From quality assurance to continuous quality improvement.

During the 1970s and 1980s, quality assurance (QA) was a provider-oriented, defensive response to requirements of external agencies, such as the Joint Commission on Accreditation of Healthcare Organizations, [1,2] and a reaction to increased liability of health care centers for the performance of all individuals, including physicians. [3-5] There is little, if any, evidence to suggest that the quality assurance era had a significant positive impact on American health care. In contrast, continuous quality improvement (CQI) is a consumer-oriented, proactive response to negative public perceptions of the health care business, which range from disappointment to distrust. [6] Eleven essential points differentiate QA and CQI.

Point 1

The CQI goal is to improve the norm of practice and behavior, as opposed to settling for compliance with a standard based on normative behavior and dealing with outlier behavior as "incompetence." Artificial goals and standards are not the thrust of CQI. Rather, the "standard" is to place oneself in the position of customer/patient/family member and to provide the same concern, service, and careful judgment that one would expect if roles were reversed.

Quality assurance did not concern itself with improving the norm of clinical practice or institutional management. The emphasis was on complying with the standard, which meant continuing the norm. For example, health care centers are allowed by Joint Commission "standards" to have a number of incomplete patient records equal to one-half the average number of monthly discharges. The following statistics from one health care center are typical [7]:

Average monthly discharges     1,744
Allowable incomplete records     872
Incomplete records, October      856
Incomplete records, November     834
Incomplete records, December     836

The schematic in figure 1, page 4, can be used to explain and illustrate the CQI emphasis, using physician performance as an example. QA focused only on "incompetent" performance (Q). CQI focuses on improving overall performance (R) and on developing data to define the sought-after physician (S). Of course, mechanisms are retained for dealing with outlier performers when they are encountered.

Point 2

CQI uses positive and supportive approaches to generate improved performance. Because quality assurance was primarily concerned with outlier performance, "problem-solving" actions were punitive and threatening. For example, in the context of a traditional medical staff, a legalistic procedure of "preparing charges," conducting "investigations," and holding formal legal hearings became synonymous with "peer review" and "quality assurance." External agencies tended to adopt the same punitive approach to health care centers. A congressionally commissioned Institute of Medicine study recently concluded that "PROs...focus on outliers rather than the average provider [and] use a sanctioning process that is largely ineffective." [8]

In contrast, CQI proponents suggest that supportive approaches can generate improved performance, personal pride, and interest in the success of the organization. One key to achieving this potential is to help managers and medical staff leaders learn interpersonal techniques that effectively combine firmness, fairness, forgiveness, and "how-to" suggestions so that the desired change in behavior is forthcoming. [9]

In the context of the traditional medical staff in a health care center, a first step must be to modernize the "corrective action" provision of medical staff bylaws. The trend is to reserve administrative/legal remedies for use with recalcitrant, unresponsive practitioners. The new emphasis is on "choosing a remedy," depending upon the nature and frequency of the physician performance problem encountered. [10,11]

Point 3

In QA, individual attitudes were considered too "soft" a concern to be important. All that mattered was the size of one's database. Organizational attitude (corporate culture) was simply a marketing buzzword.

In CQI, individual and corporate attitudes are as critical as valid, useful data. Is the management style of each executive and manager conducive to productivity through common effort? What are the stated goals of the governing body and the executive staff? Should the corporate culture even define reasonable profit-taking without exploitation? Candidates for senior management positions may encounter more questions, in the search and interview process, about their personal goals and beliefs. Some health care organizations may still choose executives purely on the basis of financial expertise. But others may look for additional qualities, such as recognition of the "social contract" accepted by those who choose to be in the health care business.

Point 4

Either "Quality" must be defined, operationally, or the word must be replaced by a more specific, more useful implementation guideline. The health care business can no longer afford the luxury of assuming that "quality" cannot be defined. In a philosophical, academic context, "quality" remains as elusive as ever. But a purely philosophical appraoch to quality will not serve the needs of health care executives, who need a list of specific parameters to confirm effective performance of organizations and the people in them.

For example, five qualities of a preferred health care organization might be stated as [12]:

* Appropriate care environment--technology, qualified and caring people, continuing education, communication.

* Corporate culture of providing dependable services in a businesslike manner, without exploitive profit-taking.

* Availability, including to those most in need.

* Accountability; willingness to share information.

* Efficient management.

Similarly, five qualities of a preferred physician might be stated as [12]:

* Clinical knowledge and skills, carefully applied to each patient .

* Cooperative attitude.

* Accessibility.

* Accurate, timely, and legible completion of patients' medical records.

* Efficient practice habits.

It isn't necessary to agree, in this early stage of methods development, that such parameters can be quantitatively measured. It's only necessary to acknowledge that descriptive aggregate data about each of these aspects of quality can be shared (to reassure those who depend on health care services and those who pay for them, and thus to regain public and political support) and can be sed internally as a basis for improving systems and helping individuals meet appropriate expectations.

Unfortunately, arguments about definitions of quality may continue to delay selection and implementation of CQI methods. In fact, the word "quality" has been so overused and abused that it may no longer be useful. Replacing "quality assurance" or "quality assessment" with "continuous quality improvement" is helpful, but it may not be enough to clarify needed changes in attitudes and behavior. Performance measurements (such as "performance indicators" rather than "quality criteria" and "performance data" rather than "quality review") may be easier to define and implement.

Point 5

The separatist approach to quality assurance, utilization management, and risk management must be replaced by cooperative effort. A major result of CQI, properly implemented, will be learning how to make each other look good by sharing data, conclusions, and solutions to common problems.

Point 6

There will be growth in research designed to compare CQI methods, particularly research related to the use and abuse of data. Quality assurance lived two frustrating decades without adequate attention to determining, through objective research, which methods were best. When "12 audit studies" were required by the Joint Commission, with accreditation hanging in the balance, who was going to stray far from the audit study approach? With insurance companies insisting on the industrial model of risk management, who was going to risk pointing out that there are differences between the health care business and hard goods businesses, as well as similarities? With PSROs (now PROs) insisting that the main measure of quality was average length of stay, with hospital payment hanging in the balance, who was going to debate methods theory?

The CQI era will provide opportunities for those who wish to pursue research on CQI methods. Many questions must be answered, objectively.

* How should the cause-and-effect analysis of an incident or pattern be conducted?

* Is it still true tht only physicians are qualified to evaluate critical data?

* Is there a difference between judgments made by "clinical analysts" and those made in the traditional committee setting?

* Is it possible to develop "second-level criteria" to be used at the analysis step, as opposed to "first-level criteria" or "performance indicators," which simply create a list of data items to collect?

* Which "performance indicators" are most useful? How many must there be for a health care center? For a surgeon? For a nurse?

* Are indicators of aggregate institutional performance the same as, or different from, useful indicators of individual performance?

Point 7

Expect CQI methods to be less complex that QA methods. Quality assurance was a committee activity, so products were the usual products of committees: minutes, reports, statistics, plans, lists, etc. CQI emphasizes effective management, on a day-to-day basis, by responsible individuals who are well-oriented to what is expected of them, work in concert, and "document" using simple, streamlined formats.

In the quality assurance era, there was no relationship between volume of paperwork and level of performance. Everyone was responding to the same paperwork requirements, so every organization had its mountain of "documentation." When CQI is properly implemented, there is an inverse relationship between paperwork and performance. That is, the better the performance, the smaller the volume of paperwork required to confirm it. Only when there are problems with performance must time be spent with detailed documentation, reflecting problem-solving efforts and their results. Expect the data needs of CQI to be less than the voluminous QA data systems of the 1970s and 1980s. CQI systems will be sophisticated, but not complex.

Sophistication, without complexity, will also evolve through distinguishing two separate tasks. One is confirming and comparing aggregate institutional performance. The other is confirming and comparing individual performance. Outcome data, such as infection rates and mortality rates, may prove adequate for the first task but not for the second. [13]

Point 8

Medical staff functions, notably credentialing and peer review, must change to reflect CQI assumptions and goals. The CQI era will not tolerate the traditional pro forma, ineffective functioning of the organized medical staff. [14,15] Credentialing and recredentialing methods must be patient-protective and must never be economically contaminated. [16] Attention to qualifications and performance must extend beyond physicians to other health care practitioner and beyond the traditional hospital context to modern health care delivery models such as managed care plans.

The impact of CQI on peer review may be even greater. The emphasis will be on objective conclusions, translated into a language and format that is as clear to nonclinicians as it is to clinicians. Figure 2, below, illustrates one version of a "Questioned Incidents and Trends Analysis Sheet" currently in use in several health care centers. [17] This or some other method of recording clinical conclusions about individual performance in a computerized, analyzable manner is essential to effective accomplishment of "medical staff functions."

One understood, the conversion of peer review to "developing physician-specific performance data," with trending after the "because why" step, should please "quality physicians." For example, the traditional peer review analysis of a case has focused on the question, "Did the doctor make a mistake?" If not, there was no problem and no record was kept. If so, the negative assumptions of quality assurance/peer review/corrective action led to punitive measures and generated resistance to performance evaluation activity.

The analysis step in CQI is a "because why?" step. All contributing factors are examined, not just physician performance. As other admit the need to improve systems and performance, physicians may finally feel they can admit the need to improve, without fearing fatal (in the sense of the physician's reputation and livelihood) consequences. [18]

In addition, proper implementation of the "1,2,3,4,5" or similar method provides the mechanism needed to develop positive performance data. This feature must be carefully pointed out to physicians, because it is different from implementation of a similar "1,2,3,4,5" classification of cases by PROs. The difference is that the PRO's implementation assumes that the equation is:

Finding = Problem

Thus, as implemented by the PRO, even as "1" would be a degree of "severity." In CQI, the beginning assumption is:

Finding = Question

(Is this a problem or not? Why or why not?)

0 = No questions.

1 = After careful review of the stated finding(s), the performance is good, for the reason stated.

2 = Reasonable degree of controversy in a very subjective area of clinical practice.

Thus, tabulation of total 0's, 1's, and 2's is at least a primitive beginning toward defining reasonable and sought-after physicians and institutions (see figure 1).

In addition, this broader "because why" step offer physicians and organizational providers the opportunity to develop and refine information about the impact of public policy and patient factors on quality of care.

A major task is to help physicians understand why objective conclusions and the findings on which the conclusions are based, along with a brief statement of the justification for the conclusions, must all be shared. The reason, of course, is that there is no other way to generate confidence in the validity of these conclusions.

Physicians would rather present the conclusion, and say, "We're doctors; here are our conclusions; trust them." Physicians must be helped to understand that everyone expects as monthly statement from the bank. No one accepts a written communication, "Dear Depositor: We have done our internal accounting, and all your money is still here."

Point 9

In the quality assurance era, absolute confidentiality was a central concern. This is not surprising, given the provider-oriented nature of quality assurance; the development, by definition, of only negative information; the accompanying harsh, threatening "corrective action" steps; and a litigious climate. A major goal of continuous quality improvement is to provide information to payers, beneficiaries, employees, patients, family members, and the press. The public now seems to believe it is entitled to enough information to make intelligent to enough information between the practice of two or three physicians whose names are obtained from a physician referral service and to compare the aggregate performance of the two or three hospitals in the community.

A variety of information-display formats now allow sharing of aggregate information confirming effective provider performance, without providing the case-specific information sought by players in the malpractice tort lottery, itself so badly in need of reform.

Point 10

Some trends in medical staff bylaws and organizational structure are compatible with the CQI concept. Medical staff bylaws once dealt only with legal issues and the complex and convoluted internal political structure of the medical organization. Increasingly, bylaws now also reflect sound organizational theory and acceptance of social responsibility. In many health care centers, medical staff bylaws are now are more stable document, not in a state of constant revision. One reason is that details of methods and policies have been placed in a variety of bylaws-related rules, policies, and methods manuals.

There has been a major move away from sluggish management of medical staff matters through a bulky committee structure toward day-to-day responsiveness of medical staff leaders. Elected medical staff leaders in community health care centers are learning to work with medical directors/vice presidents of medical services. This trend promises eventual improvement in the executive function of the medical staff. Within a few years, large and cumbersome medical executive committees may give way to a physician executive, working with an elected chief of staff and a physician advisory panel.

Ideally, CQI attitude changes will infect physicians. Aggressive, adversarial behavior, sometimes generated by punitive and legalistic approaches, may become less common. Self-governance may finally be understood as "taking responsibility for yourself," rather than as a distracting legal fuss. Helping physicians understand "the givens," and the attitudes necessary to be among the chosen in the CQI era, will be a major challenge for physician executives.

Point 11

Continuous quality improvement is a major Joint Commission emphasis. Surveyors have already learned to look at paper documentation less as an endpoint and more as a brief record of substantive efforts, related to a geniune commitment to high-quality performance. But don't over-focus on Joint Commission requirements. The positive benefits of CQI will never be gained if one's approach to CQI is narrow:

* What does the Joint Commission require now?

* Which seminar should I attend?

* Which new data system should we install?

Focusing on the broader implications of CQI will result in the kind of system that the Joint Commission wishes to find in place, and functioning usefully, at survey time.

Summary

The promise of CQI is increased communication, trust, and productivity in the workplace. A reasonably expected spin-off is restoration of public confidence in, and political support for, health care providers. In three to five years, we should know the impact, if any, of the continuous quality improvement movement. Will CQI prove to be a major breakthrough, or just another "layered on" program? As the old song predicts, "It all depends on you." If CQI is not implemented with vision, the movement will simply join "I Dare To Care" in the graveyard of broken promises.

References

[1] "Quality of Professional Services." Accreditation Manual for Hospitals. Chicago, Ill.: Joint Commission on Accreditation of Hospitals, 1976, pp. 27-8.

[2] "Quality Assurance Standard." Accreditation Manual for Hospitals. Chicago, Ill.: Joint Commission on Accreditation of Healthcare Organizations, 1988, pp. 235-8.

[3] Darling v. Charleston Community Memorial Hospital, 211 N.E. 2d 253, 1965.

[4] Johnson v. Misericordia Community Hospital, 294 N.W. 2d 501 Wis. Ct. App. 1980 aff'd., 301 N.W. 2d 156 (Wis. 1981).

[5] Jackson v. Power, 743 P.2d 1376 (Alaska 1987).

[6] Shortell, S., and McNerney, W. "Criteria and Guidelines for Reforming the U.S. Healthcare System." New England Journal of Medicine 322(7):463-6, Feb. 15, 1990.

[7] Personal communication.

[8] Lohr, K., Editor. Medicare: A Strategy For Quality Assurance, Vol. I, Washington, D.C.: National Academy Press, 1990, p. 3.

[9] Thompson, R., and Thompson, D. Productive Confrontation: What, When and How. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1987.

[10] Thompson, R., and Springer, E. Answers to Twenty New Tough Questions About Peer Review, Credentialing, and Medical Staff Bylaws. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1988, p. 96.

[11] Thompson, R. "Kinder, Gentler 'Corrective Action': Easing Physician Fears While Simultaneously Improving Effectiveness of 'Peer Review." North Carolina Medical Journal 50(9):502-7, Sept. 1989.

[12] Thompson, R. The Board Member's and CEO's Practical Guide to Medical Staff Structure and Responsibilities. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1989, p. 7.

[13] Thompson, R. Next Steps in Implementing Quality Improvement. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1990, p. 11.

[14] Eisele, C., Editor. The MEdical Staff in the Modern Hospital. New York, N.Y.: McGraw-Hill, 1967, pp. 3-4.

[15] "Survey Statistics." Perspectives, 10(3/4):13, March/April, 1990.

[16] Thompson, R., and Haddad, L. The Medical Staff Orgnization: 1990's Answers to 20 Tough Questions. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1990.

[17] Thompson R. Next Steps in Implementing Quality Improvement. Dunedin, Fla.: Thompson, Mohr and Associates, Inc., 1990, p. 58.

[18] Hilfiker, D. "Facing Out Mistakes." New England Journal of Medicine 310(2):118-22, Jan. 12, 1984.

Richard E. Thompson, MD, is President of Thompson, Mohr and Associates, Dunedin, Fla. (813/937-0170). He is the author of a new College monograph entitled Keys to Winning Physician Support: A guide for Executives and Managers.