Moving from quality assurance to continuous quality improvement.
Striving to improve the quality of patient care is an ancient medical tradition. The intensity of this concern has increased greatly during the 20th Century and perhaps is at its highest level ever. A number of major organizations are supporting projects to explore or develop new methods of measuring and improving quality. The currently accepted approach to monitoring and improving the quality of hospital care (generally called quality assurance, or QA) is about to be replaced by a new set of procedures and terminology. The most popular replacement candidate has been referred to as total quality management (TQM) or continuous quality improvement. At most, there may be only minor differences between TQM and CQI, and so they will be collectively referred to here as CQI.
Major Differences Between QA and CQI
QA grows out of medicine's tradition of individual physician responsibility for the patient. In this tradition, the quality and results of medical care depend primarily on the physician's knowledge, judgment, and selection of support staff. When things go well in medical care, it is the physician who receives praise. When things go wrong, it is the physician who is chastised and corrected. Because medical care is a technically complex matter and quality is estimated principally by comparison with the treatment results of similar cases, quality standards are generally set through consensus of experts.
For many years QA focused mainly on individual cases and on peer review of case management. Customs such as clinico-pathologic conferences, tissue reviews, grand rounds, etc. have been formalized into the JCAHO hospital requirements for Surgical Case Review, Blood Usage Review, Drug Usage Evaluation, Medical Record Review, and Departmental Monitoring & Evaluation of Quality and Appropriateness of Care. The essence of each of these is regular, systematic examination of cases by committees of peers--clinicians who are at least as knowledgeable as the physician under review.
Over the years that the QA approach has been used, several weaknesses have become obvious. The criteria for case scrutiny and evaluation are usually unspecified and often are unspecifiable. Peer reviewers look at cases as if they were conducting treatment themselves and subjectively judge how far the actual treatment varies from their ideal. Review is of cases that fall below some standard and ignores borderline cases or those that are adequate but could be further improved. The objective is to find and deal with inadequate or mistaken clinicians-- often through punishment or disgrace-- thereby making the process unpleasant for all concerned and reducing the likelihood that useful learning will take place. Because the process is informal and unstructured, there is no mechanism to be certain that even serious transgressions will be corrected or prevented in the future. Finally, QA is indirect. It emphasizes "process" and "structure"--how the treatment is carried out and with what tools-- rather than "outcome"--the result that is the goal of treatment.
CQI starts with very different traditions and assumptions. It comes from mass production, assembly-line industry where the bulk of the work is repetitious and anonymous. Industry typically expects little if any ethical or social obligation between worker and consumer and not even much between worker and employer. Obligations may be specified in a collective bargaining agreement or in a job description rather than by a professional relationship. The worker often sees employment as simply a way to earn a living and sees the employer as a large faceless organization interested mainly in profit. The CQI challenge is to forge disinterested workers into a team concerned with turning out a first-rate product. The product is typically one, such as an automobile, whose advantages and quality are obvious to the consumer, and consumer displeasure with quality has an immediate impact on sales. Corporate leadership's principal obligation is to make a profit, and there is no personal relationship or obligation between corporate leadership and the consumer.
CQI has introduced a number of major innovations. First, quality is seen as nothing more or less than satisfying the consumer's desires in the product. There is no need for expert review or scientific analysis in evaluating quality. Second, quality is achieved through participation of all members of the production team, not just leaders, experts, or inspectors. Worker participation produces worker motivation to excel and also makes available the fund of information that workers have about the production process and the product itself. Third, quality depends on the process by which the product is turned out, not just on the raw materials and the design. The process itself must be examined for hindrances to quality and then improved. Fourth, quality monitoring cannot be episodic and cannot just pick out unusual events. Monitoring must be continuous and involve sampling to provide a picture of the quality level of all of the products, not just the poor ones (outliers in health care terms). Fifth, ad hoc teams of workers are used to find and solve quality problems. Teams do not look for culprits to penalize but rather for flaws to correct in the production process. Finally, the CQI system is supported with a staff of "quality coaches" who are experts in the group process of the improvement teams, but not necessarily in the technical aspects of health care delivery.
No one is certain about what will happen when CQI replaces QA in health care on a large scale. Even in industrial settings that are widely proposed as examples of how effective CQI can be, institution of CQI has not been industrywide as is proposed for health care. Unfortunately, some published conjecture lauding CQI strains credibility. For example, though the need for CQI is often placed in the context of the crisis in health care costs and funding, how much can CQI conceivably save even if it works perfectly? Much of health care expenses seems not related at all to quality of direct medical care but rather to the means of funding care and to advances in technology. In industry, quality monitoring can save money through decreasing waste, such as avoiding production of car doors that do not fit properly and therefore must be discarded. This kind of waste is not a major issue in health care.
CQI is proposed as a way of lowering the costs of malpractice suits and claims. Because a number of studies show that most malpractice suits do not involve poor quality care and most adverse events in patient care do not lead to malpractice suits, can effective CQI conceivably have an effect on the number of suits?
Clearly, additional planning and education need to be done lest organizations such as JCAHO get so far in front of their constituents that chaos results. Before CQI is accepted on an industrywide basis, some key questions need to be answered.
Will CQI's Increased Cost Over QA Lead to Improvements of Equal Value?
QA has never been a major cost element in the health care field. Is the field willing and able to pay the increased costs of CQI? Although the extent of cost increases from CQI are far from obvious, there are some bases upon which to make predictions. Berwick states that, in the ideal hospital "every single employee and physician in the hospital receives a minimum of two weeks training a year" and that 250 "improvement teams" complete their projects in a year.[*] Such activity will produce an enormous increase in the cost of quality efforts. Not all hospitals will choose to carry out the extent of training or activity advocated by Berwick, but a certain minimum will be necessary, particularly at the beginning. As Berwick implies, training of physicians will be a critical element in installing CQI. Even two days of training per year--not the two weeks Berwick mentions--will be a significant expense either to the hospital or to the physicians themselves if they donate their time.
In addition to direct monetary costs, there are two non-monetary costs that must be considered:
* The initial turbulence throughout the quality system as new ways of doing things are introduced and old ones are discontinued. Such turbulence will consume additional administrative time and perhaps at least temporarily lead to a decline in quality.
* The additional cost of the loss of quality expertise as the QA staff is dismantled and replaced by CQI staff with a completely different focus.
A conservative estimate of the average increase in cost of training and monitoring is 75 to 100 percent. This assumes that physicians will donate the time needed for training and that the members of the improvement teams will also donate their time or at least not diminish their productivity in their regular jobs.
Improvement in the quality system that will result from the new measures is much harder to estimate. Inasmuch as JCAHO is presently enforcing QA standards in every accredited hospital, CQI will not be operating in virgin territory. A 20 to 50 percent range of average improvement seems reasonable. CQI proponents need to justify this exchange of costs for benefits.
What Positive Elements of QA Should Be Preserved?
CQI is not a modification of or evolution from QA. It is a new way of operating. With the installation of CQI, several elements of QA will disappear. QA has always emphasized the work of health care professionals and ignored, or deemphasized, the work of the support staff and the opinions of anyone but the professionals. CQI depends for its definitions of quality on consumer or patient satisfaction, even though most patients have very little awareness of the technical subtleties of health care. It also depends almost exclusively on improvement teams or task forces composed of cross-sections of workers. Many of these workers are not experts in the fine points of specialized care. Though the work of all members of the health care team is important, all work is not equally important. Our contemporary approach to health care management and funding in this country is predicated on the belief that physicians are responsible for planning and directing health care while other professionals, especially nurses, play major roles. Semiprofessional and nonprofessional members of the team certainly influence quality, but they are generally not
held personally accountable, do not have the same ethical and legal obligations, and are not paid as are professionals. Physicians must be heavily involved in any quality system and their work must receive the heaviest scrutiny, with professional nursing next in line. To do this effectively, the CQI system must be designed, as the QA system has been, to maximize physician and nurse participation.
QA in hospitals and similar organizations has come to rely upon personnel employed to operate the system. They are typically called QA coordinators, and they are often nurses or medical record personnel who have specialized in QA. They have a body of knowledge both theoretical and practical, their own national organizations and certification, several journals, and an area evolving into a new profession. Because of their health care backgrounds, they often fulfill a crucial role in hospitals as informal quality advisors to physicians and administrators and do problem investigations. They serve as a focus of quality activity as well as of accountability in a setting where the official QA directors--e.g., physician department heads--are usually not employees and are available only part time.
CQI uses quality coaches, experts in the general process of improving quality, not just in health care. They assist the teams but take no responsibility for the work of the teams and do not have the health care knowledge to step in to back up a team even if they wished to do so. CQI stands or falls on the basis of teams, all of whose members have other primary responsibilities. Under CQI, there is no place where a suspicion of quality failure can be quietly reported for discreet investigation, or where accountability to get the quality work done is lodged.
QA and CQI use two different views of how to make changes to improve quality. CQI uses the carrot; QA, the stick. CQI seems to be based on two assumptions:
* Quality problems are mainly system problems, not people problems.
* When a group identifies a quality problem and a solution, all members of the group, and apparently even those who are not members of the group, will change their behavior to achieve the solution. QA assumes quality problems are mainly professional people problems and that negative sanctions and reeducation are necessary to get people to improve. In the absence of data specifically supporting one view or the other, common experience suggests that both are partially correct and neither one alone is sufficient. Moreover, common experience suggests that the public will not tolerate a health care quality system unwilling to search for transgressing individuals and unwilling to take action against those who are found.
Furthermore, consideration must be given to the special case of physicians whose work must be subject to quality scrutiny because of a variety of legal and accreditation requirements. Physicians are traditionally very reluctant to allow scrutiny of their work, except by colleagues and then only in confidence. Will physicians be willing to have improvement teams composed of cross sections of employee groups review their work and suggest improvements? If not, how will other physician review occur and be fed into the CQI system?
All health care organizations run quality monitoring systems parallel to QA but often not recognized as such. This system, operated by the organization's top administrator, will be called quality control (QC) in this article. Administration as a purchaser of services, supplies, employee time etc. carries out its responsibility to see that it gets its money's worth. Thus, through departments such as personnel, purchasing, food service, housekeeping, etc. and the supervisory mechanisms of these departments, quality is constantly monitored and problems are responded to. QC focuses on support services and leaves professional services to the QA system. Will CQI replace QC, and, if so, can administrators accept that? If not, how will CQI and QC mesh?
QA has grown beyond the tradition of examining only individual cases where treatment results were not acceptable. JCAHO has added continuous statistical monitoring to the single case approach. However, examination of adverse or "sentinel events" remains an important part of QA. Without single case examination, there is a possibility of losing sight of important instances of substandard care and allowing the substandard caretaker to continue giving poor care. Because CQI assumes that most problems are system problems, sentinel events are not generally occasions for investigation, nor is there a staff available to investigate as there is with QA.
Finally, CQI needs to relate to or encompass several of the major quality initiatives now in process that have been designed to mesh with QA. For example, JCAHO is developing its Agenda for Change, which shifts the quality focus to the outcome of care from just the manner in which the care is given. The Agenda involves an approach completely different from CQI in several aspects. Will the Agenda be discarded? Will other outcome focused systems be ignored?
Does CQI Fit the Culture and Conditions of Health Care?
Some elements of CQI may not be suitable or acceptable in health care. Industrial CQI pays special attention to variability in end products and assumes that variability is generally evidence of poor quality work. Because in assembly line work all raw materials are uniform, variations at the end must have been introduced by workmanship. The situation in health care could not be more different. None of the "raw material" is uniform and, although variability at the end certainly can result from the work process, it may have been present from the beginning. Variability as a measure of poor quality and CQI as an attempt to eliminate variability must be rethought.
A means of appropriately determining high quality must be developed. In industry, it is fairly simple to measure quality. Aside from asking consumers, who have much of the information they need to form useful opinions about quality, there are usually precise engineering standards available. Health care is not amenable to such definitions.
Health care has traditions, ethical standards, and legal accountability that establish direct obligations between individual patients and individual clinicians. QA has built upon this set of obligations to monitor quality on a typical industrial model. Improvement teams, task forces, and similar groups are not part of the health care tradition and may encounter strong resistance. Early stages of most new programs are full of excitement and attract as supporters those who are most inclined to welcome change. Evaluators of such programs must take into account what is a temporary placebo-like effect in estimating what the program is likely to be when in full-scale, long-term operation.
Combining CQI ond QA
Unfortunately, things have moved so fast that there has not been enough time to examine each major element of CQI and QA to determine what works best in a health care setting. While examination still needs to be done, this is not the place to begin it. However, certain elements of QA are absent from CQI and appear to have been particularly useful. These elements should get special consideration for amalgamation into CQI.
First is concern with and involvement of professional personnel, particularly physicians, who are not employees of the hospital. This includes investigation of single adverse clinical occurrences--sentinel events--and responses to these investigations. Even though searching for bad apples may be unpleasant and sometimes futile, it is an activity the public expects of us, and every so often it leads to major improvement in care. Furthermore, it takes advantage of and reinforces the health care traditions of personal responsibility and ethical obligations. Other parts of this element are definition of high quality by expert consensus and evaluation of quality in individual cases by peer review. The tremendous effort currently going into preparation of practice guidelines cannot be ignored. Patients and nonprofessional staff simply do not have knowledge on the same level as that of expert practitioners, though their opinions certainly ought to be included. Finally, there is a need for special ways to train and involve private practitioners in CQI training and deliberations. Although such individuals do not have the time or the motivation to get involved to the same extent as full-time employees, their involvement is critical. The effectiveness of the time they' are able to put in must be maximized.
Second is the QA support staff-- the coordinators--whose essential characteristic is that they are trained in both health care and quality monitoring. They are willing and able to carry out a wide variety of tasks necessary to keep the system operating. Because of their status as health professionals, they are trusted to carry out the health professional traditions. Quality coaches whose training is in group process and whose objective is primarily to help other people keep the system operating cannot replace QA coordinators.
Third is operation of the QC system as part of administration's obligation to be a responsible purchaser of goods and services. This aspect of quality monitoring is virtually ignored in discussions of CQI, yet it cannot be discarded, any more than QA can. At the least, QC has to be permitted to operate independently, although ideally it should be integrated into CQI together with QA.
Medicine has much to be proud of in its accomplishments in improving the quality of care over the years. Though continuous quality improvement represents further progress in this quest, it is wasteful to discard successful QA and QC elements in our present approach. Continuous quality improvement must grow out of quality assurance and be adapted to health care traditions and culture rather than simply adopted from industry.
[*] Berwick D. "Peer Review and Quality Management: Are They Compatible?" QRB 16(7):246-51, July 1990.
Jack Zusman, MD, is Professor, Florida Mental Health Institute, University of South Florida, Tampa. He is a member of the College's Forum on Quality Health Care.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Medical Quality Management|
|Date:||Jul 1, 1992|
|Previous Article:||Choices and Conflict: Explorations in Health Care Ethics.|
|Next Article:||Reducing health care costs: a case for quality.|