficiency and fewer defects are laudable goals in delivering care in a hospital setting, and thus Lean Six Sigma has been combined into one overarching strategy. In general, the Lean part of a project focuses on what is known as a “value stream map,” which organizes all parts of the process into “value added” and “non-value added” steps. The steps that do not add value are considered waste, and there is an attempt to eliminate or minimize those steps. Lean itself has a standard group of solutions that are often effective to deploy in process redesign. The Six Sigma part of a project focuses on the rate defects in the product. (Six Sigma takes its name from a six standard deviation in a bell curve, which amounts to 3.4 defects per million opportunities.) Six Sigma is in many ways similar to the PDSA cycle, but with a control phase at the end. It relies on a routine of Define, Measure, Analyze, Improve and Control (DMAIC).²¹ Often the control phase is monitored by an SPC chart. An in-depth explanation of Six Sigma is beyond the scope of this text, but the understanding of how Lean and Six Sigma complement each other and can be used simultaneously is valuable to those pursuing quality improvement.