By Ginny McPartland
Fifth in a series
If the 1970s and 1980s brought a quality of care frenzy, circumstances in the 1990s conspired to create a veritable quality tsunami. Health care leaders reacted dramatically to a 1989 paper by industrial quality guru Donald Berwick and began to second guess health care reformers in Washington. Realizing their survival was at stake due to market pressures and government and employer demands for quality data, physicians and other health plan leaders dove head first into the quality quagmire.
Berwick’s message was revolutionary. His call to action was to take away the punitive side of quality review and bring all medical disciplines into a discussion of how to improve care processes and thus ensure better quality. Berwick’s influence was to abolish “The Theory of Bad Apples” in quality assurance and replace it with “The Theory of Continuous Improvement.”
Berwick, a physician in the Harvard Community Health Plan, had gotten his inspiration from Japanese industrial quality experts. Their carefully defined philosophy called “kaizen” led Japan to high quality factory production success. “An epigram captures this (Japanese) spirit: ‘Every defect is a treasure.’ In the discovery of imperfection lies the chance for processes to improve,” Berwick wrote in the New England Journal of Medicine Jan. 5, 1989 edition.
He adds: “Quality can be improved much more when people are assumed to be trying hard already, and are not accused of sloth. Fear of the kind engendered by the disciplinary approach poisons improvement in quality, since it inevitably leads to disaffection, distortion of information, and the loss of the chance to learn.”
Flawless care requires support for decisions
Physicians rely on a support system to carry out high quality health care, Berwick noted. “Flawless care requires not just sound decisions but also sound supports for those decisions . . . In hospitals, physicians both rely on and help shape almost every process pertaining to patients’ experiences, from support services, such as dietary and housekeeping, to clinical care services, such as laboratories and nursing. Few (processes) can improve without the help of the medical staff.”
Berwick urged significant investment in assessing and improving quality of care. “In other industries, quality improvement has yielded high dividends in cost reductions that may occur in health care as well. . . The most important investments of all are in education and study, to understand the complex production processes used in health care; we must understand them before we can improve them.”
With Berwick’s message ringing in their ears, Kaiser Permanente leaders jumped into high gear. In September of 1990, the Kaiser Permanente Committee embraced Berwick’s bold new approach, which had been dubbed “TQM,” for Total Quality Management. At the same time, the Joint Commission on Accreditation of Hospitals also adopted TQM principles in its reviews.
Vohs and Lawrence put force behind the Quality Agenda
The 1991 Annual Report, titled the Quest for Quality, was devoted to chronicling the TQM phenomenon and explaining its genesis and hope for the future. The report acknowledged the work done previously by pioneers Sam Sapin, MD, and Leonard Rubin, MD, but declared the need to step it up:
“The role of quality assurance historically has been to identify problems within the system,” explained Susan Leary, director of quality assurance in the Program Office. “But with TQM,” she says, “we’re given new empowerment to go out and get involved in the planning process and to make system-wide changes once those problems are identified.”
A 53-page binder produced at the launch of the Quality Agenda defined the campaign as “A Roadmap for the Future.” The guide was to speed up KP quality improvement efforts and to intensify efforts to share good ideas and innovations across the regions.
The campaign emphasized the need to get all employees to understand and take part in quality initiatives. The roadmap outlined five specific tasks: 1) creating of a positive work environment; 2) measuring what we do; 3) improving what we do; 4) developing new approaches; and 5) telling our story.
KP’s first program-wide Total Quality Management conference
In 1992, the first annual interregional conference on Total Quality Management featured workshops conducted by Don Berwick, MD; Brent James, MD, renowned quality improvement expert and statistician of Intermountain Healthcare in Utah; and David Eddy, MD, the man who invented the computer model that could compile a wide range of health data and simulate a realistic clinical situation.
Eddy was hired by the Southern California Permanente Medical Group in 1991 to use clinical research data to evaluate the benefits and harms of different clinical interventions. Eddy compiled actual patient outcomes and ascertained which treatment would likely bring the desired results.
“One treatment (for lower back pain) might have a 30 percent chance of returning a patient to work, while another has only a 10 percent chance,” Eddy said. “But the first treatment might have greater risks. How do we decide if the greatest benefits of the first treatment are worth the risk? To determine this, we’d like to ask patients what they prefer. They’re the ones who will live or die by these decisions.”
Physicians need help synthesizing complex medical research
In 1993, Lawrence published the Quality Agenda in Action, a report on quality initiatives across the program. Highlights included KP’s work with HMO groups and six large employers to develop the HEDIS (Health Employer Data and Information Set) quality measures and specific data collection methodologies for various treatments and preventive screenings.
The update also called out the Northern California Permanente Medical Group for its own launch of a program to establish best practices, vetted through research, and to implement and evaluate them. The Interregional Nursing Task Force brought together nurses from all regions to set up a system of best nursing practices.
A five-year study conducted by Southern California concluded that normal childbirth after a Caesarean section was possible and safe; while another study of 2.5 million patients in Northern California showed that screening for rectum and distal colon cancer with sigmoidoscopy decreased the rate of death from these conditions by 60 to 75 percent. In all, more than 350 TQM projects had been launched across KP’s 12 regions in the four years prior to the publication of Lawrence’s report.
Next time: How do physicians know they’re doing the right thing?
By Ginny McPartland
Fourth in a series
How good was the problem-based approach to quality assessment developed in the early 1970s by Kaiser Permanente’s pioneers Len Rubin, MD, and Sam Sapin, MD? Soon after its unveiling, the Comprehensive Quality Assurance System was to be put to the test.
In 1979, at the behest of the federal Office for Health Maintenance Organizations (HMOs), the first incarnation of the National Committee for Quality Assurance (NCQA), was formed. Sponsored by the Group Health Association of America and the American Association of Foundations for Medical Care, the committee invited Sapin and Rubin to join.
In short order, the committee adopted Rubin’s problem-focused review method. NCQA’s emphasis was on identifying and correcting problems, and traditional audits were not required, reported Sapin who served NCQA as a board member and surveyor from 1980 to 1987.
Sapin and Rubin knew the review method worked because they had used it to evaluate KP care in both Northern and Southern California. The KP scheme had two levels: first, identifying possible trouble spots by a variety of means and judging the problems according to 56 monitoring criteria; and second, to fix the problem through process change.
“During the 1980s, these regular reports appeared to generate more quality assurance activity than did the previous classic (traditional) medical audits,” Sapin said.
National quality group loses financial support
NCQA floundered in the early 1980s due to the withdrawal of financial support. “NCQA’s status is presently precarious unless the parent organizations, the HMOs which are surveyed and some of the states, provide funds for its operation,” Sapin reported to the KP board of directors in 1983.
Even though member HMOs and the Office of HMOs inWashington, D.C., were satisfied with the surveys, there was an undercurrent pushing for a review agency independent of HMOs. James Doherty, CEO of the Group Health Association of America for 15 years, said in 1996, “HMOs needed to subject their operations to external review by an independent quality assurance body.”
NCQA regenerates and launches renewed mission
In 1990, the NCQA managed to get funding to reconfigure as an independent agency with a $308,000 grant from the Robert Wood Johnson Foundation and matching funds from HMOs. The board was reconstituted to have 20 members, the majority representing purchasers (largely employers) of care, health plans or consumers.
Six physicians, including four medical directors of managed care plans, and Dr. Thomas R. Reardon, a trustee of the American Medical Association, also served on the new NCQA board in the 1990s, according to a 1996 New England Journal of Medicine (NEJM) article by John K. Iglehart, then NEJM national correspondent and editor of the Health Affairs journal. (Iglehart was KP’s vice president of government relations in Washington, D.C., from 1979 to 1981.)
The author notes, “Although strong ties still exist (with managed care leaders), the NCQA is a conduit through which employers apply pressure on health plans to continually raise their quality horizons. This pressure creates a tension that reverberates throughout the NCQA’s relationship with health plans.”
Consortium hammers out first HEDIS measures
With the reconfigured NCQA, Kaiser Permanente and six other large employers went to work to fashion quality performance measures. These measures, which cover inpatient and outpatient care, would come to be known as HEDIS or HMO Employer Data and Information Set.
In the 1993 Quality Agenda in Action report, KP CEO Dr. David Lawrence wrote: “HEDIS is the basis for. . .a national effort of 30 major managed health care plans and a group of consumers and business representatives. . .to develop a system that will enable (purchasers) to compare health plans on the basis of quality indicators.”
NCQA released its initial set of quality measures in 1991, and about 330 health plans measured their performance according to the HEDIS system and reported their results to employers, Iglehart reported in his NEJM article.
He wrote: “The NCQA standards are evolving. . .A recent version (HEDIS 2.5) incorporated more than 60 performance indicators that cover quality of care, access to and satisfaction with care, the use of services, finances and management. Most indicators, however, assess administrative performance or utilization rather than quality of care.
“The nine quality measures focus on process, particularly the use of preventive services, which can be readily measured. Only two indicators measure a health outcome (low birth weight) or a proxy for a health outcome (hospitalization rates for patients with asthma),” Iglehart wrote.
On its Web site today, NCQA touts its HEDIS system as the industry standard for comparison of health care providers. “HEDIS allows for standardized measurement, standardized reporting and accurate, objective side-by-side comparisons . . . We work to make sure that all measures address important issues, are scientifically sound and are not overly burdensome or costly to implement.”
Examples of current HEDIS measures include: Advising smokers to quit; antidepressant medication management, breast cancer screening, cervical cancer screening, children and adolescent access to primary care physician; children and adolescent immunization status; comprehensive diabetes care; controlling high blood pressure; and prenatal and postpartum care.
Next time: 1990s begin with supercharged KP quality agenda
By Ginny McPartland
Third in a series
As the 1970s drew to a close, physicians and quality reviewers nationwide were probing and struggling to make a faulty “medical audit” system work to evaluate and improve the health care of millions of Americans, young and old, well and sick, rich and poor.
In three Kaiser Permanente regions, physicians and quality auditors had received federal grants to apply the medical records retrospective review or traditional method endorsed by state medical associations, the Joint Commission for Accreditation of Hospitals (JCAH), and Medicare and Medicaid officials.
Southern California, Hawaii and the Northwest KP regional medical groups obtained three-year grants under the EMCRO (Experimental Medical Review Organizations) project to put the traditional medical audit system into practice and find out how well it worked. It didn’t.
Northern California quality pioneers forge their own model
In Northern California, quality guru Len Rubin, MD, PhD, had taken a different path. “I think they (Northern California) were a bit smarter than us (SCPMG),” admitted Sam Sapin, MD, KP Southern California quality leader in the 1970s and 1980s.
As early as 1973, Rubin was piloting a new quality assessment method in all 13 of the Northern California KP medical centers. Rubin’s system, called Comprehensive Quality Assurance System (CQAS), had reviewers checking medical records of patients who had just been discharged from the hospital. The emphasis was on finding deficiencies in real time and auditing problem cases for questionable practices that may have contributed to bad outcomes.
“He (Rubin) really pioneered the problem-focused approach to quality of care assessment and assurance,” Sapin said. “His motto was and is, “Find out what’s wrong, not what’s right, then fix it.” Rubin published his CQAS protocol in 1975 for the American Group Practice Association.
Best to measure process or outcome?
In developing his model, Rubin was acutely aware of the difficulty of defining the relationship between process – delivery of care and drug administration – and the outcome of those treatments. Rubin argued that there are many “outcomes” in the continuum of care and many factors don’t affect the ultimate outcomes of good health, ability to return to work, or at the other end of the spectrum death or chronic illness or disability.
“As can be seen, there are endpoints here (or outcomes) relating to many departments (laboratory, record room, physician, pharmacy, patients, etc.). Often the outcome of one process is the input for another,” Rubin wrote in his 1975 CQAS publication.
“Further confusion lies in the time-dependence of outcome. The outcome may differ enormously, depending upon whether it is being measured immediately after treatment, several months after treatment, or several decades after treatment,” he continued. “Latent drug effects only now becoming known completely invalidate previous judgments that some outcomes were good in the long term. He concluded: “There is no way to measure ‘ultimate outcome.’ ”
Different schools of thought cloud issue
Sapin reported that at this time debate was raging between the “outcome” and the “process” people. “The process people say outcomes are too difficult to measure and interpret correctly; one should only compare outcomes of cases of comparable severity and one should also take into account all of the intervening variables (e.g. patient compliance, income level, lifestyle) which cannot be controlled by the provider and yet will affect the outcomes.
“On the other hand, the outcome people say don’t bother to assess process because many processes do not correlate well with outcomes and some may even lead to bad outcomes. The safety and efficacy of many of the things we do to and for patients have never been scientifically validated,” Sapin said in his 1983 presentation to the Board of Directors, “Historical Perspectives on Quality of Care.”
In 1973 as Rubin tested and refined his method, Jim Vohs, KP’s CEO, set up the KP board of directors’ Subcommittee on Correlation of Quality of Care and established a new department to focus on quality. Vohs also established the Interregional Quality Assurance Committee (IRQAC), with quality representation from all KP regions.The group came together at least once a year to compare quality notes. Giving program-wide quality a high priority, Vohs accompanied surveyors on all the committee’s facility visits in the regions.
Rubin’s system had an almost twin at the state level
Ironically, in the early 1970s the California Medical Association (CMA) and the California Hospital Association (CHA), responding to the rise of malpractice insurance rates, devised a “problem-focused” or “generic-screening-criteria” method similar to Rubin’s. Even though both agencies required the traditional medical audit for their member organizations, they decided to use a different method to identify outcomes that could become subjects of malpractice lawsuits.
The CMA-CHA method called for the review of medical records upon a patient’s discharge to identify adverse outcomes. In reviewing charts, they measured quality according to 20 standards to find problems. Examples of the screening criteria were unplanned removal of an organ, repeat of an operation during the same hospital stay and development of a heart attack after admission. CMA-CHA’s “Medical Insurance Feasibility Study” was published in 1978.
By 1980 the traditional medical audit had become a dinosaur in the world of quality assessment and assurance in favor of a problem-focused method.Review organizations move away from medical audits
In 1980 JCAH quietly abandoned the traditional medical records audit. JCAH’s new published standards required hospitals to have a documented quality assurance program but no specific audit program was mandated. Members of the federal Professional Services Review Organizations (PSROs), local agencies designated by Medicare and Medicaid to study quality of care, were also disappointed by the medical audit results and phased out the method.
In its 10th anniversary issue in 1984, the JCAH journal re-published Sapin’s 1980 groundbreaking article, “A Region-wide Quality of Care Monitoring and Problem Delineation Plan.” The article authored by Sapin, Gerald Borok, MPH, and Cheryl Tabatabal, rejected the traditional medical audit and described the problem-focused review scheme he and Rubin had developed and piloted.
In a 10-year anniversary reflective commentary, William C. Felch, MD, Quality Review Bulletin Editor and New York internist, recognized the new KP-generated system and touted the plan as “ambitious, carefully thought out and planned.” He added: “The question five years later is how did it all work out?”
Next time: Birth of the National Committee for Quality Assurance
Margaret “Maggie” Knott was an exceptional physical therapist who became world famous for practicing and teaching proprioceptive neuromuscular facilitation (PNF). This was a highly successful treatment modality for those suffering from severe physical impairment, many of them polio or rheumatic fever patients. In 1956 she and Dr. Dorothy Voss published the first textbook on PNF. Dr. Sedgwick Mead, then-director of the Kaiser Rehabilitation Center, described her as “One of the most extraordinary persons I have known.” PT Knott later became director of the Center. She passed away in 1978.
-Lincoln Cushing, Heritage staff writer