THE HASTINGS REPORT: JULY-AUGUST 1998

SPECIAL SUPPLEMENT

What Could Have Saved John Worthy?

This case study is one product of a two-year research project, "Value Perceptions and Realities within Managed Health Care," carried out by The Hastings Center with funding from The Prudential Foundation.

Copyright (c) 1998 The Hastings Center

Reproduced with permission of The Hastings Center

PREFACE

The advent of "managed care" has significantly altered how health care is delivered in the United States, The generic term managed care covers a wide territory, encompassing tools for allocating health care resources such as utilization review and even an emerging philosophy of health care as well as a great variety of different health care organizations. Many of the ethical dilemmas that arise in managed care are shared with fee for service medicine. To the extent that change has given rise to truly new ethical challenges, those challenges arise in the way managed care reorganizes the delivery of health care services, not around individual patient-professional relationships, but around sites of care (whether the offices or network practitioners, clinics, or hospitals or other facilities) and formalized relationships among a range of providers. Moreover, "managed" caregiving is organized in explicitly bureaucratic hierarchies in which information management is systemwide" at the level of the health plan, and monitoring of clinica1 and administrative practice is ubiquitous. All on the foundation of a philosophy of care that, however well or poorly articulated, responds to the needs of individual patients in the context of population-based

One of the most challenging features of this reorganization is that it prises apart decision making roles and responsibilities that fee for service health care largely vests in the privileged relationship between doctors and patients. Managed care integrates other parties into decisionmaking much more directly--including those who purchase health care services and design benefits packages and coverage policy, and those who administer the delivery of services, and those who regulate health care. If we are adequately to address the ethics of managed care, then it is crucial that we understand has "a decision" about clinical care is in fact distributed among many decisionmakers and significantly shaped by a constellation of other decisions taken throughout the system. If we are adequately to address the ethics of managed care, then it is crucial that we understand how "a decision" about clinical care is in fact distributed among many decisionmakers and significantly shaped by constellations of other decisions taken throughout the system.

Ethical analysis grounded in the patient-doctor relationship, though important, cannot well address this systemic dimension of managed care. What is called for is a richer ethics of organizations, including an understanding of the goal(s) a health care organization is intended, or expected, to serve; the different tasks, such as determining how and by whom health care services will be delivered, that structure a health care system, and the values, loyalties, and habits of thought appropriate to those tasks; and the organizational structures and practices that promote or inhibit excellence in carrying our those tasks. The case study presented here begins the process of exploring these systemic issues.

This case study is one product of a two-year research project, "Value Perceptions and Realities within Managed Health Care," carried out by The Hastings Center with funding from The Prudential Foundation. The project working group based its analysis in part on members' expertise and the published literature in ethics and health policy. Importantly, it also drew on interviews with clinicians, patient-members, and corporate level staff in a variety of managed case health plans who shared with us their experiences facing the new challenges of health care. °

WHAT COULD HAVE SAVED JOHN WORTHY?

PROLOGUE

One evening John Worthy, age forty-seven, brought home information

about the new health insurance plans his employer, Factory Inc., was offering. Factory had decided to offer a choice of health plans: GoodCare, a managed cane plan, or GoodCare Prime, which had a point of service option. Those who wanted to pay more for it themselves could stay with the company's old indemnity insurance.

After dinner, John and his wife Jane looked over the materials and talked about which plan they should choose. John and Jane were in pretty good health, and their younger daughter was rarely sick. Their older girl was covered by the health service at college. GoodCare's HMO had no deductible, only a small copay for office visits, and there were several doctors in their area on the plan's list of providers. They decided to sign up with GoodCare. They filled out the enrollment form and chose a primary cane provider whose office would be convenient to get to; John took the papers back to work the next morning.

Two weeks later they received their member cards and handbook from GoodCare, along with a letter encouraging them to make a "get to know you" appointment with the family doctor they'd chosen. There was no copay for the initial visit, but with one thing and another the Worthys didn't get around to scheduling appointments.

CASE

One morning two months later, John Worthy woke up with a headache.

When Mrs. Worthy called at 10:30 to say that John was experiencing a rather severe headache that came on suddenly a couple of hours previously, FRAN DAVIS, John's new family doctor, was concerned. Unfortunately, Dr. Davis was in the middle of office hours with a very full waiting room. The physician and the Worthys were still strangers to one another, and neither the Worthys nor Dr Davis were very familiar with the HMO's rules and procedures, since Dr. Davis's medical group was relatively new to GoodCare too.

A year ago the medical group, weary from battling payors and paperwork, sold their practice to Physician Management Services (PMS). In exchange for the practices physical assets, a long-term contractual commitment, and a percentage of gross revenues, PMS negotiates contracts with the many HMOs and other payers for whom their physicians provide services, takes care of all the paperwork, periodically upgrades the practices computer information system, and keeps the group within its budget. PMS receives a capitated fee from GoodCare for each enrolled patient to cover all primary care, plus outpatient laboratory and x-ray services. Specialists and hospital care:, including emergency room visits, are covered separately by the HMO. Physician Management Services pays each physician a salary, but also transfers some of its economic risk to its physicians in thc form of bonuses and penalties tied to productivity and utilization. They established protocols to reduce excessive care, and set a 5-patients-per-hour productivity standard. Recently PMS has also been educating physicians about their tendency to send too many patients to the emergency room----a common response among physicians who are capitated For primary care, and whose waiting rooms have become crowded with managed care patients who pay little or nothing for office visits. Last year, high ER costs prompted GoodCare to begin assessing financial penalties on primary care providers who overused this resource.

Dr. Davis was aware that if GoodCare decided, after the fact, that John Worthy did not truly require emergency care the patient, or even she as the primary care provider, could be financially at risk. Nevertheless, because she was unfamiliar with Mr. Worthy's medical history, overwhelmed at the office, and somewhat concerned by his symptoms, Dr. Davis suggested that Mrs. Worthy take John to the nearest emergency room if his headache really seemed that severe.

EDWARD R. POST, the emergency room physician, was concerned. When Worthy presented a little before noon, his temperature was normal, and blood pressure was 158/96. Funduscopic examination showed no indication of elevated intracranial pressure. John said he'd had severe headaches in the past, but none for maybe the last two years. He acknowledged being a long-time heavy smoker and said he seemed to feel some pain behind one eye. During the examination he also said that his neck might feel just a little stiff; because his headache was so bothersome, it was hard For him to figure out exactly what hurt. Dr. Post could not directly feel any neck rigidity, but believed that given the severity of the headache as John reported it be should have further tests. Post explained to the Worthys that the problem was most likely just a cluster headache, a severe tension headache, or sinusitis; he did not believe that John had meningitis. But he indicated that there was a possibility, although very remote, of early bleeding in the brain, a tumor, or some other much more serious problem. And so he strongly urged John to undergo a CT scan. The Worthys asked whether the expenses would be covered by their health plan. Post replied that this hospital was not affiliated with GoodCare, and so he could not offer any assurance about reimbursement. Nonetheless, he urged the Worthys to stay for the further diagnostic evaluation.

The Worthys talked things over, but decided to leave the hospital because they were afraid the proposed tests would not be covered. Dr. Post couldn't justify insisting that John stay for tests. He suggested that since their primary care provider was busy they consider calling a neurologist in their heath plan, The Worthys thought about visiting the in-plan hospital at a considerably greater distance from their home, but opted instead to return home, from where Mrs. Worthy phoned GoodCare.

CONNIE S. ROGERS customer service representative at GoodCare HMO, was concerned. She explained to the Worthys that if John visited the emergency department or entered the nouplan hospital near his home, and if it was nor declared to be a medical emergency, the entire hospital visit could be disallowed. The Worthys recalled Post's suggestion to see a neurologist and asked if the HMO had any available. Rogers replied that GoodCare's neurologist was a Dr. Newman, and gave the Worthy: his number. The Worthys phoned, but were told by Dr. Newman's receptionist that they could not have an appointment until they first received an authorization from their primary car provider. Jane Worthy, increasingly frustrated, phoned Dr. Davis to secure the necessary authorization, unfortunately, Davis's phone line was busy, and calls had been temporarily transferred to an answering service. Jane left a message but, unsure when the message would reach Davis, she phoned back to the HMO.

Connie Rogers again answered. Mrs. Worthy, furious almost beyond words insisted on talking with the president of the HMO. She was becoming panicky that her husband's worsening condition might bode something life-threatening, and was not at all sure that he could tolerate the eighteen-mile drive to the in-plan hospital. Sensing that litigation might be in the offing, Rogers indicated that although the president was not available, she try to reach GoodCare's medical director and would talk to him on their behalf. She then phoned the medical director and explained the situation as best she could.

MICHAEL DEPP, the medical Director of GoodCare, was concerned. As soon as he finished talking with Connie Rogers, he phoned Dr. Davis's office and got through promptly. Upon being updated on the Worthy's current plight, Davis indicated she would resolve the problem and phone Depp back with follow-up information, Davis first phoned the hospital emergency department and learned from Dr. Post that John had not appeared to be in an acute emergency condition, and that the Worthys had declined further evaluation for financial reasons. Dr. Davis inferred from the subsequent events that John Worthy's condition seemed to be worsening, however, and that prompt referral to a neurologist was medically justified, Dr. Davis was not familiar with Dr. Newman, and did not know whether he was as experienced as other neurologists she knew in town. Nevertheless, she wanted John to be seen expeditiously by someone who could provide all the care he needed. After phoning Newman's office with the necessary referral authorization, Dr. Davis then phoned the Worthys to inform them that they were clear to visit the neurologist.

Finally, Dr. Davis phoned Michael Depp with the promised follow-up and took the opportunity to vent her own frustration. Depp listened patiently and sympathized with her difficulties. He acknowledged that the "Mickey Mouse run-around," as she'd put it, sometimes did pose real inconvenience. But he explained that GoodCare had tried other forms of cost containment for many years with little result. Thus, they installed the primary care case management model, the preferred hospital and specialist list with discounted fees, and their utilization management system. Patients were still free to visit any provider, and to receive any care they wanted. But the HMO could only assure full coverage for care it deemed necessary and appropriate.

When Mrs. Worthy called Dr. Newman's office a second time to set up the authorized appointment, she told the receptionist that her husband had a "really, really bad headache," and couldn't wait until the first available slot next week. Although the receptionist did not completely appreciate the seriousness of the situation, she did juggle out a next-day-appointment.

Jane Worthy was concerned. She drove John hack to the ER at 11:00 PM, but even before she pulled up to the doorway, his tortured breathing had stopped. Successive attempts by the hospital staff to revive him failed. She was deeply sorry that her husband died of a massive, brain hemorrhage as his cerebral aneurysm ruptured.

WHAT COULD HAVE SAVED JOHN WORTHY?

FRAN DAVIS

Family doctor

I made the decision not to see Mr. Worthy in the office and recommended

mended instead that he go to the emergency room if the pain "seemed that severe." I didn't actively decide not to see him; I just could not fit him into the schedule on that particular day. Mrs. Worthy called when the office was "Monday morning jammed" as we were trying to accommodate all the patients who had become ill over the weekend and were "triaged" by our cross-coverage system of physicians and "physician extenders." Knowing that my schedule was completely booked, I had very few options.

Ordinarily, new patients are given thirty-minute initial appointments to review their relevant medical, surgical, and social histories as well as current medications, allergies, and active medical problems. The initial appointment also helps the patients and me become familiar with one another, and provides me the opportunity to describe my philosophy of practice, how the office functions, what my patients can expect of me, and what I expect of them. When I receive calls about or from patients I've never met, I have neither a context within which to place their symptoms nor any other pertinent information, including their frequency of and attitudes about illness, their social and community support systems, etc. Unfortunately, unless a patient calls me I have no way of knowing which patients have newly designated me as their primary care provider. The plans our practice contracts with have not been helpful in this regard. Such was the case with Mr. Worthy

Given the scheduling needs and expectations by Physician Management Services that I will have a rather large (and growing) panel of patients, my availability for evaluating a patient on an emergency basis varies. Patients with subacute and or urgent problems are often seen on a "same day" basis, and certainly within a day or two. (I do try to keep a few appointments open each day for true emergencies, though they often get filled with patients who have less urgent problems, or don't get filled at all. This can lead to quiet [read unproductive] times in the office and I'm told we can't comfortably afford that.)

My own skills enable me to diagnose and treat over 90 percent of my patients' symptoms, but for some patients a referral is indicated. For the kind of problems Mr. Worthy was experiencing I sometimes refer the patient directly to the emergency room for evaluation and treatment Determining just how "emergent" a problem is is difficult, however, and often falls to the discretion of the patient. I have only a small number of patients from GoodCare and cannot begin to remember how GoodCare's policies and preferred referral patterns align with or differ from those of the other eleven plans we work with. In the same way, defining what's appropriate and "necessary is very challenging. But it's increasingly important (for my financial well-being and the plans') that I authorize only "appropriate" referrals. As a result, I sometimes manage patients' symptoms or diagnoses for which my skills are adequate but not as good as a specialist's.

Contracting with managed care plans keeps our practice going, but it's also costly. Completing insurance forms, obtaining (pre)authorization, monitoring adherence to plan guidelines on preventive care, etc. forces us to devote greater resources to the administrative functions of the office. That, combined with lower reimbursements from the plans, has led our office to replace highly skilled nurses with less skilled medical assistants. We can no longer afford to employ a knowledgeable nurse to elicit the history from Mrs. Worthy that might have suggested the need for ER evaluation. When Mrs. Worthy called the office she spoke with my receptionist, who is pleasant and tries to be helpful but has no medical training.

I suppose that if I had spoken to Mr. Worthy myself I might have been able to determine how serious his condition potentially was and authorize referrals to the neurologist or ER right away

My practice is increasingly monitored for utilization (rates and costs of outpatient visit, specialist, and ER referrals, x-ray and pharmacy costs, etc.), adherence to preventive health measures, and the quality of my charting. As much as I hate to admit it, the close scrutiny also affects my advice to patients with regard to referrals. Not only am I increasingly held accountable for the costs of specialty care, some managed care plans expect me to challenge the consultants' recommendations. This kind of interaction, however, has tended to sabotage my relationships with the specialists upon whom I often rely.

When Mr. Worthy was having such difficulty navigating the system I received a call from Dr. Depp, GoodCare's medical director. He asked me to clarify the clinical situation, the course of events, indications for ER/neurology referral, etc. I did not have any of that information. When I suggested that we simply refer Mr. Worthy to the neurologist based on the patient's experience Dr. Depp challenged me. I then had to call the ER track down the physician who cared for Mr. Worthy, review the clinical situation, get back to Dr. Depp, get back to the Worthys, generate a referral form to fax to Dr. Newman, and call his office to facilitate the urgent referral. The investment of time, energy, and money, and the disruption in office operation was, enormous.

Perhaps I should have referred Mr. Worthy to the emergency department that was covered by his plan, or called the ER physician to inform

him of Mr. Worthy's imminent arrival. Perhaps I should have known all the relevant rules of GoodCare at the time Mrs. Worthy called, in which case I could have helped John avoid the situation he confronted at the ER. Perhaps, perhaps, perhaps... Although GoodCare is quite new to me that doesn't absolve me from responsibility for knowing their referral rules.

I am somewhat worried about the legal ramifications. I might be told I should have had a higher quality triage system. Perhaps in my zeal to maintain a "productive" practice I did not allow for sufficient appointment openings. GoodCare and/or I may be sued. The publicity could do terrible damage. Though I have great reservations about participating in GoodCare, I cannot truly afford to give them up. I also worry that GoodCare may deselect me, since the contract does say I can be deselected "for cause"--without defining what constitutes "cause"--and if they do, perhaps other plans will follow suit.

GoodCare's Policy on access to the ER created very real problems for the Worthys and me. Relying on patients' discretion is often not clinically appropriate. Requiring patients to see me for an ER referral typically elicits anger at me (not the managed care plan) and I find myself having to try to explain a plan's policy. And while patients can be expected to read their contracts after finally choosing a plan, I don't think it's reasonable to expect them to anticipate their future medical needs and assess how well a particular plan may respond to those needs.

Of course, I'm not sure the outcome would have been any better had I evaluated Mr. Worthy in my office and/or referred him to the plan emergency department immediately. Still, I realize now more than ever that when the clinical situation demands it I may have to interrupt the schedule of my very "productive" office and inconvenience many other patients for the benefit of the one. Admittedly, focusing on the needs of an individual patient is more difficult in the context of managed care's philosophy of "population-based medicine." At the same time, I can see that if I simply refer patients to the ER for problems that I could have evaluated and treated, especially an ER that is not participating in the patient's plan, I am contributing to some very considerable charges to the HMO.

It's not unreasonably idealistic to suggest that Mr. Worthy should have been able to call the office, be triaged sufficiently over the phone to determine how emergently he should be evaluated, be referred to the nearest ER if the clinical situation seemed to warrant it, have a thorough evaluation at that ER, and then be treated as the clinical situation dictated. If in retrospect neither the symptoms nor diagnosis were truly emergent (according to even a "reasonable patient standard"), patients and physicians alike may have to share the expense.

And it may be that employees will have to bring greater pressure to bear on employers, and employers in turn convey that pressure to plans to be responsive. And employees and patients may have to take far greater responsibility for reviewing the benefits packages and policies of plans they are offered so that their choice is a more informed one.

EDWARD R. POST

Emergency department physician

I acceded in Mr. Worthy's decision to decline my recommendation

to have a CT scan for several reasons.

A severe headache that comes on suddenly a couple of hours previously may be brought on by many causes, most of them benign. Rarely severe headache may be caused by the very thing that caused Mr. Worthy's death. Though the probability is small, the consequences are devastating and brain hemorrhage should be in the back of my mind whenever I pick up a patient chart with the chief complaint of "headache." A CT scan might have saved Mr. Worthy's life.

When a patient declines to follow my recommendations for diagnosis and treatment of a medical problem I face a dilemma: how strongly should I try to convince the patent to go along with my recommendations? When I think the patient risks a high likelihood of harm I actually have no dilemma. I reason; I argue; and if necessary I cajole in trying to persuade the patient to follow the recommended course of action. I am not above asking any relatives present with the patient to enter the debate in an attempt to convince him or her (I confess I have skated on the edge of patient confidentiality in this.) And if a patient still refuses, I ask him or her to sign a form acknowledging that he or she left the ER "against medical advice." When there is an extremely small chance of some dire consequence, as in Mr. Worthy's case, I am less aggressive about persuading the patient to follow recommendations.

I believe strongly in patient autonomy and the right to choose. Many of my emergency physician colleagues have two rules when dealing with patient refusals: 1) in an emergency, when in doubt, treat;, 2) always doubt. I know better that the emergency/exception to informed consent is circumscribed to true emergencies where the patient lacks decisionmaking capacity and no one is legally authorized to decide for the patient. Thus though I may feel strongly about which is the best course of action, I do not try to trump the competent patient's decision, whether it is as important as deciding to forgo resuscitation or as minor as to forgo stitches for a wound.

Once a patient decides to refuse my recommendation for something that would help diagnose a condition that is uncommon but serious, such as an intracranial bleed, conversation with the patient becomes crucial. This kind of conversation was made more difficult by the fact that, like most of my emergency patients, Mr. Worthy had not met me before. Though I try to build a rapport with each patient, it is quite difficult to find the right way to express the gravity of such a decision and the proper weight my recommendation should hold--and to be confident that the patient truly has capacity to make such a decision and has given the risks and benefits due weight.

Mr. Worthy was not the first patient who refused to have tests I have recommended. Patients have refused to undergo tests because of the pain associated with the procedures. Mr. Worthy was, however, the first patient who refused a CT scan (a painless procedure) based on the costs that he perceived he would have to pay out of pocket

In hindsight, I wish I had spent more time trying to convince Mr. Worthy to have a CT scan. Of course, I really did not know whether his plan would have paid for the tests. Too many plans contract with the hospital and too many different patients come through the emergency department for the ER staff to have any knowledge of an individual's coverage. I could not have reassured the Worthys that their HMO would pay

And even under the best of circumstances, the diagnosis of intracranial bleed can be missed. CT scans cannot be done for everyone with a headache, and even if a CT scan is done, the bleeding might not be extensive enough at the time to show up. A lumbar puncture might then be necessary to find it, and even then this test might not disclose impending doom.

The chance that those tests won't be reimbursed poses a dilemma for our hospital. We are located just on the edge of the poorest area of our city. For the most part, our patients are "working poor" without insurance. Many of them do not pay their medical bills. In essence, no insurance means no payment for us. And in the past year our county hospital, the mainstay of health care given to the uninsured poor, was closed, which has greatly increased the number of uninsured in our emergency department census. The private hospital that has contracted to care for indigent patients is a long busride away. In a perceived emergency, a patient will go to the nearest hospital, and we are required by federal law to medically screen all patients and to treat those wish emergencies. Also, we are seeing more uninsured patients

back in our emergency department who are turned away from primary and specialty physicians' offices because of problems with payment.

In the past we made up for the lack of payment by shifting costs to those who could most afford to pay--patents with private insurance, As managed care plans have come to dominate the market in our city they have affected our reimbursement mix by discouraging the use of the ED by their "insureds" and decreasing the amount paid when the ED visit is approved. As a result we have had less ability to shift the cost of the unpaid bills of our poorer patients.

One of the not-so-secret truths of emergency medicine is that most of our practice is not taking care of emergencies. Plenty of patients come to the emergency department with minor complaints that could be seen in their doctors' offices the next day. We estimate that only 10 percent of our cases are truly emergent, that is, must be treated in the next minutes to hours--such as heat attacks or appendicitis. Forty percent of our cases are urgent (must be treated in the next twenty-four hours, e.g., fractured bones without significant displacement), and 50 percent are nonemergent and nonurgent (e.g., viral sore throats, chronic back pain).

I have a certain amount of sympathy with the managed care plans' objectives of decreasing unneeded emergency visits. Even my colleagues and I feel that the emergency department is abused when patients come for minor problems. The visits that prove to be nonemergent, though, turn out to be quite important for the emergency department. They cover the costs of keeping the department open and fully staffed every day and night of the year. And they pay for the other patients who do not pay their bills.

Another problem is the increasing bottom-line orientation of our emergency department. Our hospital, felling the effects of competition, has trim to attract paying patients back to our ED by advertising that patients will wait no more than two hours from the time they are seen by a nurse until their tests and treatment are completed. This advertising of a fast "turn-around" time has not yet come with a money-back guarantee, but that is not out of the question.

And the hospital puts great emphasis on patient satisfaction, so much so that in the new contract that the hospital has proposed for services by the emergency physician groups, patient satisfaction surveys will be used as performance standards by which the contract may be cancelled and the physicians put out of a job--with only ten days' notice. Thus we physicians are under pressure not only to make the right diagnosis and treatment, but also to make people happy with our "customer service." The sicker patients become a drag on the patient flow, and the system now seems to penalize the truly ill for the benefit of the "worried well." If those who may have to pay their own emergency department bill because their managed care plan denies coverage have to wait hours on end to receive this expensive care, they are more likely to complain about the wait.

I suppose I could have tried to contact Mr. Worthy's primary care physician, or to talk to the gatekeepers at the HMO. But even on weekdays during normal business hours this has proven to be a long, tedious, and usually fruitless process.

If I had insisted on the CT scan, I believe Mr. Worthy would still be alive today As he would if his HMO adhered to a "prudent layperson" standard to authorize payment for emergency services---covering care when a reasonable person believes his or her condition requires emergency evaluation and/or treatment.

CONNIE S. ROGERS

Customer service representative

As a customer services representative for GoodCare my job is

to help subscribers work with the system. That usually involves referring people to the proper medical or business office. It is my responsibility to answer questions about health insurance coverage and the process for obtaining care. I have a college education, but I don't have a clinical background and I'm not really a medical person.

I am also the first contact for complaints, and I try to settle them at my level. I want the customer to be satisfied and my boss not to be disturbed.

When Mrs. Worthy first called my office the medical problem seemed routine. She told me that they had been to an out-of-plan emergency room, so I understood that the physicians who had evaluated Mr. Worthy had not deemed his medical problem an emergency---if they had, he would have been treated there and then. The Worthys were new to the plan; they did seem confused about procedure, and that is where I am well trained to help. I clarified some of our rules that govern reimbursement processes, approval, and denial, explaining that to be sure things like a CT scan would be covered the Worthys should see their primary case physician first. Because Mrs. Worthy told me this primary care doctor didn't have an appointment available that day I tried to he helpful in identifying Dr. Newman as an in-plan neurologist for them to consult.

I'm successful at my job when I help the subscriber through the system, so I was especially alert when Mrs. Worthy called back. I do care about the customer--and I have a family myself--I could understand that she was very angry and why she demanded to speak to the president of GoodCare. But the president would not be the person to call for help with Mr. Worthy's medical problem. From the way Mrs. Worthy described the situation it seemed they needed a medical doctor to authorize an emergency room visit and tests. And if I put Mrs. Worthy through to the president I would not be doing my job well either. Even when business or access to care issues are involved, the president tells us he wants us to offer him solutions, not problems.

In the short run it might have been easier for me if I had transferred Mrs. Worthy to the president. But I could hear Mrs. Worthy's frustration and concern and transferring her wouldn't have helped. Frankly, I was also quite worried they were working up to sue us, so I offered to call the medical director and try to get help.

My understanding of the problem for me to deal with--or at least refer---was not about Mr. Worthy's medical condition itself. The problem for the HMO and me was getting him seen by a plan provider or having the Worthys assume financial responsibility for care given outside die plan. They needed to understand that we have no financial duty outside the plan unless the medical problem is an emergency defined by the doctors. Our HMO is typical: we contract to arrange and pay for services, and we do what we promise--that is, we do what we have contracted to do. I believe consumers should take some responsibility for knowing what they have bought. Doctors do the doctoring. Since the emergency department doctor had not said this was an emergency, the main duty was to help Mr. Worthy get to a plan provider for evaluation and treatment.

I never expected that Mr. Worthy could die. I am so sorry, and it troubles me because I don't see what I could or should have done differently to help. As the contact person for GoodCare I have a very narrow set of responsibilities and authority, I took as much initiative as I could to help Mr. Worthy get care. The next time someone calls with a problem getting emergency care I'll still have to rely on the doctors.

MICHAEL DEPP

Medical director, GoodCare

I made the decision that GoodCare would employ a primary care case management program for all of its members and would only approve payment for emergency department visits that GoodCare determined to be appropriate. These policies were designed to promote formal reliance on a primary care provider as the first point of contact in the health care system and to position the primary care provider to be the care manager for services for an HMO member. And they are intended to encourage members to seek emergency services only when they are truly experiencing a bona fide emergency. This latter policy was not designed to discourage visits to the emergency department that were for a legitimate emergency, but to encourage discreet use of services. This is a standard policy of all the HMOs with which I am familiar.

Sound, credible research over many years has found that the preponderance of visits to emergency departments me for routine or urgent care that falls far short of reasonable criteria for emergency care. Typically ED care is sought for convenience purposes or because a person has not developed an established relationship with a primary care provider. An emergency department visit, which normally entails initial triaging and potentially involved and expensive diagnostic testing, may cost several hundred dollars to arrive at the same

conclusion a phone consultation with a regular source of care might produce at a dramatically lower cost.

It is also well known that because emergency departments specialize in episodic care, there is a great likelihood that patients will not obtain timely follow-up care from that site. We fully understand that hospital emergency departments ate reluctant to turn patients away because of fear of being accused of "dumping," and we understand that their prices are high because they must offset the high cost of uncompensated care rendered in the ED. We are prepared to pay these prices when it is truly necessary for one of our members to go to the emergency department.

Our plan takes the role of the primary care provider very seriously. As a general internist, I know very well how valuable a primary care physician can be to the uninformed patient in the complex medical care process. The design we created is built around being able to select competent, qualified primary care physicians to play the key role as patient care managers. We encourage primary case providers to do everything they can, within their own realm of competence, and to refer or authorize care when presented with something beyond that realm. We carefully screen qualifications of physicians applying to become primary care: providers--over 90 percent of them are board certified--and we have a detailed primary care provider contract that explicitly spells out the roles and responsibilities for the primary care provider position. We require primary care providers to have twenty-four-hour coverage available to our members, and we do spot-checking to establish compliance with this responsibility. Regularly administered member satisfaction surveys also pose questions regarding the off-hour accessibility of primary care providers, and this ultimately has an influence on physician compensation. And primary care providers are required to observe specified standards of appointment and wait times that are incorporated into their contracts with GoodCare. We have to assume the primary care provider would not sign a contract that he or she did not fully understand.

Out capitation system for primary care offers incentives to encourage the primary care provider to counsel his or her patients to use the emergency department appropriately. And we provide the primary care provider with detailed profiles of the care rendered to their panel members so they can identify panel members who have had ED visits that have been disallowed. All of these policies apply whether we are contracting with individual physicians or, as in the case of Dr. Davis, with a practice management company like Physician Management Services. In that case, we pay a capitation to the management company for each GoodCare member who selects a PMS physician, and we continue to produce individual physician profiles and to monitor member satisfaction based on the physician with whom they are enrolled. Unfortunately, we do not have control over how PMS chooses m compensate its physician or what sorts of productivity standards are used since the physicians are employed by PMS, not GoodCare.

The HMO has also adhered to a firm policy of requiring primary care provider approval for all specialty referrals because we want the primary care provider to be the "captain of the team" and to protect our members from getting unneeded specialty services. Primary care management is the foundation of our care delivery system and we do everything to bolster this, we will not be able to maintain the integrity of the product we are promising to our members.

Our plan bas a limit on the number of patients for which a primary care provider may be case manager to ensure that they are able to give our members adequate attention, though I must admit we cannot control the overall size of a primary care provider's practice or the pace at which patients are seen in his or her office. We know that many physicians sign many managed care Contracts and we realize that physicians may feel overwhelmed by some of the variation among the plans with which they are affiliated. This is unfortunate and it is one of the reasons we want our physicians to have a sizable GoodCare presence in their practice--so they will be attentive to our members.

Because Dr. Davis is not exclusively devoting her time to our plans, we have to rely on her--and PMS to determine how much time she will spend with our members rather than other patients in her practice. I sympathize with the pressure she must feel trying to generate a sufficient practice volume to satisfy her employers.

Our plan's quality improvement committee, of which I am ex officio chair, has reviewed our ED authorization policy twice in the past three years and reaffirmed its soundness. It was also reviewed by thc National Committee for Quality Assurance (NCQA) at the time we had an accreditation site visit earlier this year. In fact, we have identified reducing inappropriate ED visits as one of the key projects we will undertake next year to demonstrate to NCQA our commitment to continuous quality improvement. As I recall, the only complaints we have received about this policy have come From hospital emergency departments in non-network hospitals, which have suggested that we are too restrictive and that and that appeals process for denials was ambiguous. But, frankly, we think this criticism is largely due to the fact that we did not include those hospitals in our provider network.

There are also sound economic and behavioral reasons why members should bear the responsibility for unauthorized ED visits. Research clearly indicates that cost participation discourages unnecessary care. GoodCare has adopted the most widely accepted clinical criteria in the managed care industry for use in determining what is a true emergency. Reducing unneeded care enables us to hold down premiums to employers, who should not have to pay for unneeded or inappropriate visits. In turn, employers can provide health benefits with little or no premium contribution by employees if we can hold unnecessary care to a minimum. I want to emphasize that we make this policy clear to all employer groups, who have consistently reaffirmed their unwillingness to pay for unneeded care. We also provide very explicit information in our member handbook--it is in bold print on page 37--that members will be financially responsible for care received in a hospital emergency department that has not been judged to be a legitimate emergency by GoodCare. If there are questions about this or any policy in the handbook, members are encouraged to call our toll-free information line to obtain clarification. It is hard for me to see how we could be doing anything more to make this policy clear to everyone.

I have very few misgivings about our policy on emergency department care. It is a responsible one that is consistent with practice across the industry. I have complete confidence that our primary care providers will not be unduly influenced by incentives to avoid necessary referrals or skimp on care, and that they will render sound clinical judgments when contacted by patients and/or hospital EDs. After all, these primary care providers are physicians first, and members of our network second, so I fully expect each to adhere to the high standards of the medical profession. Besides, they know the malpractice risks they might face if they fail to act in a diligent fashion on behalf of their patients.

As primary care providers they are expected to develop a relationship with all the patients in their practice, and we even permit a member's initial visit to the primary care provider after joining the plan to be made without any copayment. Ultimately, we firmly believe a good physician will know when to override or ignore a policy that puts his or her patient's care at risk and will do the right thing. I certainly wish we had had more than one neurologist available, but the only way we can get any kind of negotiating leverage with these guys is to try to have a sizable portion of the patients in their practice.

I do have some concern, however, that our members may make bad judgments in the face of financial incentives. I sometimes worry that consumers fail to value health care adequately, and thus may avoid getting services that they have to pay for out of their own pockets. We know that a lot of preventive care does not get delivered for this reason, which is why we have extended our benefit package in these areas. But people respond to clinical symptoms in different ways. The worried well come in for unneeded visits or people with low threshold for pain or uncertainty may rush into emergency departments. Other people delay needed care to save a few dollars. Likewise, some people pick plans to minimize their out-of-pocket costs while others gamble with high deductibles and copayments to save a few dollars. It all depends on how they want to spend their money.

Care seeking is an inexact and variable process, and we are working very hard to make it more rational and predictable. In another market we are piloting a new nurse call advice system, coupled with distribution of a self-care manual that we hope will be successful in raising the level of understanding of signs and symptoms among our members so they will respond appropriately. We are hopeful this form of "demand management' program will screen out unneeded care of all kinds, including ED visits, without discouraging necessary care. This education piece should be really helpful in avoiding the unfortunate situations like the one with Mr. Worthy. But it won't be perfect, since some people just make had judgments.

As the medical director of this HMO, I can honestly say we have developed a care delivery system, member policies, and provider compensation arrangements that fit well together and have achieved success in providing a relatively rich benefit package at a modest cost. The fact that our HMO has had net growth in membership of more than 10 percent in each of the past five years and never had disenrollment rates in excess of 3 percent is proof that we am meeting the needs of our customers, We have not even raised our premiums in the past three years because of our success in reducing unnecessary use and negotiating fair prices with our network providers.

Looking back on what happened, 1 don't really think our system failed Mr. Worthy. Of course, I wish be had gotten in to see a neurologist sooner--though there is no guarantee that that alone would have saved his life. We will have to take another look at whether the neuro-capacity in our network is too tight, and see what we can do to get another on board. And I wish Dr. Davis had known the patient and his wife better, because I am confident that she would have been able to appreciate the urgency of the situation better than it appears that she did. Dr. Davis is a fine physician, but our QI committee will have to examine her practice profiles somewhat more closely now became of what may become an ugly situation with the lawyers. I am sure we will face criticism and recrimination from the hospital, but I think our executives may be able to defuse that by offering them the opportunity to join our network as they have been seeking to do so for some time. I also believe we will be okay with Worthy's employer as they have been a good account since they began with us, and they offer a menu of choices to their employees so no one feels railroaded into GoodCare. We might have some trouble maintaining our membership there if word gets out before the next open enrollment period. I'm sure our competitors will exploit