Naomi Morris, MD, MPH
Professor and Director
MCH Training Program
School of Public Health
University of Illinois at Chicago
2035 W. Taylor Street, M/C 923
Chicago, IL 60612
Manning Feinleib, MD, DrPH
Research Professor
Institute for Health Care Research & Policy
Georgetown University Medical Center
2233 Wisconsin Ave. NW, #525
Washington, DC 20007
Milton Terris, MD, MPH
Editor
Journal of Public Health Policy
208 Meadowood Drive
S. Burlington, VT 05403
Naomi Morris: The talk I’m giving this morning is called "Respect: Its Meaning and Measurement as an Essential Element in the Quality of Patient Care." This was really the first paper that had policy in it that I had ever written. It grew out of my concern for people in an age when there were all kinds of stories circulating about health care management organizations. So, I started thinking about the fact that no one really questions the right of people to be treated with respect by health care providers. Respectful treatment, in my mind, could be considered a basic human right. But, the potential for individuals to be dehumanized by personnel who are under pressure to process relatively large numbers of people as rapidly as possible is well known by clients attending big ambulatory clinics, especially those serving the medically indigent. Now, more and more of the medically indigent in regions V and VII are being served through managed care organizations (MCOs). In many MCOs patients have no control over the physician they see, the site of care, or the level of care. The reward system for MCOs or insurance systems in general favors the provision of less service and fewer referrals. I consider this situation another threat to the treatment of patients with appropriate respect. This can happen to the best of us. It really isn’t limited to the medically indigent. Measuring patient satisfaction is being added to HEDIS 3, which is the information required by the National Committee for Quality Assurance (NCQA) for those HMOs who want to be accredited by that particular body. It’s being added because it’s claimed to be a means of measuring the quality of a care provider that would be meaningful to other people.
However, I have studied patient satisfaction in the past, and I know that it can be very misleading. People who perceive little choice, like in a Medicaid HMO, or in a big hospital outpatient clinic, or if they feel powerless, which is typical in poverty populations, they may be apt to say they’re satisfied when they truly are not. They give the socially desirable answer in order not to prejudice the provider against them. Or, they may accept a less than desirable situation because they don’t think they’re going to find one any better. Or their other option is simply not to return for care they may need. And that’s why I think that in the evaluation of the quality of care, you have to go beyond a self-administered patient satisfaction measure to the objective assessment of how people are treated. Because I think health departments should be involved even more than in the past in monitoring services provided from public funds, I want to briefly share with you how to observe how people are treated in service settings. People from the state health department have always gone out to local health departments to sort of look over their records and get a feel for how that place is doing their job, so this is not a new idea.
Over twenty years ago I was part of a research team studying whether the new federally funded family planning projects were meeting their legislated goals. These happened to be projects funded by the federal Bureau of MCH. The team selected projects from 12 cities for study. Our project, known as FPEP, the Family Planning Evaluation Project, had several experiments in the design. In what I show you today, FPEP just provides the context for demonstrating the use of an instrument that we developed, called the Clinic Observation Record (COR). It is intended to measure respectful treatment. We expected patients who were treated with respect to use the services more appropriately. Our measurement objective was to compare the differences among clinics and programs in the amount of respect shown to patients. Results were to be used to improve the performance of those clinics as well as for evaluation.
The method: Initially, it was difficult for us to operationalize the concept "respect." We visited clinics, we asked patients questions, and we thought about our own experiences. We concluded patients are different, circumstances are different, and the same type of treatment doesn’t please every one to the same extent. Patients often said they were satisfied with services the team felt were inadequate. We finally defined respectful treatment as behavior which accords the patient high status in a sociological sense.
There were two observable aspects in a health care setting: staff behavior in the one-to-one interactions, and the environment in which the visit took place. We therefore created a two-part Clinic Observation Record. Part One was largely a checklist of behavior by one staff member with one patient described in interaction categories. Each clinic staff member was observed with five different patients. From five observations for every staff member an overall picture of clinic operation was obtained, and overall pictures of different functions performed by various staff were derived. The first part of this clinic observation record has 24 items scored in a way to indicate the amount of respect observed in a scale from least, with a score of zero, to the most, which we gave one to two points, and infrequently, five points. We were not measuring the patient’s preference. This was all our own concept of respect that was used.
Part Two of the Clinic Observation Record looked at the fixed aspects of the clinic setting and was observed only once for any particular clinic. It included such things as space, seating, lighting, restrooms available, reading materials, piped in music, play areas for children, a drinking fountain, and foodsnacks available. Policy items were also listed, such as whether there were specialists present for high-risk clients, or were they only seen by medical students. What were the days and hours that the services were available? What type of teaching was done to the patients concerning contraceptive methods? What about the receptionist? Was her function as a barrier or as a patient advocate? And what were the services provided beyond family planning? This was all arranged in a checklist format.
The Clinic Observation Record (COR) was simple enough for a high school graduate to score consistently after about a week’s training. We assumed the scorer would not be a member of the clinic’s staff, but for the purposes of self-improvement the clinic could use this instrument and have their own person do it. For comparison with the COR we used a five-item fixed choice self-administered patient satisfaction scale which we called PATSAT. Each instrument was considered meaningful in its own right. They didn’t address identical things, but they were related. The PATSAT items included satisfaction with time spent, and we knew the exact time recorded both by the patient and by a person clocking the patients in and out of the clinic. The PATSAT items also included satisfaction with services in general, the doctor’s explanations, the doctor’s action toward the patient as a person, and the nurse’s actions toward her as a person.
The findings are revealed in four tables, and I’m going to show you some high points in the first three, and then the fourth one. The first one shows you items from Part One of the COR, with the proportions of respect characterizing these behaviors. The data come from the number of cities you see listed in the first column and the number of clinics in that city. Full respect would be 1.0, so the fractions you see are the proportion of 1.0. You can see that the way the patient was beckoned was scored: if they were waved in or called by their first name, or whether somebody said, "Number Three," these didn’t come out very high. With regard to permission, we thought it would be appropriate for the patient to be asked permission for an observer to be present during the process she was going through. That didn’t happen very often. The score for that item was .202. The next thing was whether the staff addressed the person face-to-face when talking to her. That didn’t come out very well either. These are family planning patients, and just about every one here knows what I mean about time lying in the stirrups waiting for something to happen. That item didn’t turn out as bad as some things, but it wasn’t wonderful either. The tone of voice scored pretty well. As far as the opportunity to ask questions, the opportunity was not too great, with less than half given the chance to ask questions. All questions patients did manage to ask were answered, so that came out with a high score. The exam assistance means someone else besides a doctor was there during an exam to be helpful to them, and it looked like there usually was. The explanation for what was going on during the exam only occurred about half the time.
The second table shows the average proportions of respect on Part Two of the COR characterizing clinics in different cities. This only shows you four cities, but if you read the article in Dr. Terris’ journal it will have more details on more clinics. You can see in the first two the percentage is lower than in the second two. There was a significant difference between the cities.
If you look at the third table, it shows you the average time spent in minutes in selected cities and the proportions of respect in Part One of the Clinic Observation Record, as well as the patient satisfaction scores for the same cities. You can see that in the cities that seem to have better satisfaction and better Part One scores, there seems to be a lower number of minutes spent in the clinic. That city that kept people in the clinic for 141 minutes had a great many clinics in it, and we all know waiting is one of the things people complain about a lot.
Table Four shows you the correlation among time spent in the clinics, Parts One and Two of the Clinic Observation Record, and the patient satisfaction score. Here we are looking at data from 34 clinics in nine cities. Part One correlated highly with the patient satisfaction score at .53, illustrating agreement between trained observers and the patients. The less time spent and the more comfortable the environment and the policies, the more satisfied the patient. The correlation between less time spent and Part Two suggest that some clinics do a better job all around. Concerning the patient satisfaction results, there were actually 1,173 respondents who filled out the patient satisfaction form. The doctor’s explanations and personal treatment accounted for the largest proportion of patient satisfaction with the clinic visit, with the nurse’s treatment contributing substantially to the score. Open-ended items on another form indicated waiting was high on the list of worst things that happened today. But, if you looked at these data using regression analysis with the two parts of the Clinic Observation Record and time as independent variables, it showed that items in Part One contributed the most, explaining at least 28% of the variance in patient satisfaction among clinics. Including time and Part Two brought the proportion to 36%. The treatment by doctors and nurses, especially with regard to what we define as respect, turned out to be more important than that waiting time variable.
We didn’t really study the referral process. It would take a more highly trained person than those we used to review the gatekeeper’s decisions, but this certainly needs to be done. It seemed to us that a mandated quality assurance process incorporating this and the observed measurement of respect with state health and human services involvement would be useful today. The public health sector charged with assuring services as a core function will have to inform itself as to deterrents to needed care, as well as to health outcomes under varying circumstances. To make this point another way, it is essential to measure health outcomes by source of care. That’s part of what the performance indicators are about too. They ultimately want to measure outcomes and how you get there. It is likely that, as in the past, not all the population will seek needed services. You don’t have outcomes in a service-derived data set for those who don’t utilize care. The health service system will only be improved by MCOs if the previously unreached proportion of the population seeks and receives the care they need. I believe that making the system more respectful of patients, as well as effective, efficient, and affordable, will help us reach that goal. Thinking again about those performance indicators and thinking about the growing size of MCOs, I think there is a challenge there for us because data will need to come from all the organizations that serve targeted geographic areas and sectors of the population.
The last thing I want to say is that if anyone is interested in seeing those instruments we used I would be happy to share them and I would be happy to help fix them up for other types of services. Thank You.
Dr. Terris: Good morning. The title of my paper is "The Urgent Need for Health Department Resources to Monitor and Regulate the Quality of Health Care." With regard to health department resources, as I listen this morning, 15 out of the 18 performance items dealt with infrastructure. It seems to me that unless the Congress and the administration put their money into public health services so that we have the infrastructure to really change things, then all of the work done filling out reports won’t mean too much.
To discuss the quality of care in the era of managed-care, which is the era we are facing now, it is essential to define the meaning of the term managed-care. There is a lot of confusion about it. It should not be confused with the care given by a "managing physician." That is a general physician who treats, refers, and monitors the care of a patient by specialists and different services at different facilities. Also, it should not be confused with group practice pre-payment plans, health maintenance organizations (HMOs), since the majority of physicians in managed care plans in the U.S. today are engaged in individual practice. Not in group practice, not in staff organizations, but in individual practices in their own offices. Managed care means that the health insurance plan receives either a fixed budget to provide care for a designated population, as in the group or staff model, or discounted fees, as in the independent practice associations and preferred provider organizations.
The origins of managed-care can be found in the characteristics of medical care insurance in the United States, which for the most part consisted of provider-sponsored plans such as Blue Cross/Blue Shield, and the commercial plans of insurance companies, both of which paid physicians on a fee-for-service basis. A small minority of the population was enrolled in group-practice pre-payment plans, which provided care at a fixed premium per person enrolled, the so-called capitation method of payment. These plans, such as Kaiser Permanente and Group Health Association of Puget Sound, employed salaried physicians and have been shown in the past to have about 40% lower hospitalization costs, and about 25% lower overall costs than individual practice fee-for-service plans, with equal or better results in terms of quality of care. I have 14 references to studies that deal with this issue.
The fee-for-service method of payment is primarily responsible for the enormous escalation of health expenditures in the United States. From $141 per capita in 1960, 5% of the Gross Domestic Product, to $3,621 per capita in 1995, 14% of the GDP. The consumer price index for all items rose from 30 in 1960 to 157 in 1996, an increase of 430%. It rose from 22 for all medical care in 1960 to 228 in 1996, an increase of 923%, more than double the increase for all items.
Since most physicians in the U.S. are paid by fee-for-service instead of salary, there is a great incentive to provide unneeded services. This incentive system also pushes up hospital costs, since many procedures are performed in the hospital. It is the physician who controls the cost and makes the decision on what needs to be done. Despite the doctors’ propaganda to the contrary, most patients don’t demand to be operated on or undergo tests. They meekly follow doctors’ orders. It takes a very rare person to do anything but meekly follow the doctor’s orders.
Take, for example, the spectacular rise in physicians’ total charges for Medicare patients from 1982 to 1987. They rose five years by 76%, from $15 billion to $27 billion. Did technology improve by 76% in five years? Did the older population expand by 76%? Did old peoples’ demands for care grow by 76%? More likely, doctors’ interests in higher incomes was responsible for this change. In this five-year period, surgical services increased by 85%, physicians’ visits by 52%, consultations by 127%, diagnostic x-rays by 133%, clinical laboratory services by 84%. How much of this was needed? Nobody knows.
In the United States most people receive their health insurance coverage with employment, with the employer paying part or all of the cost. Faced with enormous escalations in cost, employers mounted a determined and increasingly successful drive to reduce benefits and shift costs to their employees. They also turned to preferred provider organizations (PPOs) to obtain lower health insurance costs. Local business health alliances organized, and the Washington Business Group on Health, representing large employers had a membership of 160 corporations in 1990.
In 1986 the Leadership Commission on Health Care was formed. Its 34 members included nine officers of major corporations -- AT&T, IBM, the Upjohn Company, General Motors, Equitable Life Insurance Society, W.R. Grace & Company, USX, Equifax, and Rogers Holdings -- and one labor leader, John Sweeney, President of the Service Employees International Union, as well as the Blue Cross/Blue Shield Association, the American Medical Association, and a variety of academic, legal, and political leaders. The bi-partisan character of the Commission was demonstrated by the three Honorary Co-chairs, former Presidents Jimmy Carter, Gerald Ford, and Richard M. Nixon.
The commission's 1989 report, For the Health of a Nation: A Shared Responsibility, foreshadowed major aspects of the Clinton health reform platform. It proposed a "universal access" program, stating "This system builds upon the American tradition of providing private health insurance through the workplace. It is designed to encourage continued extensive reliance on that approach without mandating employers provide such coverage. Thus, the system preserves the pluralistic approach to health care financing apparently preferred by Americans." Actually, it did propose a play-or-pay approach, taxing corporations that didn’t provide the basic national level of coverage for their employees.
Missing in the Commission’s report was the emphasis on "managed competition" and "managed care" that emerged from the Jackson Hole group. Mary Jane England, who came from Prudential to head the Washington Business Group on Health began participating in the Jackson Hole meetings in 1991 because the private sector innovations she wanted to extend to public policy "were the new organized systems of care that big business was developing with large insurers. These new managed care plans integrated the financing and delivery of care while using sophisticated delivery systems to store medical records and manage claims. They represented corporate America’s vision of an efficient and rational medical system."
The methods of the Jackson Hole Group included, overwhelmingly, representatives of the insurance industry -- the Health Insurance Association of America, Aetna Life & Casualty, Kaiser Foundation Health Plan, Blue Cross and Blue Shield of Colorado, Prudential, Metropolitan Life, Cigna Corporation, Harvard Community Health Plan, New York Life Insurance, U.S. Health Care, Blue Cross/Blue Shield Association, and Travelers. Other corporations represented were the Washington Business Group on Health, Syntex Corporation, Carlson Holdings, Mobile Technology, and Medtronic, Inc. Provider associations included the American Hospital Association, American Medical Association, American Group Practice Association, and the American Academy of Pediatrics. In addition, there were several executives of foundations -- Pew, Hartford, and Mayo -- and a number of academic and other participants, of whom Alain Enthoven of Stanford, Paul Elwood of InterStudy, and Lynn Etheredge played important roles. Everybody was on this committee except the people of the United States.
Hecker has noted that "if any group had a privileged voice in the Jackson Hole group, it was the large insurers who were on the forefront of delivery system reform…To them, the future was in managing the delivery of care, not just paying for it…in managing risk, not avoiding it…" This position put the large insurance companies and managed-care conglomerates at odds with the great bulk of commercial insurers which not only profited from the status quo, but lacked the capital and expertise to develop managed care plans. Indeed, the Jackson Hole group was the expression of a gathering intra-industry struggle between insurers that handled large corporate accounts and those that operated in the small group market. This conflict came to a head in mid-1992, when Cigna left the Health Insurance Associaton of America, a politically powerful trade organization, which generally reflects the position of small claims insurers. Cigna was followed by Aetna, Metropolitan Life, Travelers, and most recently, Prudential.
The big five pulled out. These five largest private insurers, "the gang of five," formed their own lobbying group to support the approach of the Clinton administration. In an article in the New York Times on May 10, 1993, "The Changing Definition of Health Insurers: Health Insurers Bcome Health Care Providers," they are quoted as saying, "Many traditional indemnity insurers that have looked to insure only the healthiest customers would hardly be missed." This is underscored by "Alain C. Enthoven, a Stanford economist and one of the architects of the type of health-care change embraced by Mr. Clinton, known as managed competition," who said: "The traditional indemnity health insurance industry is biologically extinct…These changes are already underway, it will just happen faster if managed competition comes into being."
On the other hand, the Health Insurance Association of America responded to the Clinton plan with a $10.5 million advertising campaign that featured the soon to be famous couple "Harry and Louise," criticizing various aspects of the President’s proposal. The struggle between large corporations and small business was not limited to the insurance industry. The New York Times reported on February 6, 1994 that "the manufacturers are divided over health care with the largest companies like the General Motors Corporation, Caterpillar, and General Electric leaning toward the Clinton plan." On the other hand, "the Clinton plan would require employers to pay more for health care than some of the small manufacturers now pay…‘Even if the Clinton plan held all costs to 7.9% of our payroll, as the President proposes, that would still be more than the 6% we as company now pay for health insurance,’ said Ronald D. Bullock, the president of Bison Gear and Engineering of Downers Grove, Illinois." Bison has 160 workers.
The National Association of Manufacturers, despite the power of the large corporations, voted not to support the Clinton plan in its present form. The U.S. Chamber of Commerce, representing tens of thousands of companies endorsed the alternative plan proposed by Representative Jim Cooper of Tennessee, which would require companies to offer the required health coverage to their workers, but would not require the companies to pay the premium. So, business failed to support the Clinton. The failure of corporate America to support the Clinton proposing mandated health care by small business to increase health insurance coverage was largely responsible for its demise. But Clinton’s embrace of managed competition gave new impetus to the development and expansion of managed care plans. Enrollment increased from 34 million in 1991 to 52.5 million in 1996, from 14% of the population in 1991 to 20% in 1996. In 1991, the group practice plans had the greatest number of enrollees, 50%. By 1996, individual practice associations had 41%. Mixed plans had 33%, and group practice plans had only 26%. During this period, managed care market share increased at the expense of indemnity insurance plans, whose share decreased from 53% in 1991 to 31% in 1995.
Now, the extraordinary and continuing growth of managed care has had a tremendous impact on the quality of medical care. Some observations are in order. First, the threat to quality in fee-for-service plans comes from the financial incentive to provide more services, more expensive services, unnecessary diagnostic and therapeutic procedures, and unnecessary, potentially harmful surgeries. In managed care, the opposite is true. The financial incentive is to provide fewer services and less costly services, resulting in the threat to quality coming from the failure to provide needed services, denial of claims for needed services, increasing staff-patient loads, pressure on physicians to prescribe less expensive diagnostic and therapeutic procedures, and pressure on hospitals to lower costs by substituting cheaper, poorly trained or untrained personnel for qualified nurses and technicians. On the one hand, there is widespread downsizing of nursing staffs in hospitals. On the other hand, managed care plans have begun to substitute nurse practitioners for physicians as providers of primary care. The key words of managed competition are: cut costs.
Second, not only do for-profit plans now dominate the field of managed care, the difference between for-profit and non-profit is disappearing, as non-profit plans in hospitals increasingly convert to for-profit status. The non-profits have begun to join their for-profit competitors in exhibiting behavior of corporate America in the post-Cold War era. Merger mania. The replacement of small independent local organizations by the branches of large national corporations, and the relentless drive for greater profits by downsizing, by firing large numbers of employees and hiring unqualified workers to reduce labor costs.
Third, the managed care plans are causing a major shift in the exercise of power control in the health care industry. In the pre-managed care period physicians were dominant and reaped the profits. Now, management is increasingly dominant, and management and shareholders harvest the profits. According the New York Times of April 11, 1995, the biggest pay package of a chief of a for-profit HMO, $15.5 million in salary and stock options, went to Norman Payson, M.D., president and chief executive of HealthSource. He was followed by Daniel Crowley of Foundation Health, $14 million, Roger Greaves, of Health Systems International, $9 million, Malik Hasan, M.D., $9 million, William McGuire, M.D. of United Health Care, $6.8 million, Leonard Abramson of U.S. Health Care, $4 million.
Fourth, the salaried group practice plans, which dominated the field in the Seventies and Eighties, now account for only 26% of managed care enrollees, as compared with 41% in individual practice associations, and 33% in mixed plans. This means that most physicians affiliated with managed care plans are still being paid by fee-for-service, which creates incentives to provide more services, while management demands they provide fewer service. Physicians in salaried group practice plans, on the other hand, have no incentives to provide unnecessary services. The studies comparing utilization and quality of services in managed care and fee-for-service plans were based largely on salaried group practice plans. They are no longer adequate to deal with the current situation where comparisons must also be made between traditional full price fee-for-service plans and their "discount houses," the non-salary managed care plans. The employer bargained with the discount houses, had them compete over who would give them the lowest price.
Fifth, traditional fee-for-service is still dominant in the U.S.; only 20% of the population is currently enrolled in managed-care plans. There is an urgent need to monitor and regulate the quantity and quality of health care provided by all types of plans. Nor can this be left to private agencies, which do not and cannot represent the interests of the public.
In 1989, the National Association for Public Health Policy, with which I am affiliated, adopted a policy statement, "A Progressive Proposal for a National Health Care System," which outlined the elements of such a program. "Federal standards are essential to improve the quantity and quality of medical care. They also serve to help equalize the services provided across the nation. These standards should be developed in consultation with the state and local health departments, not with the insurance companies."
"In addition to standards, specific measures for quality control and improvement are necessary. The services of all provider organizations shall be evaluated, and governmental action to improve them shall be taken on the basis of the following procedures:
A: Assessment of qualifications of health care personnel, not only in terms of educational background and previous licensure or certification, but through field testing of knowledge and performance.
B: Field inspection of availability and suitability of facilities and services.
C: Determination of compliance with performance standards through computerized reporting and statistical analysis of numbers and types of visits, diagnoses, clinical and lab procedures, hospital admissions, length of stay, etc.
D: Rigorous analysis of data on the effectiveness of care, including reduction in risk factors for the disease, degree of elimination of preventable diseases, avoidance of iatrogenic diseases such as drug reactions and hospital-acquired infections, reduction of operative fatality rates and case fatality rates, reduction of length and degree of illness and disability, and degree of rehabilitation of patients.
E: Supplementation of the statistical analyses through studies of samples of patient records by qualified investigators to determine the appropriateness and adequacy and quality of care.
F: Prompt and effective action by government agencies in response to appeals by Consumers Councils and Employee Councils for investigation and action on deficiencies in the treatment of patients.
G: Consultation, training, and other assistance by the government to remedy deficiencies and improve the quality of services.
H: Appropriate government regulatory action to correct outright abuses and gross deficiencies in the quality of care provided."
It is essential that the federal, state, and local health departments take steps now to expand their database in order to provide the necessary information to monitor and regulate the quantity and quality of health care received by the public. This will require adequate funding and both political and technical leadership which at this point is lacking. It should become a major priority issue for the public health movement in the United States.
Thank you.
Note: The references used in this presentation are available in the published paper, Journal of Public Health Policy 19 (1998): 5-14.
Manning Feinleib: Good morning. I should start off by saying I’m from the federal government, and I’m here to help you. I’m also in academia now, and I’ll try to give you some insights from the academic perspective. First, I should say, regarding the definition of managed care, the definition I’ve been using is quite different than the one that Milton Terries has been using. I was there for a good part of the Clinton administration’s efforts to introduce a new health care system, and to me there were three fundamental ingredients to the managed care proposals. One was guaranteed access to care. That was an important buzzword at the time. Another was appropriate care, that we should have some way of monitoring that the care was appropriate. And then, cost efficiencies. Milt has pretty much convinced me that only one of those is valid, and that is the last one. He also pointed out that one of the reasons for this emphasis on the reduction of costs is that the fee-for-service system had promoted a system that was too fat. There were too many things being done at needless cost that just put money into physicians’ pockets, and were not being used very appropriately. If this was the case, there should be a lot of room to cut costs without greatly impairing the appropriateness of services. You may be aware that in the last four to five years there was an effort to improve professional guidelines in a variety of areas to get a better handle on what appropriate care should be. This also went afoul of the congressional process. The Agency of Health Care Policy and Research did almost too good a job showing that a lot of the practices that were being done in the country were not very effective or useful, that there was no evidence to support them, and that they should probably be discouraged. All of these things have come into play as we come into a transition period in what is essentially a free market economy to find out how we can provide adequate care for the people of the United States that is effective, not just a system for diverting the medical expenditures from the physicians’ pockets to the CEOs’ pockets. Somewhere in between we hope to get to a point that does ensure good care on a cost efficient basis. It is surprising how little we actually know what is good medical practice. It is very hard to document in the mode of evidence-based policies now what is the appropriate care for so many conditions that are treated in everyday life. I think we’ll see a lot more emphasis from NIH and others to try to get a handle on what kind of services, procedures, etc. should be offered.
There is already a great deal of data available from national and state sources on maternal and child health. Most of these are currently derived from the National Vital Statistics System, which is based on the routinely collected birth and death certificate data. I hope you are all familiar with it. These reports provide a great deal of information, for example, percent of mothers with first trimester prenatal care, or percent of mothers who smoked during pregnancy. This is also a routinely collected item of information at the state level, and you can be sure that our concern about large proportion of teen mothers who smoke during pregnancy will become of vital importance as we play out the scenario for the tobacco settlement. One of the key targets is going to be what happens among these teenagers, and that will affect policy for about the next quarter of a century with regard to the whole tobacco predicament we’re in.
Here is a list of indicators developed by many people in this room, who have been working on the model indicators development committee. You have developed a wonderful list of items with five major domains, with about 20 categories with items that should be obtained. The detailed aspects, which we will see in a little while, number around 200 items of specific information we’d like to incorporate into this data set. This will be a tremendous undertaking. I’d like to thank Joan Kennelly for providing me with the following part of this document, which I think will put us in good stead for the next couple of decades.
Ten years ago, when we were formulating the national health objectives to be achieved by the year 2000, we came up with a considerably shorter list of objectives. One of the essential criteria for selecting these objectives was whether we’d be able to track the accomplishments during the Nineties. Although we realized not all of the objectives had established data sets in 1990, we thought we could develop them early in the decade. This was not to be the case. In fact, it is only in the last few months that we came out with the final operational definitions for priority area 14 (Maternal and Infant Health), and I don’t think they’ve been agreed on yet in some of the other areas. Some of the data systems that were in place in 1990 are no longer available due to budgetary constraints, or changes in programmatic priorities. For example, the birth defects monitoring program was discontinued in 1994. This was to have been our major data set to monitor fetal alcohol syndrome, spina bifida, and other neural tube defects. CDC is now working with about 20 of the states to develop pilot programs to make up for this loss.
Since January of this year, the FDA has required that grain products in the United States be fortified with folic acid in order to prevent neural tube defects. It will be essential to public health and nutritional policy analysts to see whether or not the intended effect of this legislation is successful in reducing the number of neural tube defects by about 1,500 each year. So, we will need a pretty detailed surveillance system. In the meantime, we will probably have to rely on birth certificates to do this, and this is not a very good monitoring or surveillance system. You will notice that for two of the objective areas, 14 and 16, there is no data source available. For some of these objectives, we are still in a very sad state. However, for many of these objectives, we can still rely on the vital statistics system and several national surveys that are designed to poll appropriated random samples of the population to fill the need. Whether or not all of these surveys can be operated at the state or local level still remains to be determined.
This report released by the MCH Model Indicators Working Group offers what this particular group thought should be the model indicators. A great deal of effort has gone into this, and I’m very glad that the states have taken over the leadership of this, because I must say the federal establishment has not been very supportive financially, and has not provided the best leadership role. I know the states have the feds outnumbered by 50 to one, and they have done a fantastic job of taking the initiative.
Here is an idea of what are called the core indicators, about two dozen of the most essential items with which every state should be involved. The first category is disease and injury morbidity. You see some of the conditions about which they are concerned. They fall into a number of areas. Here we have items related to vaccine-preventable diseases such as measles, and other preventable infectious diseased such as gonorrhea and HIV. Items such as anemia prevalence, overweight prevalence, detecting lead poisoning in children, and hospitalizations for asthma. Some of the other items, including incidence of invasive cervical cancer and attempted suicide rates are also indicators. In the area of mortality data, they concentrated on a variety of measures that are readily available through the death certificates. These should be available at the state and local level. Some of the others, particularly those related to perinatal issues, include morbidity and percent of low birth weight infants. The tobacco dependence rates will obviously be an essential issue that will be mandated through any settlement regarding the tobacco problem, and hopefully funds will be available to have significant programs at the state level. There is another level of recommended indicators with about 75 listed, and I won’t go over these now. Then there are about 100 indicators at an optional level which are included in the final report.
How will this data be disseminated and used once it is collected? I would like to tell you about a remarkable accomplishment I just became aware of just last week thanks to Naomi Morris. This is the website for the Maternal and Child Program at the Utah Department of Health. I would like to thank Dr. Wu Xu for sharing this information with me, and I will tell you in a minute how you can access this Internet site and get in touch with her. The acronym for this program is MatCHIIM, standing for Maternal and Child Health Information Internet Query Module. I think they’ve done a very impressive job. This is just the beginning of a process which will provide a lot of information about all of these topics, and here we see the first available indicator page is available. They provide information about low birth weight with nice graphics, including the Healthy People objectives, Utah and national rates, and them some information and goals, etc. One of the most remarkable things about this website is that you can go deeper and deeper in depth into any issue. This is a wonderful item to have readily available. It is the kind of information that will influence decision makers when they want to get detailed data, and I think, above all, the general public and the media who can be so instrumental to putting the pressure on the people who give the money to support these kinds of programs. This will require continuous funding support to create and maintain these items, as well as to update them on a regular basis. The web address is http://hlunix.hl.state.ut.us/matchiim/measures .
Now let us turn in the few remaining minutes to the main question I was asked to address. What are the state data needs to assess quality in managed care. I would like to look at this in two points of view. First, for what indicators do we expect appropriate performance of managed care organizations to make a difference? And second, what are the potential sources of data for these indicators? If we look at the 17 items for the Year 2000 objectives, I think that you will find that managed care should be having an impact on most of them. I had put down in my notes for this talk an optimistic approach, to say that good managed care with effective coverage and access to care, and good quality professional guidelines, should have a positive impact. In view of Dr. Terris’ remarks, I will now put it in a negative light, to say if we cut costs too much, which of these will be impacted negatively? It’s a half-full or half -empty proposition, up for debate. The point is, that if managed care is having an impact, if you’re an optimist you should say most of these things should be improved. If you’re a pessimist, cutting costs will result in deterioration. I think you will agree that if managed care is doing a good job, we should continue to see declined infant mortality. This should be readily available from our vital statistics data. I don’t know whether fetal alcohol syndrome will be impacted by managed care, whether it’s doing a good or bad job. The high-risk population is not heavily impacted by the type of health care delivery service received. Low birth weight should be improved with good managed care, but my observation over the last couple of decades is that we haven’t done a very good job of reducing this. On the other hand, if the managed care organizations are seeing the mothers regularly, we should see that weight gain in pregnancy should be pretty much to accepted targets and guidelines. There should be much less severe complications with pregnancy if they are doing a good job, and they should be interested in this because complications cost a great deal of money. Likewise, cesarean deliveries should be less prevalent. If you go through the other items, and I think you’ll find interplay of these effects, if you have good prenatal care and counseling services, early diagnosis of abnormalities, there should be cost savings involved. We should expect managed care organizations to take an interest in most of these things. Above all, we hope that in monitoring cost effectiveness, they will be able to provide data on the occurrence of these items and make them available to state health departments and researchers.
I pointed out several times that one of the most important sources of information is the vital statistics system which originates at the state level. Every ten years the standard certificates are redesigned. They are reviewed for input regarding how well birth certificate information is being gathered and used, and what change should be made. Here is a partial list of some of the suggestions for additions to certificates as we go into the year 2000. A couple of the items which may be particularly appropriate for the purposes of this session include item four, the primary source of payment for delivery, as well as item six, the type of practitioner providing prenatal care. If these items are routinely collected items on the birth certificate in the next century, and when they are coupled with a linked infant birth and death file, we should have a wonderful system for looking at the effectiveness of various types of medical care and payment coverage.
I should also mention that there will also be an effort during the next few years to have more and more states do an electronic birth certificate. All of these data should then be available more quickly and reliably. Going back to the list of model indicators, you can see that certain items have been highlighted. I hope states will give a great deal of emphasis to these areas so we can look at the impact. I don’t want to suggest now any special surveys. This is something we all recognize to be a problem, but I think these are essential areas in which to obtain data.
Finally, the area of primary care has to be better defined. I would just like to give you one short list to indicate what this is and what might be done. This is a list of what’s often called ambulatory care sensitive conditions, conditions that, if treated effectively in the early stages, do not require hospitalization. We should not be seeing these in the hospital, and they should not be showing up on death certificates. These are the kind of conditions that I think, if managed care organizations pay attention to them, it will save money, and they can prevent some of these conditions. You shouldn’t see kids coming in with ruptured appendices. You shouldn’t see diabetics treated inappropriately. Kids should be getting their shots so they don’t have to be treated later on. Asthma is now a target area. Effective treatment in an ambulatory setting should preclude hospitalization. I think managed care organizations will take an interest in lists like these, but from our point of view we hope they will obtain data which will help us to monitor their progress.
Thank you.