As we struggle to provide the best care possible for our patients, new methods are needed to determine the best currently known practices and for assuring that each patient receives this care (1, 2). This has led us to develop 2 new sciences, evidence-based medicine (3) and the pragmatic science of standardizing care processes (4, 5). These sciences must be carefully integrated into the traditional practice model, wherein care providers use their expertise at the bedside to tailor care to the needs of individual patients. The standardization of care is used to assist the care providers by providing current knowledge and best known care processes to promote effective, efficient care while safeguarding against errors and harm. Such standardization, however, should be integrated with the knowledge and skills of clinicians who may have unique perspectives of the individual patient and his or her wishes and of the environment (6).

Many organizations have established guidelines similar to the guidelines for antithrombotic therapy derived from the American College of Chest Physicians (ACCP) Consensus Conference in 1998 mentioned in the current article by Dr. Robert Baird. Such guidelines serve as a wonderful source of the best scientific evidence currently available in the area being addressed, along with the opinions of experts in the field. They suffer, however, in not being explicit enough to be useful at the bedside, and their practicality, acceptability, and safety in the environment where they are to be used is unclear.

This deficiency has resulted in the new pragmatic science of standardizing care processes or quality improvement. In traditional science, a single hypothesis is tested, keeping the experiment rigid and constant. By demanding strict entrance and exclusion criteria, the influence of confounding elements is reduced, and by implementing double blinding, bias is prevented. The goal of traditional science is to demonstrate cause and effect.

In the pragmatic science of process improvement, using evidence created by traditional science, a change hypothesis is generated. Then a series of small changes that are believed to result in favorable outcomes is tested in small, rapid cycles while measuring the direction of the results with control or run charts to ensure that improvement in care is occurring and harm is avoided. At times, change concepts may be applied in parallel, such as changing the process of sedation and enteral feeding and implementing a weaning protocol simultaneously, with the desired outcome of reducing ventilator days. Cause and effect is not the purpose, as many changes are occurring simultaneously. The cumulative process change determines the result rather than any one element. Bias is stabilized using control charts for measuring change but is not totally eliminated. The aim is to demonstrate significant improvement in care in a real-world population of patients with the biases and confounders present.

The advantage of testing rapid, small changes is that the experiment can be controlled, and when necessary the process can be changed to prevent harm, increase acceptability of the frontline care providers, and allow the molding of the process to fit the patient population and the local clinical environment. This technique allows for challenge by the clinical team and results in their buy-in and ownership of the protocols. During testing, ambiguity and vagaries are identified and the protocols clarified for clinical use. But most importantly, frontline workers engage in the process and learn new methods of improving care that can be applied to other problems. This creates a new culture, one of constant improvement and safer care.

The current article by Dr. Baird entitled “Quality improvement in the intensive care unit: development of a new heparin protocol” is an excellent example of the pragmatic science of quality or process improvement. A problem was identified with the control of heparin use, with many different methods being applied, usually without regard to the current scientific evidence. The different methods resulted in inadequate care, with a mean time of 63 hours to control the patient's anticoagulation in the desirable range. This time was much longer than necessary and was placing patients at risk.

The process improvement team identified the scientific evidence available and proceeded to develop from the guidelines an explicit protocol that would be usable, practical, and acceptable in their environment. As they tested the initial protocol, they found problems. As a result, they modified the protocol to fit their environment, changing the initial bolus from 18 to 14 U/kg/hr, thus improving safety. They allowed the frontline workers to decide if they wanted to use the activated partial thromboplastin time or heparin assay to make the protocol more acceptable. When evidence that one test was superior to the other became available, however, they adopted the better methodology. The ACCP consensus conference guidelines that are lengthy and nonspecific were condensed into a practical 1-page protocol in the form of an order sheet that could easily be implemented in the intensive care unit. This facilitated ordering heparin while assuring proper monitoring and dosage change as needed.

The results were clearly in the right direction and although, at the writing of the paper, there were not enough data for statistical significance, there are enough data to show improvement. The intermediate or process outcomes of improving the percentage of patients receiving the correct heparin bolus from 8.6% to 90% and the optimal infusion doses from 3.4% to 100% is spectacular. Even more important, the clinical outcome of proper anticoagulation being achieved in 34 vs 63 hours is clearly an improvement in care, as the time to proper anticoagulation is correlated with a reduction in pulmonary emboli.

Determining the best known care through literature review and the process of evidence-based medicine techniques is not always easy but is usually doable. Implementing the process is much more difficult, as it requires change in current practice styles and a new way of viewing the practice of medicine. Implementation requires a cultural change that is created when care is improved as the new pragmatic science of standardizing care is applied and its use is expanded into other areas by the frontline teams.

1. Kilo CM, Kabcenell A, Berwick DM. Beyond survival: toward continuous improvement in medical care. New Horiz 1998;6:3-11.
2. Nolan TW. Understanding medical systems. Ann Intern Med 1998;128:293-298.
3. Cook D. Evidence-based critical care medicine: a potential tool for change. New Horiz 1998;6:20-25.
4. Berwick DM, Nolan TW. Physicians as leaders in improving health care: a new series in Annals of Internal Medicine. Ann Intern Med 1998;128:289-292.
5. Brock WA, Nolan K, Nolan TW. Pragmatic science: accelerating the improvement of critical care. New Horiz 1998;6:61-68.
6. Clemmer TP, Spuhler VJ, Berwick DM, Nolan TW. Cooperation: the foundation of improvement. Ann Intern Med 1998;128:1004-1009.

.