Chapter 1 Summary and Policy Options

Chapter 1

Summary and Policy Options

CONTENTS

PageIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Selected Life-Sustaining Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Focus on the Elderly Population . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Who Are the Life-Threatened Elderly?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The Specter of Rationing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .The ’’High Cost of Dying”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Quantity v. Quality of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Selected Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Current and Future Resource Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Quality of Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Access to Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Decisionmaking Problems and Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Congressional Issues and Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Access to Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Decisionmaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Quality of Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

B o xBoxI-A. Principles for Decisionmaking Regarding the Use of Life-Sustaining

Technologies for Elderly Persons, as Developed by ProjectAdvisory Panel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure

3456889

1010111516192828303235

Figure No. Pagel-1. Heterogeneity of the Life-Threatened Elderly . . . . . . . . . . . . . . . . . . . . . . . . . . 8

Tables

Table No. Pagel-1. Utilization of Life-Sustaining Technologies for Patients of All Ages, and for

Elderly Patients, in All Settings Combined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11l-2. Charges for Life-Sustaining Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Chapter 1

Summary and Policy Options

INTRODUCTION

The dramatic advances in life-sustaining medi-cal technologies during the past three decadeshave been accompanied by rapid expansion intheir availability and use. As equipment and pro-cedures have been refined and experience accu-mulated, the necessary personnel, facilities, andreimbursement have expanded, and the clinicalcriteria guiding use have been broadened. Thetypes of patients who become candidates for life-sustaining treatments have changed, and theirnumbers have increased sharply. Many of thesepatients are elderly. As the population ages, asonce “extraordinary” measures become common-place, and as ever-more powerful technologiesemerge, it becomes increasingly important tounderstand the problems as well as the potentialassociated with the use of these technologies andto devise policies that reflect this understanding.

Technologies that support or replace the func-tioning of a vital organ are capable of saving andsustaining life and, sometimes, capable of restor-ing health and independence. However, an indi-vidual’s response to treatment can seldom be pre-dicted with certainty; thus, it is never clear thata “life-sustaining” technology will sustain the lifeof a particular patient or, if it does, for how long.The quality of the life that is sustained may beeven harder to predict. Patients and other inter-ested parties may evaluate differently the bene-fits and burdens associated with treatment versusnontreatment and with one treatment versusanother. An important factor that further com-plicates matters is that many patients with life-threatening conditions are not able to understandtheir treatment options or to express preferencesregarding them.

Public discussion about the use of life-sustainingtechnologies, either for individual cases or healthcare policy, is relatively new, but newsworthy. Atany one time, many thousands of elderly personsare receiving life-sustaining interventions. The vastmajority of cases go unnoticed except by the pa-tients, family members, and others directly in-

volved in making and living with difficult treat-ment decisions. However, a few of these cases gainnotoriety and public attention as it becomes appar-ent either that treatment was unwanted or futileor, conversely, that some new medical break-through or personal triumph over adversity hasoccurred. Under public scrutiny, these cases makeclear the interdependence of private health caredecisions and the public policies that determinewhether treatment choices are legal, ethicallyacceptable, economically feasible, and fair.

The legal, ethical, and economic questions raisedby decisions about the use of life-sustaining tech-nologies have been studied by scholars and pol-icymakers both inside and outside the govern-ment. The first major government publicationsaddressing access to and decisions regarding theuse of life-sustaining treatment were prepared inthe early 1980s by the President’s Commission forthe Study of Ethical Problems in Medicine and Bio-medical and Behavioral Research. Related researchhas been performed or sponsored by the Officeof the Surgeon General, the General AccountingOffice (GAO), the Congressional Budget Office(CBO), the National Institutes of Health (NIH), andthe Office of Technology Assessment (OTA). Thesestudies demonstrate the lack of consensus regard-ing appropriate use of life-sustaining technologiesand the lack of consensus even about appropri-ate procedures for making those decisions.

This assessment draws on the earlier studies,but it is different from them in two importantrespects. First, it is focused on particular tech-nologies, The discussion goes beyond abstract con-siderations related to the care of the critically andterminally ill to identify specific problems and po-tential solutions related to selected technologiesused to treat or manage life-threatening condi-tions. Second, this assessment is focused on a speci-fied age group, i.e., persons over age 65, ratherthan on all potential patients. The major purposeis to provide an array of options for public policythat will support wiser clinical decisions about the

3

4 ● Life-sustaining Technologies and the Elderly

use of these technologies. Toward this goal, theassessment presents information about topics asdiverse as the cost of equipment, competing ethi-cal principles, the experience of patients and theirfamilies, and the training of health care profes-sionals. The assessment synthesizes available andnew information, from a new perspective, andfrom this it develops a set of issues and relatedoptions for congressional review.

Selected Life-SustainingTechnologies

Life-sustaining technologies are drugs, med-ical devices, or procedures that can keep indi-viduals alive who would otherwise die withina foreseeable, but usually uncertain, time pe-riod. While these technologies share some com-mon ethical, legal, and health care delivery prob-lems, each has unique characteristics that eitherraise special questions or suggest possible solu-tions. Five specific technologies used to treat ormanage life-threatening conditions are the focusof this assessment:

1. Cardiopulmonary resuscitation (CPR) re-fers to a range of technologies that restoreheartbeat and maintain blood flow and breath-ing following cardiac or respiratory arrest.Resuscitation procedures range from basiclife support, which uses manual externalcardiac massage and mouth-to-mouth venti-lation, to advanced life support, which mayinclude application of prescription drugs andsophisticated devices such as an electricaldefibrillator, temporary cardiac pacemaker,and mechanical ventilator. Resuscitation hasextremely wide potential application becauseit can be applied to virtually any person whoseheart stops beating,

2. Mechanical ventilation is the use of a ma-chine to induce alternating inflation and defla-tion of the lungs, to regulate the exchangeof gases in the blood. The most common typeof ventilator (or “respirator”) delivers inspira-tory gases directly into the patient’s airwaythrough tubing that connects the patient tothe machine. The technology is used to sus-tain patients whose spontaneous breathingis inadequate or has stopped altogether due

3.

4.

5

to acute or chronic diseases of the neuromus-cular, necrologic, or pulmonary system, ordue to anesthesia or trauma. This assessmentis particularly concerned with mechanicalventilation that becomes prolonged o rchronic.Renal dialysis is an artificial method ofmaintaining the chemical balance of the bloodwhen the kidneys have failed. The blood iscleansed of impurities, either by cycling theblood through a machine and back into thepatient via catheters (hemodialysis), or bycycling dialyzing fluid into and out of the ab-domen using the patient’s peritoneal mem-brane as a filter (peritoneal dialysis). Dialy-sis is used for patients in acute renal failureand those with chronic end-stage renal dis-ease (ESRD).Nutritional support and hydration refersto artificial methods of providing nourishmentand fluids. The two modes of delivery are en-teral (or tube feeding), in which nutritionalformulas are delivered via a tube into thedigestive tract, and parenteral that includesall methods other than enteral but is primar-ily intravenous feeding in which nourishmentis delivered via catheter into the bloodstream.Total parenteral nutrition (TPN) is an intra-venous procedure that supplies sufficientnutrients to maintain a person’s weight in-definitely. Tube feeding and TPN are used pri-marily for people who are unable to take suffi-cient amounts of food and fluids by mouthor who are unable to digest and absorb themadequately.Antibiotics are a large set of drugs used tocure or control numerous bacterial, viral, andfungal infections, including minor ones. Dif-ferent families of antibiotics have been de-veloped for use in combatting different typesof infections. Antibiotics maybe administeredtopically, orally, intravenously, or intramus-cularly, in discrete doses or continuously. Allantibiotics are potentially life-sustaining, By“life-sustaining antibiotic therapy” OTA meansnot a particular drug or family of drugs butthe use of any antibiotic against a life-threat-ening infection.

With the exception of antibiotics, none of thefive technologies examined in this assessment can

Ch. 1—Summary and Policy Options • 5

cure the underlying condition that precipitatedits use. Thus, among patients who receive theseinterventions and survive, health status and func-tional capacity vary widely. While some patientsregain adequate natural function of the affectedorgan, others become permanently dependent onthe life-sustaining technology (and they may besimultaneously dependent on more than one life-sustaining technology). They may require continu-ing medical care and, often, other forms ofassistance.

The life-sustaining technologies OTA has stud-ied are only a few of many possibilities. They wereselected to illustrate significant ranges across suchdimensions as burden, cost, and risk. For exam-ple, antibiotic therapy administered intravenouslyis relatively painless and nonrestrictive, especiallyin comparison with mechanical ventilation, hemo -dialysis, and TPN. Mechanical ventilation fre-quently involves continuous, round-the-clock ap-plication, while hemodialysis is typically appliedthree times per week for 3 to 5 hours per treat-ment. Resuscitation is, ideally, applied only once.Costs and expenditures, which are related to fre-quency and duration of treatment, range fromminor to catastrophic. Available reimbursementmay be near total or minimal. The technology maybring risks of serious complications (e.g., renalfailure associated with mechanical ventilation) or,provided proper procedures are followed (e.g.,to prevent catheter-related infection for TPN), itmay be generally safe. While invasiveness and highcost may tend to restrict use, low risk and lowcost (or generous reimbursement) may lead tooveruse. All these factors bear on clinical deci-sionmaking.

The five technologies examined in this assess-ment also illustrate the variety of settings and cir-cumstances in which life-sustaining treatment canbe administered. Most of these technologies arenow technically possible and available not onlyin acute care hospitals and intensive care units(ICUs), but in nursing homes, patients’ homes, andother community settings. While incubation formechanical ventilation is usually done by highlytrained professionals in an emergency room orICU, some stabilized ventilator patients can man-age in their own homes. Basic resuscitation tech-niques can be performed by trained bystanders

wherever a cardiac arrest occurs, but advancedCPR requires emergency transfer to a hospital.

Focus on the Elderly Population

This assessment focuses on elderly persons whoare already receiving or who might become can-didates for life-sustaining medical technologies.For purposes of this assessment, the elderly pop-ulation is defined as all persons aged 65 andover. OTA recognizes and emphasizes, how-ever that defining the elderly population onthe basis of any chronological age criteriontends to mask the heterogeneity of that popu-lation. Sixty-five, or any chronological age, is apoor indicator of biological function, physiologi-cal reserve, cognitive ability, or health care needs.The use of age 65 is justified, however, by its prom-inence in available health and demographic sta-tistics and its relevance to eligibility criteria in cur-rent Federal and State health care programs,especially Medicare. To minimize the loss of ana-lytical and descriptive rigor from using a singleage criterion, this assessment refers wherever pos-sible to subgroups of the elderly population (e.g.,65 to 74, 85 and over).

While many important considerations in the useof life-sustaining technologies apply regardless ofthe patient’s age, some factors distinguish theelderly as a special population. These include:

●

●

●

●

Elderly people, as a group, are at greater riskof life-threatening illness than younger people.Because both the prevalence and severity ofchronic conditions and their associated dis-abilities increase in old age, elderly personswho experience a life-threatening illness aremore likely than younger persons to alreadybe in a state of compromised health and re-duced functioning that negatively affects theirquality of life.Elderly people are more likely than youngeradults to be victims of a dementing illness,and they have high rates of other disorders(e.g., depression, drug toxicity) that may tem-porarily or permanently impair their abilityto make health care decisions.Comorbidity (the coexistence of more thanone disease) and age-associated loss of func-tion complicate the prognosis and treatment

6 ● Life-Sustaining Technologies and the Elderly

●

●

●

●

●

of life-threatening conditions in elderlypersons.There are questions about the quality ofhealth care currently available to elderly pa-tients. Many health professionals in practicetoday are poorly prepared to care for seri-ously ill elderly people whose presentationof disease and response to treatment may dif-fer from that of younger adults.As a group, elderly people utilize a large shareof all health care resources and consume thelargest share of public health care dollars.Elderly people, as the major beneficiaries ofMedicare, may bear the brunt of Federal ef-forts to contain health care costs.In contrast to other segments of the popula-tion, especially newborns and young children,the law recognizes the autonomy of elderlyadults.Elderly persons are more likely than youn-ger adults to have contemplated the mean-ing and value of their life and its end.

The significance of the above factors will beheightened as the elderly population increases inabsolute and relative size, and in average age.Demographers predict continuing growth of theelderly population, from approximately 25.5 mil-lion people and 11 percent of the U.S. populationin 1980 to 35 million and 13 percent in 2000. More-over, conservative projections indicate that thepopulation aged 75 to 84, which accounted for30 percent of the total elderly population in 1980,will reach 35 percent in 2000. During the sameperiod, the proportion of persons 85 and olderwill increase from 9 to 15 percent of the popula-tion over 65.

Who Are the Life-ThreatenedElderly?

In order to emphasize the diversity of the pop-ulation at risk and to illuminate problems in mak-ing decisions about their care, OTA has deviseda classification system consisting of four catego-ries of ‘(physical status” and four categories of"(decisionmaking capacity.” Most of these catego-ries are not articulated in practice, but they in-fluence a person’s ability to make treatment deci-sions for himself or herself and may also influence

the decisions that are made by others on a per-son’s behalf.

Variation in Physical Status

A life-threatening condition may be—and inelderly persons frequently is—superimposed onpreexisting physical and/or mental disorders, orit may occur in an otherwise healthy and activeindividual. It is inappropriate for clinical deci-sionmakers or public policymakers to lumptogether all elderly persons who becomecandidates for life-sustaining technologies.Rather, the life-threatened elderly should beseen as individuals with widely varying phys-ical and mental status. Physical conditions maybe acute or chronic, have different prognoses(both of survival and restoration of functional abil-ity), and have a course that is either decisive orunknown.

1. Critically ill persons are those in the midstof an acute life-threatening episode (e.g.,cardiac arrest, stroke) or persons believed tobe in imminent danger of such an episode.They are medically unstable, and if they arenot treated, are expected to decline.

2. Chronically ill persons have one or morechronic conditions that may or may not belife-threatening but that reduce chances ofrecovery and restoration of function in theevent of an acute disease. Included in thisgroup are persons who have a life-threateningchronic condition that has been stabilized,with or without a life-sustaining technology,or that is in remission (e.g., chronic renal fail-ure treated with dialysis; cancer in remission).Many chronic conditions that are not imme-diately life-threatening are mildly or severelydebilitating; some (e.g., hypertension) increasethe risk of acute life-threatening illnesses orthe risk of complications associated with acutedisease.

3. Severely debilitated persons have seriousor multiple impairments or comorbidities.Their functional capacity and physiologicalreserve are severely compromised. They aremedically stable but highly vulnerable to newphysiological stresses (e.g., at heightened riskof infections, iatrogenic illness, complicationsof treatment, and accidents).

Ch. 1—Summary and Policy Options ● 7

4. Terminally ill individuals are those forwhom a prognosis of death has been made.Designation as terminally ill usually requiresdiagnosis of an illness that has a predictablyfatal progression that cannot be stopped byany known treatment.

A widely accepted definition of “terminal illness”includes the expectation that death will occurwithin 6 months. This definition has been adoptedby Medicare. In practice, however, accurate prog-nosis is extremely difficult, and this difficulty addsto the dilemmas regarding treatment decisions.Contrary to popular belief, a terminal illness isnot always identifiable as such, and most patientswho are dying have not been declared “terminallyill. ” Only retrospectively can these designationsbe reliably made.

Variation in Decisionmaking Capacity

Cognitive ability has two elements of special im-portance in the context of this assessment. First,a person may be cognitively normal and fully ca-pable of making decisions, severely cognitively im-paired and completely incapable of making deci-sions, or somewhere in between; thus, there aredifferences in the boundaries or content of cog-nition. A person who is confused or disorientedto time and place, or even judged by a court tobe incompetent)’ may still be capable of makingand expressing preferences regarding his or hermedical treatment. It is this relatively narrow con-ception of cognitive ability, i.e., decisionmakingcapacity with respect to medical treatment,that is central to this report. A second importantelement of cognitive ability is temporal. Like phys-ical status, cognitive ability may be stable or fluc-tuating, and a person’s decisionmaking capacitymay be expected to improve or worsen. Thesedistinctions result in four theoretical categoriesof patients, as follows:

1. Individuals maybe capable of making deci-sions about their medical care (and all otheraspects of their life), and their decisionmak-ing capacity may be assumed to be stable.

2. Individuals may be currently capable ofmaking decisions about their medical care,

IIn this assessment, OT’A uses the term “incompetent” SpeCifiCal]J’to designate an assessment of co,gnitiir ability that has been declaredby a legal procedure,

3.

4.

but this status is assumed to be unstable ordeclining. Persons whose lucidity fluctuatesand those with progressive dementing dis-orders are examples.Individuals may be currently incapable ofmaking decisions, but it is expected that theirdecisionmaking capacity will be restored. Thiscategory includes patients who are uncon-scious, severely depressed or confused dueto reversible causes (e.g., anesthesia, drug tox-icity, pain).Individuals may be permanently incapableof making decisions about their medical care(and everything else). In these persons, thereis no sign of ability to absorb and evaluateinformation or to express a preference, andthere is no realistic prospect of change. Ex-amples include patients in a persistent non-cognitive state, irreversible coma, and per-sons who are severely demented.

Combining the physical status categories withthe decisionmaking capacity categories producesa paradigm of 16 patient groups. However, an in-dividual’s placement in this scheme is subject tochange (see fig. l-l). This complexity accounts,in part, for the problems inherent in generaliza-tions about the use of life-sustaining technologies.

The combination of a patient physical and men-tal status may affect both the decisionmaking proc-ess and the decision that is reached. For example,in some States, a patient’s request for nontreat-ment is granted only if the patient is deemed bothdecisionally capable and terminally ill. Or, a criti-cally ill patient, regardless of decisionmaking ca-pacity, might be excluded from the decisionmak-ing process because of the need for immediateaction.

Accurate evaluation of decisionmaking capac-ity is critical, but problematic. Assessment pro-cedures are not reliable and not necessarily com-parable as applied in different institutions,Assessment of cognitive status may be particu-larly difficult when the patient’s physical statusis reduced by illness, drugs, or other medical in-terventions, or when the patient is depressed. Pa-tients whose ability to communicate is impairedor unstable present added problems for accurateassessment.

8 • Life-Sustaining Technologies and the Elderly

Figure 1-1 .—Heterogeneity of the Life-Threatened Elderly

Decisionmakingcapacity Decisionmaking

capacity

D

Decision making Decisionmakingcapacity capacity

KEYPhysical status: Decisionmaking capacity:

A Critically ill 1 Capable of making decisionsB Chronically ill 2 Currently capableC Severely debilitated 3 Currently incapableD Terminally ill 4 Permanently incapable

BACKGROUND

The findings presented in this chapter shouldbe understood in relation to the various socialphenomena that made an assessment of life-sustaining technologies timely in the first place.The historical context of this study is a stressfulone, in which many things are changing rapidlyand dramatically. The speed of technological ad-vance is unprecedented, the elderly populationis growing geometrically, health care is beingtransformed. The words and concepts that arepart of this scenario—quality of life, autonomy,euthanasia, suicide, rationing, doctor-patient rela-tionship, malpractice, old age-evoke strong, oftenconflicting, responses. Other important conceptsare distinguished by their unfamiliarity: advancedirective, living will, durable power of attorney,surrogate decisionmaker, prospective paymentsystem, brain death. In this fluid environment,lags are inevitable: between new knowledge andits adoption, between technical capability and deci-sionmaking guidelines, between medical practiceand legal protections.

In other parts of this report (especially chs, 2,3, and 4), many of these concepts and trends arediscussed in depth. They arise in the context of

patients’ legal rights and ways to exercise them;the cost of health care and efforts to contain them;how medical technologies are developed and ac-cepted into practice; ethical bases for allocatinghealth care resources; ethical and legal issues con-cerning the withholding and withdrawal of treat-ments that sustain life; increased presence of thelaw and economics in medical practice; attitudesabout illness, death, and dying; growth of theelderly population; and the emergence of geriat-rics as a specialty within medicine, nursing, andother health professions. The background infor-mation presented in this chapter only suggeststhe range and importance of the social issues thatdrive concern about life-sustaining technologies.

The Specter of Rationing

The looming national debt and efforts to reduceit draw public attention to and impose new con-

21n this as5es5ment, “withholding)) is a decision to not initiate alife-sustaining intervention for a particular patient. Withholding maybe the enactment of the patient’s express wishes or the judgmentof other persons that the application of medical technolo~v is notwarranted. “Withdrawal” means the discontinuance or removal ofa life-sustaining intervention that has been initiated.


straints on older questions about the allocationof public resources in general and health care re-sources in particular. At the global level, the totalresource pool must be divided among all compet-ing national interests, i.e., health, defense, edu-cation, foreign aid, the environment, crime, andso on. At the next level, health care resources (in-cluding financial, human, and technological re-sources) must be allocated among a myriad ofpotential beneficiaries and causes. Here the com-petition is between prevention and cure, acutecare and long-term care, research and services,etc. Finally, at the micro-allocation level, specifichealth care resources must be distributed amongthe individuals who claim them. If there are 3 bedsin an ICU and 4 patients, or 10 donor kidneys and20 patients awaiting transplantation, difficult de-cisions must be made. At every level, our currentfiscal consciousness intensifies the need to makewise choices—and to be able to demonstrate thebenefits.

Many people see the present economic climateas a harbinger of inevitable rationing of scarceresources. In some circles, there is discussion ofexplicit criteria for allocating resources based, forexample, on age, prognosis, or cost. Elsewhere,rationing is rejected outright as unnecessary and/or evil. Other solutions can be found, it is argued,if priorities are adjusted at the global level anddemand for health care resources is modified (e.g.,by improving disease prevention and eliminatingthe use of unnecessary medical procedures).Whether one favors or abhors health care ration-ing-or believes it is already here—the strong re-action this concept evokes is one of the major rea-sons for concern about high-technology healthcare.

The “High Cost of Dying”

Considerable attention has been drawn to thehigh cost of health care for the elderly popula-tion (in 1984, annual personal health care expend-itures for Americans over 65 were projected at$120 billion, almost half of which would be paidby Medicare) and, in particular, to high Medicareexpenditures for patients in the last year of life.The latter has been interpreted and widely re-ferred to as the “high cost of dying.” The implica-tion has been that a great deal of money, in fact

“too much” money, is spent on patients who areelderly, and too much of this on patients who dieanyway. These figures have captured consider-able attention and led many people to ask whetherthe benefits justify the cost. Further, because itis widely assumed that life-sustaining technologiesare a major factor in the cost of care for personswho die, the value of this kind of treatment is oftenquestioned. Projected increases in the elderly pop-ulation and the increased costs these portend in-tensify the debate about what level of care is tobe provided at public expense.

Concern about the “high cost of dying” persistsdespite recent analyses that put this cost in a differ-ent perspective. First, understandably, the costof care is highest for people who get the most care,that is, those who are the sickest. Thus, what somedecry as the high cost of dying others recognizeas simply the cost of health care for very sick peo-ple, some of whom live, some of whom die, andmany of whom are elderly. Equally important,analyses of Medicare expenditures show that themajority of elderly people who die do not incurhigh Medicare costs in their final year. And, ofthose elderly patients whose health care costs arevery high, while approximately half die, the otherhalf survive. Analysis of Medicare expendituresover the past 20 years also shows that the rateof increase has been about the same for patientswho survive as for those who die, suggesting thatthe increase in expenditures is not due to dis-proportionate use of expensive life-sustaining tech-nologies for those who die.

In 1983, to contain high Medicare expenditures,Congress mandated a new basis for payment ofinpatient hospital claims. Under Medicare’sPart A prospective payment system (PPS), paymentfor inpatient hospital care is based on predeter-mined amounts for patients in given diagnosticcategories. Hospitals thus may show profit or loss,depending on their ability to keep their costswithin the established payment limits. Hospitalsand the physicians they employ now have strongeconomic incentives to be more selective in thetype, amount, or duration of treatment providedto Medicare patients, especially those whose costof care is likely to exceed available payment. Earlystudies of the effects of PPS reveal that the aver-age length of stay in hospitals has continued its

10 . Life-sustaining Technologies and the Elderly

pre-PPS decline. While the potential cost savingsto Medicare are significant, serious questions havebeen raised about possible negative effects on ac-cess to and the quality of care.

Quantity v. Quality of Life

Advances in medical technologies provide con-siderable ability to alter the timing and circum-stances of death. Indeed, modern diagnostic andtherapeutic technologies have changed the verydefinition of death and have influenced both pro-fessional and popular expectations. Recognitionof the manipulability of death enables us to pre-sume a significant measure of control and to con-template a death that is more or less “acceptable. ”

Questions about life-sustaining medical care fre-quently revolve around judgments about whatconstitutes acceptable “quality of life” (and, im-plicitly at least, “quality of death”) and deep-seatedbeliefs about the relevance of this consideration.Evacuations of “quality” are subjective and per-sonal; what is an acceptable quality of life to oneperson may be a fate “worse than death” to another.Similarly, life-sustaining treatment that somewould gladly endure, others would reject as “tooburdensome” or “undignified.” Thus, it is clearthat references to the quality of life must distin-guish whether the referrent is the patient’s uniqueexperience and evaluation of their own life or thevicarious experience and assumptions of someother person.

SELECTED

Summarized below are the findings OTA deemsmost significant either because they relateuniquely to elderly persons, affect large numbersof citizens, have legislative implications, or makeoriginal contributions to the debate about life-sustaining technologies. The findings are pre-sented under four general categories: 1) currentand future resource use; 2) quality of care; 3) ac-cess to care; and 4) decisionmaking problems andprocesses. Further information on all these topics,as well as many more specific findings, appearin chapters 2 through 10 and in background papersassociated with this assessment.

Many people believe that life, whatever its qual-ity, is sacrosanct. Under this view, the possibilityof sustaining life justifies, or even dictates, the useof all potentially effective means. In contrast, manyother people believe that the present and expectedfuture quality of life are valid, even essential, con-siderations in decisions about whether or not toapply life-sustaining treatments. These fundamen-tal disagreements about quality v. quantity are fre-quently expressed in the terms of treatments that“prolong life” v. treatments that “prolong dying.”In fact, the distinction between prolonged life andprolonged dying is like the difference betweenthe proverbial glass that may be seen either ashalf full or half empty. The actual referrents arethe same. (In this assessment, OTA uses the terms“prolonged life” and “prolonged dying” only whenquoting other sources.)

Accompanying new attitudes toward death, andcontributing to them, is the dramatic shift in theplace of death. While the majority of deaths usedto occur at home, by 1984,61 percent of all deathsin this country occurred in hospitals and othermedical centers. This shift has major implicationsfor the types of care available to patients, the iden-tity and number of persons involved in their care,and the kinds of decisions that must be made. Iron-ically, while hospitals were once feared as “placesto die” because so little could be done to avertdeath, some people now fear hospitals as placesto die because so much can be done.

FINDINGS

For the most part, the findings presented hereapply to all of the technologies OTA studied—butthey would not have been evident, or not docu-mentable—without focusing on individual tech-nologies. Thus, an overriding conclusion of thisproject is that assessments of individual tech-nologies can provide information for bothpublic policy and clinical decisionmaking thatabstract considerations of life sustaining tech-nology cannot. Future studies and debate abouthealth care decisionmaking might usefully adoptthis more focused approach.

Ch. 1—Summary and Policy Options ● 1 1

Current and Future Resource Use

Finding: Data on current utilization of life-sustaining technologies are highly unreli-able. Future utilization cannot be accu-rately predicted.

OTA’S attempt to estimate the utilization of fivelife-sustaining technologies reveals, above all,shortcomings in the available data and existingdata collection systems. With the exception of thedata collected and maintained by the Health CareFinancing Administration (HCFA) on Medicare’sEnd Stage Renal Disease program, reliable dataon the numbers of patients are not available.

Estimates of the total number of patients of allages and the number of elderly patients treatedwith dialysis, resuscitation, long-term mechani-cal ventilation, and nutritional support are shownin table 1-1. Total utilization ranges from a fewthousand persons, in the case of mechanical ven-tilation, to 1.4 million persons, in the case ofnutritional support. Utilization among elderly per-sons ranges from approximately 2,200 for venti-lation to 680,000 for nutritional support. With theexception of the dialysis data, these figures shouldbe regarded as preliminary, probably minimal, in-dicators of the size of the respective patientgroups. The dialysis data are taken from HCFArecords; the other data are based on a combina-tion of industry estimates, published reports, andOTA contractor reports, and were compiled byOTA.

For life-sustaining antibiotic therapy, numeri-cal estimates of utilization are too tentative to re-port. Although some data exist on the use of anti-biotics in general, the number of cases in whichtreatment is life-sustaining, and the number ofpatients who are elderly, cannot be estimated.

Differences in data collection methods, defini-tions, time periods, etc., dictate special cautionin comparisons of data for the individual life-sustaining technologies described in this report.(The reader should not conclude from table 1-1,for example, that 1 in 100 resuscitated patientsrequires prolonged mechanical ventilation or that20 times as many people are treated with dialysisas mechanical ventilation.) The figures reportedfor mechanical ventilation are cross-sectional data;they do not reflect the fact that new morbiditycreates a constant stream of patients, i.e., the pa-tients on mechanical ventilation at the time thesedata were collected might be replaced severaltimes over during the course of a year. The datafor dialysis, on the other hand, represent all pa-tients treated during a calendar year.

Also, patients with life-threatening medical con-ditions may be treated, simultaneously or sequen-tially, with several life-sustaining technologies.Many ventilator patients require nutritional sup-port, and it has been estimated that 45 percentof all infections acquired in hospitals (nosocomialinfections) are related to medical devices. Thus,totaling the number of patients receiving each ofthese life-sustaining technologies would overstate

Table 1.1 .- Utilization of Life-Sustaining Technologies for Patients of All Ages,and for Elderly Patients, in All Settings Combined

Total number Patients over 65

of patients Percent(all ages) N u m b e r of tota l

Dialysis a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90,621

Resuscitation b . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370,000to 750,000

Mechanical ventilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 , 7 7 5C

to 6,575 d

Nutritional Supportf . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,404,500Enteral (tube) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 848,100Parenteral (intravenous) . . . . . . . . . . . . . . . . . . . . . . . 556,400

27,641 31%0

204,000 550/0 est.to 413,000

1,250C 34 ”/0to 2,200d,e

680,000 48°/0 est.450,000 530/0 est.230,000 40°/0 est.

alg&j HCFA data for Medicare’s ESRD Program.bcontractor es.tirnates, hospitalized patients W’W’.c1 985 data for w states, patients dependent on ventilator 14 days or 10n9er.dNational estimates extrapolated from survey in MassachusettseElderlY defined as over 70flg84 lndust~ data and contractor estimates.

SOURCE Office of Technology Assessment, 1987

12 . Life-Sustaining Technologies and the Elderly

the number of patients receiving any life-sustain-ing technology. In addition, the data in table 1-1leave out patients who were treated, but toobriefly to appear in the figures (e.g., patients ven-tilated for less than 14 days).

Future demand for life-sustaining technologiescannot be predicted without accurate informa-tion on current utilization and monitoring ofchanges in use. This problem is vividly illustratedby the brief history of Medicare’s ESRD program.Prior to Medicare coverage for dialysis, the num-ber and distribution of dialysis machines and per-sonnel were so limited that treatment was avail-able only to the wealthy and the hand-picked;patients over age 45 were seldom considered. Fol-lowing enactment of Medicare’s ESRD program,the number of dialysis patients of all ages climbedfrom 5,000 to over 90)000 between 1972 and 1985(and the number of patients over 65 multipliedby at least a factor of 25)—figures far in excessof the original projections.

Future utilization of life-sustaining medical tech-nologies will be influenced by a number of fac-tors, some of which work in opposite directions.The aging of the population, improvements in thetechnologies, and availability in new settings willdrive increased demand. Although these increasesmay be great, they are likely to be tempered bycost-containment measures, preventive strategiesfor specific diseases, changes in procedures andguidelines for treatment decisions, and changesin public attitudes. Increasingly, cries for “deathwith dignity” and the “right to die” are associatedwith the rejection of high-technology interventionsnear the time of death.

Finding: For resuscitation, mechanical ven-tilation, dialysis, nutritional support, andlife-sustaining antibiotic therapy personsage 65 and older constitute large propor-tions of all patients, but small proportionsof the total elderly population.

This finding can be stated with confidence de-spite the numerous caveats about specific num-bers. While persons 65 and older constitute about11 percent of the total U.S. population, they com-prise over 30 percent of all patients receiving di-alysis, nutritional support, and mechanical venti-lation (see table l-l). In hospitals, an average of

55 percent of all patients who are resuscitatedare elderly. In addition, because elderly personsare known to be at the highest risk for life-threat-ening infections, it is reasonable to assume thatthey also comprise a large proportion of individ-uals receiving life-sustaining antibiotic therapy.

It is important, however, to keep these findingsin perspective. While the vast majority of nurs-ing home patients receiving nutritional supportare elderly, only 5 percent of all elderly personsare in a nursing home (at any one time), and onlya small proportion of nursing home residents (2to 5 percent) receive nutritional support. Theproportion of elderly persons who receive otherlife-sustaining technologies is much smaller.

Finding: The costs associated with life-sus-taining interventions are uncertain, butcertainly high.

In general, available data on the costs of life-sustaining technologies are piecemeal and notcomparable. The best data are those compiled byHCFA on the ESRD program. For the other tech-nologies OTA studied, even the concept “cost” hasbeen interpreted inconsistently, depending onwhose costs are of concern. Thus, some publica-tions that claim to report “costs” actually describewhat economists call “charges” (i.e., billed amount)or “expenditures” (i.e., payments). Some reportsinclude in their accounting only the specific serv-ices and supplies essential to the life-sustainingtechnology; others count the total cost of the hos-pital stay during which a life-sustaining technol-ogy is used. There has been no attempt to quan-tify the full economic impact using a definitionof costs that includes factors like lost income ofthe patient or of family caregivers. What is clearis that the costs to providers, charges to patients,and expenditures by patients and third-parties forlife-sustaining technologies all are high.

The total cost of care is closely associated withhow long the life-sustaining technology is needed.Less obviously, the costs associated with theinitial life-sustaining intervention may bedwarfed by the ongoing costs associated withsurvival of patients whose health care needsremain great despite or because of the inter-

Ch. 1—Summary and Policy Options • 13

vention. This is the case, for example, for se-verely debilitated people who acquire a life-threatening infection that is effectively treatedwith antibiotics and who subsequently require anextended stay in a nursing home. Their health andquality of life may remain poor, despite continu-ing institutionalization and health care.

Another major correlate of cost is the settingin which care is provided. (It must be recognized,of course, that the services, equipment, and ex-pertise available in hospitals v. nursing homes v.the patient’s home are not the same.) It is gener-ally assumed that cost (along with charges andexpenditures) is highest in the acute care hospi-tal and lowest at home. The movement of high-technology care outside of ICUs and outside ofhospitals altogether has been encouraged by,among other things, efforts to reduce health careexpenditures. For patients whose needs can bemet by a combination of self-care and unpaid fam-ily members, with only occasional professional at-tention, the charges and expenditures for homecare are certainly below those associated with hos-pital care. However, if round-the-clock profes-sional nursing and other attributes of intensivecare are needed, it can actually cost patients andpayers less to keep the patient in the hospital ICUthan to try to ‘(bring the intensive care unit intothe home.” Similarly, care in a nursing home some-times costs less than care at home.

Available data on charges associated with theuse of three life-sustaining technologies in the hos-pital and in community settings (including homecare and other community settings), as reportedin published studies and OTA contractor reports,are summarized in table 1-2. These data show thewide range in charges for one technology versusanother, for hospital versus community care, andfor different patients within each setting. Dailycharges for life-sustaining treatments range from$4 to $500 for different forms of nutritional sup-port. The most expensive of these technologiesis mechanical ventilation, with average daily hos-pital charges of more than $800.

For life-sustaining antibiotic therapy and resus-citation, available data are particularly sketchy.Intravenous antibiotics are estimated to cost $30to $200 per day, exclusive of the cost of any profes-

Table 1.2.—Charges for Life-Sustaining Technologies

Hospital inpatient Community settingDialysis

Per treatment – $68-$200Per year. . . . . . . . – $20,000-$30,000

Nutritional supportEnteralPer day . . . . . . . . . . . $4-$132a b .Per year, ., ,$1,450-$28,200 $3,000-$12,000ParenteralPer day . . . . . . . . $25-$500b —Per year. . . . . . . . . ., ,$9,125-$182,500 $50,000-$100,000

Mechanical ventilationPer day . . . . . . . . . $824C —Per year, ., . . . . . . . . . . $300,760C

$21,235-$216,000aoally flospltal Charges for enteral nutrition average $43, for parr?nk?rd nUtrltlOn the avera9e char9e

IS $196 Der davblncludes” formu~a, equipment and staff Iime, not hospital stay. 1985 datacAverage charges, Including hospital stay, for patk?nk In 37 States. 1985 data

SOURCE Office of Technology Assessment, 1987

sional services or institutionalization. For resus-citation, OTA found no reliable cost estimates at all.

Until accurate data are available on the costsand utilization of life-sustaining technologies anduntil the factors that alter cost and utilization arebetter understood, health care planning and publicpolicy will be uninformed. Accurate baseline dataand projections of demand for life-sustaining treat-ments are basic to planning of health care facil-ities, professional training, community resources,technological research and development, and de-cisions about coverage and reimbursement, in-cluding catastrophic health insurance plans. Bet-ter information is also a prerequisite to seriousdiscussion about the need for, or criteria to beused in, rationing of access to health care.

Finding: Reimbursement is a major deter-minant of specific treatment options.

Most of the five technologies OTA studied en-compass several treatment options, more than oneof which might be suitable for a given patient.For some patients with chronic renal failure, ei-ther transplantation or dialysis might be appro-priate, and then, more than one method of dialy-sis might be effective. For some patients whorequire ventilator support, either positive pres-sure or relatively simple, negative pressure de-vices might be appropriate; similarly, for some pa-tients, nutritional support and antibiotic therapy

14 ● Life-sustaining Technologies and the Elderly

may be provided effectively by any one of sev-eral routes. The availability and level of reimburse-ment for certain technologies not only influenceincentives for treatment v. nontreatment; they alsoinfluence the relative utilization of different treat-ment options. For example, some nutritional sup-port experts believe that higher reimbursementfor TPN results in its use for some hospitalizedpatients for whom tube feeding would be an ef-fective, less expensive option.

The availability and level of reimbursement alsodetermine which settings are available, sometimesencouraging inefficient use of resources or pre-cluding use of the least restrictive environment.Between 1972 and 1982, the reimbursement struc-ture of Medicare’s ESRD program encouragedcenter- and hospital-based dialysis over home care.Medicare coverage for nutritional support of lessthan 90 days and Medicare coverage for drugs,including intravenous antibiotics, is not availableto patients at home. The lack of reimbursementfor short-term nutritional support and intravenousantibiotics creates purely financial reasons for con-tinued hospitalization. Similarly, reimbursementfor TPN and for mechanical ventilation is some-times so much more complete for patients whoremain in the hospital that some patients whoare well enough to go home cannot afford todo so. The number of hospitalized elderly (andyounger) patients needing life-sustaining technol-ogies who could be safely treated in communitysettings is unknown.

Finding: The expansion of life-sustainingtechnologies to settings other than theacute care hospital has major implicationsfor who and how many will receive treat-ment.

Currently, the numbers of elderly patients re-ceiving life-sustaining treatments in their ownhomes, in nursing homes (tube feeding and an-tibiotics are exceptions), and in other nonhospi-tal settings are relatively small; the overall num-bers have been increasing, however, and manyobservers predict that this trend will continue.If life-sustaining technologies become widely avail-able in nursing homes and patients’ homes, theymay be offered more readily, to more patients anddifferent kinds of patients, and they may also be

more readily accepted by patients who now wouldrefuse them. Some observers warn that increasedavailability of life-sustaining technologies in non-hospital settings, especially if it is accompaniedby increased reimbursement, could lead to seri-ous overuse.

In general, patients who can be cared for in theirown home enjoy benefits that contribute to theirquality of life. In contrast to patients in the morerestrictive and strange environments of hospitalsand their ICUs, some chronic ventilator patients,home dialysis patients, and home nutritional sup-port patients retain a certain amount of independ-ence, despite physical dependence on technology.Even for patients whose functional ability is se-verely limited, care in their own home allows themto maintain considerable control over their healthcare and other aspects of their life, including so-cial relationships,

The number of elderly persons who can bemaintained on life-sustaining technologies intheir own homes is limited. Complex home carerequires, at a minimum, a patient who is medi-cally stable and cooperative, capable and dedicatedfamily members or companions, a suitable physi-cal environment, support services in the commu-nity, and adequate reimbursement or personal fi-nancial resources. These conditions, difficult forpatients of any age to meet, probably precludemost elderly patients, For mechanical ventilation,about 34 percent of patients of all ages, but only14 percent of elderly patients are cared for in theirown homes. It must be recognized, however, thatthe feasibility of home care for elderly patientsvaries with the technology that is needed. For TPN,20 percent of home care patients are elderly. And,for tube feeding, it is estimated that as many as55 percent of all home care patients are elderly.

There are numerous impediments to the optimaldistribution of patients across settings. Some pa-tients who could be safely transferred to nonhospi-tal settings remain in hospitals, often indefinitely,because caregivers are not available for home careor because of a lack of services and facilities withintheir community. There is a scarcity of nursinghome beds for technology dependent patients be-cause few nursing homes have adequate staff (or

Ch. 1 —Summary and Policy Options ● 15

adequate incentives to develop staff) to providethe level of care these patients require. Some phy-sicians and institutions remain unaware or uncon-vinced about the home care option and do notpresent it to patients. Some patients who havebeen discharged home have been forced to re-turn to the hospital because of superior reimburse-ment in that setting. Information and service net-works need to be developed to help ensure thatall settings that are medically safe receive consid-eration.

Finding: For many patients, life-sustainingtreatment in the acute care setting createsthe need for chronic care and continuingtechnological support.

Because life-sustaining technologies seldom curethe underlying condition or restore normal phys-iological functioning, some patients who survivean acute intervention require continuing treat-ment for the rest of their lives. Acute dialysis oracute ventilation may evolve into prolonged,chronic, or permanent need for these technol-ogies, with or without potential for rehabilitation.Chronic dependence on a life-sustaining technol-ogy is accompanied by continuous need for serv-ices or facilities that are typically both expensiveand scarce.

Individuals who must remain institutionalizedoccupy abed in the ICU, hospital, or nursing home,utilizing facilities, personnel, equipment, and otherresources for which other patients may be com-peting. Individuals who are able to return to thecommunity have needs that include reliable sourcesfor medical equipment and supplies, professionaland nonprofessional caregivers (including familymembers and other assistants), and maintenanceand repair of equipment. One aspect of this con-tinuing need is the high, and ongoing, cost of care.Another crucial aspect is that the necessary serv-ices and the linkages to coordinate them are un-available in many communities.

A patient’s need for long-term technological sup-port is often difficult to predict, but this possibil-ity must be recognized when the initial decisionto provide acute care is made. Some argue thatit is unethical to provide health care to acutely

ill patients if society lacks the commitment alsoto provide chronic health care and related serv-ices—especially if those needs were created bythe acute intervention. The discontinuity in ex-isting health care services leaves some technology-dependent patients and their families in a predica-ment that they did not foresee when faced withthe initial treatment decision.

Coordinated systems of care for technology-dependent persons exist in some European coun-tries, and these models may be instructive. InFrance and in England, for example, systems arein place to provide comprehensive services thatenable chronic ventilator patients to remain intheir communities. These are regional programsthat provide services ranging from group purchas-ing of medical supplies to equipment repair, pa-tient education, and emergency care. In existencefor more than 20 years, these systems are saidto be economical and to improve the quality oflife for these patients.

Quality of Care

Finding: There are some questions aboutthe quality of care related to the use of life-sustaining technologies, particularly forelderly patients.

Although OTA did not specifically seek infor-mation related to the quality of care, some issuesemerged. Perhaps most important, there is am-ple evidence that some treatment options orprocedures should be tailored to age, but thereis little evidence that they are. Despite the factthat age-related changes in the metabolism ofdrugs are now well recognized, for most antibi-otics, dose and dose interval remain standardregardless of the patient age. Similarly, althoughit is well established that nutritional requirementschange with age, the details are not well under-stood, and nutritional formulas are frequently notadjusted to these changes, especially for patientson tube feeding. For the other life-sustaining tech-nologies OTA studied, the possibility that the clin-ical outcomes for elderly patients might be im-proved if modifications were made either in theequipment or procedures has barely been ad-


dressed. Modifications in these treatments to ac-count for age-related differences depend on con-tinuing basic research in gerontology and otherfields and on dissemination of this knowledgethrough professional training.

The traditional bias of medical education andpractice places the cure of acute illness above allother goals. When cure is not a realistic goal, thisapproach often leads to inappropriate treatmentdecisions. Recent changes in the curricula of manyhealth professions recognize this problem andseek to improve care by acknowledging and focus-ing on achievable goals, such as maximization ofthe patient’s functional capacity and the qualityof life. Pertinent curricular innovations includenew courses in geriatrics, medical ethics, human-ities, and death and dying.

Other quality of care issues result from shiftsin the settings in which life-sustaining technologiesare applied and changes in the personnel who areresponsible for care. In nonhospital settings, re-sponsibility for patient care is often entrusted toless trained professionals and to laypersons. Theeducation and supervision of patients providingself-care, family members, and other lay care-givers, as well as home health care professionalsare important issues. Health care personneltrained in the use of complex technologies havetypically not been trained to work in communitysettings or to work with elderly persons. Mainte-nance and repair of equipment and availabilityof backup equipment can also be problems whenlife-sustaining technologies are used outside thehospital.

A different kind of quality of care issue con-cerns the technological hardware for certain life-sustaining technologies; this includes the primarymedical device as well as the various peripheralsupplies and components (e.g., tubing, solutions,power sources). Questions have been raised aboutthe quality, safety, and suitability of some enteralformulas for nutritional support. Also, the Foodand Drug Administration (FDA) has received alarge number of reports of mechanical ventila-tors that have malfunctioned or failed. In somecases, only voluntary standards apply to the man-ufacture of devices and products used to sustainlife.

Finding: Technological developments haveimproved the safety and efficacy of life-sustaining technologies, as well as thequality of life for some patients who aredependent on them.

Research and development (R&D) in manyarenas, including physiology, medicine, engineer-ing, electronics, biofeedback, and computer sci-ence, have brought continuous change in exist-ing life-sustaining technologies as well ascompletely new technologies to sustain life (seeapp. c). Stimulated by competition for health caremarkets, perceived need to improve availableproducts, regulatory standards, etc., R&D withinthe public and private sectors has resulted in moreand better devices and methods for diagnosing,monitoring, and treating severely ill patients inboth traditional and nontraditional settings.

General technological advances (e.g., miniatur-ization, computerization, new materials, and auto-mation) have made possible improved efficacy,safety, and reliability of many medical devices. Oneexample is the development of automatic bloodgas analyzers, considered a watershed in mechan-ical ventilation technology. Other kinds of tech-nological developments have meant improvedcomfort and independence for some patients. In-novations that reduce the size and weight of equip-ment, extend time between treatments, reducethe need for professional services, or make homecare possible enhance the quality of life for manypatients. Improved blood access systems for hemo-dialysis are a good example. Prior to developmentof the Teflon shunt in 1960, patients had to un-dergo the inconvenience, discomfort, and risk ofinfection associated with having a new surgicalprocedure for every dialysis treatment.

A potential benefit of continuing R&D is costreduction, New methods of manufacture, new ma-terials, and new markets may lower the produc-tion cost of certain equipment and supplies. If low-ered production costs are reflected in prices orin reimbursement, this would result in lower treat-ment costs for some patients. Existing incentivesto develop medical technologies that are less ex-pensive and incentives to substitute lower forhigher cost technologies appear largely tied to in-terest in the home health care market.


Access to Care

Finding: When resources are available, pa-tients with life-threatening conditions aremore likely than not to receive aggressivetreatment.

The acute care orientation in medical trainingand practice emphasizes cure and prolongationof life and justifies “doing everything humanly pos-sible” to achieve these goals, This bias to treat ap-pears to prevail for patients of all ages. It has beenreinforced by the wide availability of life-sustain-ing technologies in hospitals, reluctance to con-sider cost as an appropriate factor in individualdecisionmaking, health professionals’ and institu-tions’ fear of legal action, and the weighty uncer-tainties surrounding treatment decisions. Sincea wrong decision is irreversible, most health pro-fessionals would choose to “err on the side of life.”

While withholding of treatment is resisted, with-drawal may be even more so. Despite wide agree-ment among ethicists and legal scholars that thereis no theoretical basis for distinguishing betweenwithholding and withdrawal of life-sustainingtechnologies, in actual practice it is frequentlyeasier to withhold a life-sustaining treatmentwhose benefit is uncertain than later to “pullthe plug, ” even when the patient or patient’ssurrogate requests this. Grief, guilt, and healthprofessionals’ feelings of failure at times preventrational decisionmaking.

Some health professionals and family membersview withdrawal of aggressive medical treatmentas “giving up” or even “abandonment” of the pa-tient. On the other hand, some believe there isa greater moral imperative to withdraw treatmentthat proves to be futile or unwanted than to initi-ate an intervention that is of uncertain value. Thisposition emphasizes the need for continual reeval-uation of the medical indications for treatment.Some persons who hold this view advocate theuse of time-limited trials. For example, mechani-cal ventilation could be instituted with the provi-sion that its use be reconsidered after 1 week;dialysis could be tried for 4 months, etc. Aftera designated trial period, the patient’s situationcould be thoroughly evaluated; there would bean opportunity to assess the value of treatmentand to ascertain the patient’s wishes.

In addition to philosophical and psychologicaldifficulties, practical difficulties at times discour-age the withdrawal of life-sustaining technologies.To withdraw most life-sustaining treatments re-quires a specific physician order, frank and time-consuming conversations with the patient and/orfamily, conferences among members of the healthcare team, and formal documentation in the pa-tient’s record. At times, institutional review com-mittees, ethics committees, legal advisers, or thecourts become involved in decisions to withdrawtreatment. While decisionmaking procedures varywith the technology being considered, the deci-sion to withhold treatment is generally less ex-plicit than the decision to withdraw it.

Finding: Relative access to life-sustainingtechnologies by different segments of thepopulation cannot be assessed with avail-able data.

Health professionals’ preference to providerather than to withhold treatment and to with-hold rather than withdraw it are competing bi-ases whose impact on access to life-sustainingtreatments is not clear. Many other factors, nota-bly reimbursement, also influence accessibility ofhealth care and determine whether or not vari-ous segments of the population have equal access.Between 1965 and 1983, Medicare’s cost-basedreimbursement system facilitated the develop-ment and diffusion of medical technologies in gen-eral, and made life-sustaining technologies avail-able to hospitalized elderly patients with littleregard to cost. It is not yet clear what impact Medi-care’s prospective payment system for hospitalcare has had on accessibility of life-sustaining treat-ments. Available utilization data prove that elderlypersons have considerable access to life-sustainingtreatments, but utilization data alone do not per-mit conclusions about whether access is restricted(leading to undertreatment) or excessive (leadingto overtreatment).

Public opinion and concerns expressed by healthprofessionals suggest that overtreatment—i.e.,provision of treatment that is or becomes un-wanted or unbeneficial—is more frequent thanundertreatment. In 1985 the National Institutesof Health cosponsored a conference on Withhold-ing and Withdrawing Mechanical Ventilation in


response to wide agreement among clinicians thatthe technology is too often started and too oftencontinued inappropriately. It should be noted,however, that because treatment is easier to countthan nontreatment, overuse is probably more vis-ible than underuse.

Cost-containment pressures in general andMedicare’s prospective hospital payment sys-tem in particular force health care decision-makers to acknowledge that resources arelimited and that all patients cannot have“everything possible.” The pressure to reducecosts has spawned legitimate concerns amonghealth professionals and the public that every pa-tient will not have everything that is desirable.In the absence of guidelines for how costs are tobe reduced, it is unclear which patients will beaffected the most. Since Medicare is a programfor elderly citizens, however, the patients mostdirectly affected by hospitals’ and physicians’ ef-forts to reduce health care costs under Medicareare those over 65.

It appears that questions about equality of ac-cess should not just make the usual comparisonsof rich and poor, old and young, or black andwhite. pertinent concerns also include setting, cog-nitive ability, and age subgroup. Anecdotal evi-dence and small studies suggest that a nursinghome resident with a life-threatening infection isless likely to be treated than if that same personwere in an acute care hospital; persons with se-verely impaired cognitive ability—whose qualityof life is perceived to be poor and who cannotspeak for themselves —are also less likely to re-ceive aggressive treatment; relatively youngelderly persons and those who have a spouse aremore likely to be treated than those who are olderor alone.

Since 1983, evidence of changes in hospital ad-mission policies and the continued reduction inlength of stay suggest that limited Medicare pay-ment may have begun to influence treatment op-tions that are made available. Some Medicare pa-tients whose treatment costs are expected toexceed payment for their diagnosis-related group(DRG) have been dubbed “DRG losers,” and thereis mounting anecdotal evidence that some personshave been denied admission to certain hospitals

or denied admission to the ICU.3 Despite finan-cial incentives to limit expensive care, how-ever, there is no evidence to date that PPS hasreduced access to life-sustaining treatment.

As cost-containment measures are implementedin Medicaid and in private health insurance pro-grams, patients of all ages are more likely to re-ceive reduced care. It remains to be seen whethersavings are or will be found by cutting servicesto all patients or by cutting services to particulargroups of patients. There is wide agreement that,under PPS, Medicare patients are being dischargedfrom hospitals “quicker and sicker.” At the sametime, however, Medicare patients who are retainedin hospitals are also sicker and older than beforePPS. The meaning of these findings and the ex-tent to which they are caused by PPS is a subjectof considerable debate that is outside the scopeof this assessment.

Finding: For patients who do not want life-sustaining technologies and patients forwhom these technologies are not medi-cally indicated, treatment options havebeen relatively unexplored and are notwidely available.

Treatments whose goal is to control pain andsuffering, even at the risk of hastening death, areregarded by many people as reasonable alterna-tives to aggressive life-sustaining medical treat-ment. There is anecdotal evidence, however, thatpatients who refuse life-sustaining treatment thatis offered and patients from whom aggressivetreatment has been withheld or withdrawn aresometimes neglected by health professionals. Per-sons capable of providing alternate forms oftreatment-especially hospice care and palliativeor supportive care —may not be available. Alsothere are legal and ethical uncertainties regard-ing when and how it may be appropriate to limittreatment. Medicare reimbursement for hospicecare is currently available only in special circum-stances, only to patients who have been diagnosedas “terminally ill” and then, of course, only wherehospice facilities and/or personnel are available.

3A study in one hospital found that Medicare pptknts in sPeci”fied circulatory system DRGs, who were treated in the ICU, resultedin losses to the hospital ranging from $674 to over $24,000 per dis-charge. Such dramatic effects hate attracted considerable attentionamong health professionals and institutions.


Decisionmaking Problems andProcesses

Finding: Decisions about the use of life-sus-taining technologies are made amid greatuncertainty regarding the likely clinicaloutcomes.

Decisions about whether or not to institute life-sustaining treatments would be relatively easy ifit were known in advance whether or not the pa-tient would survive, for how long, and in whatcondition. But, variations in patients’ physiologi-cal and psychological adjustment, and in the qual-ity of care they receive, make highly uncertainthe outcomes of any treatment for any given pa-tient, Pervasive prognostic uncertainty means itis impossible to predict whether or not any treat-ment will be effective, whether a particular treat-ment is optimal, or whether a patient would sur-vive without treatment.

The inability to prospectively identify patientswho will benefit from treatment arises because,contrary to popular belief, life-sustaining tech-nologies are frequently ineffective. For acutelyill patients of all ages, aggressive treatment isassociated with high mortality and seriouscomplications. At best, one-third to one-half ofall in-hospital resuscitation attempts succeed; andonly one-half of the patients who are successfullyresuscitated survive long enough to be dischargedfrom the hospital. In acute episodes of respira-tory failure, adults treated with mechanical ven-tilation have about a 50-percent chance of sur-viving; for acute renal failure, only 20 percent ofpersons over age 70 survive. Patients receivingantibiotic therapy or nutritional support have arelatively high, but not necessarily predictable,chance of survival.

Prognosis is often especially difficult whenthe patient is elderly. The interaction of disease(especially multiple coexisting diseases) with re-duced physiological reserve makes diagnosis inelderly patients difficult and responses to treat-ment particularly difficult to predict. The clinicaluncertainties may be exacerbated by the short-age of basic scientific knowledge about aging andthe shortage of personnel trained in geriatricassessment and care.

Inability to accurately predict the outcomes ofparticular treatments can result in two kinds oferrors—i.e., treatment of patients for whom treat-ment is futile and failure to treat patients whowould survive. Reducing both kinds of errorswould not only avoid useless suffering for patientsand families, but is tantamount to more rationaland efficient use of health care resources. Studiesof the outcomes of critical care have shown thatthe cases in which costs are highest are those inwhich the outcome was inaccurately predicted.

Basic and clinical research are among the nec-essary approaches to reducing clinical uncertaintyand, thereby, to improving the content of treat-ment decisions. Information is needed about thephysiological and psychological responses ofelderly patients to particular treatments as wellas information about the outcomes without treat-ment. Dissemination of this information througheducation and training of health care professionalswould strengthen their ability to evaluate, and tohelp patients understand, the relative risks andbenefits of treatment options.

Research is underway on a variety of methodsto combine diagnostic and treatment data into sta-tistical categories that are associated with knownprobabilities of survival. Theoretically, reliableclassification systems could provide physicians animproved basis for predicting the outcome of treat-ment. An OTA workshop on such systems of pa-tient classification, held in conjunction with thisassessment, concluded that, although the currentstate-of-the-art is limited and systems remain ex-perimental, there is reason to believe that refinedpatient classification systems will effectively re-duce clinical uncertainty and provide valuable helpin making some kinds of treatment decisions.

Finding: For an individual patient, chrono-logical age is a poor predictor of the out-come of treatment with life-sustainingtechnologies.

The statistical odds of survival are worse forelderly than for younger adults who receive a life-sustaining intervention, but neither age 65—norany single age criterion—is an adequate predic-tor of physiological or psychological response to


treatment. Moreover, because physiological andpsychological diversity increase as people age, re-sponse to particular technological interventionsmay be hardest to predict in the oldest patients.

Available data for most of the life-sustainingtechnologies OTA studied substantiate that elderlypatients, as a group, have lower survival rates andmore complications. With dialysis, for example,the mortality rate among elderly patients is threetimes as high as that for all patients (45 v. 15 per-cent). On the other hand, elderly patients, on thewhole, seem to make a better psychological ad-justment to chronic dialysis than do younger pa-tients. Generalizations based on the patient’s age,while they may be statistically accurate, obscurethe fact that many individual elderly patients sur-vive and thrive after treatment with a life-sustain-ing technology.

For patients of all ages, life-sustaining technol-ogies are associated with numerous potentiallyserious complications. It has sometimes been as-sumed that elderly persons, as a group, are athigher risk of such complications and that the com-plications elderly patients experience are apt tobe more serious. In fact, data to support this as-sumption are inconclusive and vary with the tech-nology. For example, while increased risk of ribfractures is frequently mentioned in connectionwith resuscitation of elderly persons, OTA is un-aware of data to support this. Moreover, any sta-tistical association between age and rib fracturesis due not to age per se, but to age-related dis-eases that make the bones brittle (e.g., osteo-porosis).

To some degree, the worse outcomes of elderlypatients may stem from inadequate expertiseregarding aging and geriatric care. Health profes-sionals’ inattention to or misinterpretation of per-tinent clinical information can lead to unwar-ranted generalizations about elderly patients andto a self-fulfilling prophecy. If it is reasoned, forinstance, that an elderly person should not receiveaggressive life-sustaining treatment “because hewon’t do well,” he is almost certain to not do well!

Most of the patient classification systems OTAreviewed include chronological age as one varia-ble in the statistical prediction model. Even in theseabstract mathematical models, age contributes

less to the prediction than other patient char-acteristics, including severity of illness, diag-nosis, or previous health status. So great is in-dividual variability that some researchers and cli-nicians argue that the patient’s age should be dis-regarded in making treatment decisions. Othersadvocate development of a proxy forage that moreaccurately reflects the health status and reservecapacity of individual patients.

Finding: The legal and ethical uncertaintiesthat surround decisions about the use oflift+ sustaining technologies have led to in-tense interest in the development of deci-sionmaking supports and guidelines.

Profound ethical uncertainties in decisions aboutlife-sustaining technologies emanate from theplurality of cultural and religious orientations thatcharacterize this society and that affect people’svalues and beliefs about such fundamental thingsas the meaning of life and the meaning of death,individual v. public good, and the quantity v. qual-ity of life, Ethical quandaries may make it diffi-cult to discern the goal of the decision (e.g., pa-tient autonomy v. survival, etc.), the means toachieve it, or both.

Grave legal uncertainties arise because there aresituations in which no pertinent legislation exists,because legislation differs in different jurisdic-tions, and because the law is changing. Legalprecedent and case law offer valuable, but notalways consistent, guidance. Uncertainty aboutwhat actions are legal fuel health professionals’widespread fear of the law, and fear of malprac-tice litigation is an important factor in clinical deci-sionmaking. Some of this fear is well founded;some, however, results from health professionals’ignorance or misinterpretation of the law.4

Decisionmaking problems are made still morecomplex by the fact that, in most cases, there isnot one decision to be made (e.g., whether or notto start dialysis), but rather a series of decisions(e.g., whether to hospitalize, to do a particular diag-nostic test, to put the patient in the ICU, to con-tinue treatment, etc.). And, separate from the ques-

~See M.B. Kapp and B. Lo, “Legal Perceptions and hfedical Deci-sionmaking, ” prepared for the office of Technolo~V Assessment,LI.S. Congress, t\’ashington, DC, Xlarch 1986.


tions about what the decision should be are seriousquestions about how the decision should bereached. If, for example, the patient disagrees withmedical advice, what should be done? If the pa-tient is not decisionally capable, who shall be thesurrogate? The variety in patients’ physical sta-tus, decisionmaking capacity, severity of illness(emergency or not), social circumstances (espe-cially whether one is in the community or in aninstitution), and family situation (especiallywhether or not there is a designated surrogate)mean that no single approach to decisionmakingcan be applied in all instances. These difficultieshave stimulated legislative, institutional, and pro-fessional responses.

Possible roles of government in reducing theuncertainties surrounding decisions about life-sustaining technologies include Federal or Statelegislation and regulations and support for re-search. To date, the legal response has been pri-marily the enactment of new laws at the Statelevel. Living will laws have been enacted in 38States and the District of Columbia. All States andthe District of Columbia have durable power ofattorney statutes, and 15 States have statutes thatspecifically authorize the use of a durable powerof attorney for health care decisionmaking. Theseadvance directives protect the rights of patientsto participate in health care decisions even afterthey become decisionally incapable and, by clarify-ing the patient’s treatment preferences, offerhealth care providers a measure of protection aswell. Family consent laws, that specify the rightof family members to make treatment decisionsfor an incompetent person, are another option.In some States, courts have mandated specific pro-cedures that must be followed in decisionmakingabout life-sustaining technologies. Each form oflegal response does a partial job of solving theproblems that arise in decisions about life-sustain-ing treatment. The clinical and ethical dilemmas,of course, remain.

As recommended by the President’s Commis-sion for the Study of Ethical Problems in Medi-cine and Biomedical and Behavioral Research,some health care institutions, especially hospitals,have developed policies or guidelines that specifyhow decisions about life-sustaining technologieswill be made. These attempt to ensure quality of

care and to reduce risk to the institution and itsstaff. Always, institutional policies are subservientto existing State laws and mandates (or, in the caseof some Federal institutions, Federal law) regard-ing advance directives, family consent, malprac-tice, etc. In almost all cases, institutional pol-icies are procedural, not substantive. That is,they emphasize how a decision should bereached, not what it should be.

Acute care hospitals are the institutions mostlikely to have policies regarding decisions aboutlife-sustaining technologies.5 The hospital poli-cies OTA has reviewed tend to be very cautiousand to presume that treatment will be provided.Most focus on clinical criteria for particular treat-ments, especially resuscitation, and specify pro-cedures for designating and implementing Do-Not-Resuscitate (DNR) orders. Some institutional pol-icies specify alternate levels of care and then havea procedure for assigning patients to each level.Under this kind of policy, patients may be desig-nated, for example, “do not resuscitate, ” “do notincubate, ” or “supportive care only. ”

Institutional policies make explicit the presump-tion for or against treatment in a facility, who willbe involved in a treatment decision (patient, fam-ily, attending physician, other physicians, nurses,ethics committee, other facility staff), and howadvance directives will be regarded. Institutionalguidelines may address ways to protect patientautonomy, for patients who are decisionally ca-pable and those who are not, and ways to resolveconflicts.

There is now some movement toward requir-ing policies as a standard for accreditation of in-stitutions. In June 1987, the Joint Commission onthe Accreditation of Hospitals (JCAH) adopted astandard requiring hospitals and nursing homesto have a policy for decisions about resuscitationby 1988.

Ethical analysis is increasingly recognized as auseful tool in making treatment decisions. Thus,another institutional response has been the estab-lishment of institutional ethics committees or em-

.>A 1986 Surt,ev by the Joint Commission on the Accreditation ofHospitals found “that 57 percent of acute care hospitals, 43 percentof hospices, and 20 percent of nursing homes have formal policiesfor decisions about resuscitaiton,


ployment of a philosopher or theologian to assistin the resolution of troublesome cases. At leasthalf of all acute care hospitals, and higher propor-tions of large hospitals and teaching hospitals, haveestablished ethics committees to assist in decision-making for difficult cases. A few nursing homeshave also established institutional ethics commit-tees. Typically, these committees include physi-cians, nurses, administrators, attorneys, socialworkers, and lay persons who review specificcases brought to their attention. Individual institu-tional policies specify the role of these parties inthe decisionmaking process. In most instances, de-cisions made by ethics committees are regardedas advisory.

Associations of health care professionals haveshown strong interest in developing decisionmak-ing guidelines themselves, partly in an effort toavoid government intervention. Some of these areclinical guidelines, specifying when a particulartreatment is medically indicated. Some, notablythe American Medical Association’s (AMA) 1986statement on “Withholding or Withdrawing LifeProlonging Medical Treatment ,“ address the phy-sicians’ legal and ethical responsibilities in mak-ing these decisions. The AMA statement specifiesthat “life prolonging medical treatment,” which“includes medication and artificially or technologi-cally supplied respiration, nutrition or hydration”may be withheld or withdrawn when doing sois in the patient’s best interest.

Another example of the interest in guidelinesis the list of “principles for decisionmaking” de-veloped by the advisory panel to this OTA assess-ment (see box 1-A), These express the strong con-vergence of opinion—but not unanimity-of apanel of physicians, nurses, lawyers, ethicists, andeconomists regarding many of the fundamentalquestions.

Finding: In practice, many patients are notinvolved in decisions about the use of life-sustaining technologies.

The patient’s involvement in decisions about theuse of life-sustaining technologies varies widelydepending on the urgency of the medical event,the setting, the patient’s cognitive status, and estab-lished decisionmaking procedures. For the tech-nologies OTA studied, the patient’s consent to

treatment is frequently not obtained, and evenwhen consent is obtained, it is frequently not “in-formed.”

Sometimes the patient is left out of the decision-making process because the need for immediateaction or the patient’s mental state makes it im-possible to do otherwise. Victims of cardiac or res-piratory arrest, for example, are typically uncon-scious or in a severely compromised mental state;moreover, the imminent risk of brain damage doesnot permit time for discussion with other personswho may know the patient’s wishes. In such emer-gencies, when the patient’s consent for initiationof treatment is unobtainable, consent is usually“implied.“ Thus, emergency medical techniciansresponding to calls are usually obligated to tryto resuscitate every victim of cardiac arrest, notto pause and ask whether this is wanted.

In the case of resuscitation, the bias to treatis so strong that the normal presumptionabout informed consent is reversed. That is,patients (or their surrogates) are likely to be con-sulted if a DNR order is being considered, but un-likely to be consulted for consent to resuscitate.

Cognitive impairment resulting from dementiaor depression is another major factor in patients’involvement in treatment decisions. Patients who,based on formal or informal assessment, are con-sidered to have severely impaired cognition arecommonly excluded from decisions about theircare. Some of these people, however, if given theopportunity, express consistent wishes regardingtreatment v. nontreatment. Since the prevalenceof dementia increases with advanced age, elderlypatients as a group are less likely than youngeradults to be able to actively participate in deci-sions about their care.

If a patient is determined decisionally incapa-ble, a surrogate decisionmaker can be, and fre-quently is, designated. This may be done infor-mally, as when the physician turns to the patientspouse or an adult child. Or a surrogate may beformally appointed, by the patient or by a court.Some States specify a hierarchy of family mem-bers who have decisionmaking authority if a sur-rogate is needed; others have a “durable powerof attorney for health care” statute.


BOX I-A.—-Principles for Decisionmaking Of Life-Sustaining Technologiesfor Elderly Persons, as Developed by Project Advisory Panel

NOTE: Members of the Advisory Panel to this OTA assessment (sse title page) sought to express their strong

●

●

●

●

●

●

●

●

●

●

●

●

●


In some cases, it is possible to obtain the pa-tient’s informed consent, but the treatment inquestion is considered so “ordinary” that stand-ard practice diverges from the law requiring in-formed consent. Antibiotic therapy, especially inthe hospital setting, is so routine that health profes-sionals often consider consent unnecessary, andthey do not seek it. Also, health professionals’ per-ceptions of some interventions as ordinary or non-invasive mean that, in practice, different treat-ment modalities for a single life-sustainingtechnology can involve different decisionmak-ing practices. Thus, in many institutions, anasogastric tube may be placed for the provisionof enteral nutrition without the patient’s consent—even though formal consent is always requiredfor surgical placement of a gastrostomy tube forenteral nutrition or a catheter for TPN.

Many patients, particularly elderly patients, areaccustomed to a passive role in the doctor-patientrelationship and to accepting the advice of trustedhealth professionals without questioning. Personswho have developed this behavior over a lifetimecannot be expected to start seeking informationor to take an active role in treatment decisionswhen those decisions are most difficult. A 1982national survey reported that 38 percent of re-spondents of all ages, and 60 percent of elderlyrespondents, “want the responsibility of makingthe final choices about your medical treatment”to rest with their doctor. Some elderly personsprefer to entrust important treatment decisionsto their spouse or an adult child,

The urgency of many life-threatening conditionsand the fact that patients may be decisionally in-capable at the time a treatment decision must bemade point to the importance of determining pa-tients’ wishes about life-sustaining treatments be-fore a life-threatening emergency occurs. Imple-mentation of the patient’s wishes is frequentlydependent on advance planning. This may takeseveral forms, including: discussions with familymembers and/or health professionals about treat-ment options, with documentation in the medi-cal record or in a formal advance directive, suchas a “living will”; designation of a surrogate deci-sionmaker; or institutional policies that ask pa-tients to indicate their treatment choices upon ad-mission.

Many health professionals believe that the goalof truly informed consent is often illusory evenwhen there is time for discussion and the patientis fully in command of his or her cognitive facul-ties. In addition to the difficulty most laypersonswould have in understanding the details of theircondition and the treatment options, the gulf be-tween hypothetical and actual situations is wide.It is unlikely, for example, that a layperson (or ahealth professional) who has not personally ex-perienced mechanical ventilation can fully com-prehend the impact of this treatment. By the sametoken, it is impossible to anticipate what it is liketo be unable to breathe. Physicians’ observe thatmany people “change their mind [about being in-tubated for mechanical ventilation] when they arechoking to death,” and this observation contrib-utes to their skepticism about advance directives.

Even when the patient has been informedabout treatment options and his or her wisheshave been specified, problems remain in en-suring that these wishes are implemented, es-pecially if they call for limited treatment. Nei-ther an advance directive nor the instructions ofa surrogate can be followed if authorities do notknow one exists or if the document or person can-not be located. Advance directives that indicaterefusal of life-sustaining treatment are sometimesoverruled because they are considered “toovague.” This can happen if, for example, the pa-tient circumstances or the treatment being con-sidered was not anticipated when the directivewas written, and physicians think treatment willbe beneficial. Inconsistencies in State laws are amajor problem. Some States will not recognize anadvance directive that was made in another State.In many States, advance directives do not becomeoperative until or unless the patient is diagnosed“terminally ill.” Moreover, some State living willstatutes include provisions that, in the view ofsome people, contradict the common law rightto refuse treatment, by specifying, for instance,that nutritional support must always be provided.A patient preference that runs counter to the ad-vice of health professionals is often interpretedas “irrational,” and efforts will be made to changethe patient’s mind or to circumvent the-patient’srequest. In such cases, the patient’s decisionmak -ing capability maybe called into question, and ef-forts made to appoint a surrogate or a guardian.

Ch. 1—Summary and Policy Options . 25

In general, consent to recommended treatmentis easier to implement than is refusal of recom-mended treatment, and any patient wish is easierto carry out if it is consistent with the advice ofcaregivers and the wishes of family members.

Finally, decisions about the use of certainnoninvasive, common technologies are oftenmade without consideration of their life-and-death implications. Care of the life-threatenedelderly involves a continuous series of treatmentdecisions which, individually, may seem so smalland undramatic that their life-and-death implica-tions are not even recognized. Decisions aboutthe treatment of a life-threatening infection, evenin severely debilitated and terminally ill people,frequently focus on choice of the appropriate anti-biotic and omit explicit consideration of whetheror not to treat.

Finding: The physical, psychological, and fi-nancial stresses associated with life-sus-taining treatments are great, not only forpatients, but also for family members andcaregivers.

The physical, psychological, and financial stressesimposed by the life-sustaining technologies OTAhas studied differ with the technology, and theirsignificance depends on the personalities, specificresources, and exigencies of each case. Also, theimmediate and short-term stresses are differentfrom those associated with chronic care. Somepatients cope admirably with the discomforts andfears associated with acute care and, if necessary,with a technology dependent lifestyle, but othersrespond to the anticipated stress by refusing treat-ment. Others start treatment but eventually re-quest that it be withdrawn; they maybe depressedor even suicidal.

Specific effects of the technologies OTA stud-ied include inability to speak or eat (mechanicalventilation), discomfort and limited mobility asso-ciated with tubes and catheters (whether for ven-tilation, nutritional support, drug delivery, or di-alysis), and a gamut of complications ranging fromminor to life-threatening. For patients who areacutely ill, loss of sleep, disorientation, and anxi-ety are concomitants of hospitalization and medi-cation that may accompany all these treatments.

Physical restraints, sometimes used for patientswho are uncooperative or confused, are an addi-tional source of distress. Fear of a new acute epi-sode, loss of independence and control, dietaryregimens, restricted activities, and financial wor-ries may be among the long-term burdens for pa-tients who are restored to medical stability. Comor -bidities, reduced physiological reserve, and limitedsocial support, i.e., characteristics of many elderlypatients, may exacerbate any or all of these.

Family members and friends are also undergreat stress related to anticipatory grieving, finan-cial burdens, and excessive demands on their time.Involvement in treatment decisions is likely to befilled with uncertainty, selfdoubt, or perhaps guilt.If the duration of treatment is prolonged, and espe-cially if the family has caregiving responsibilities,the lifestyle of family members may be radicallychanged. Emotional burdens may be especiallygreat if the patient’s condition or treatment im-pairs or precludes the ability to communicate orif treatment cannot be administered without phys-ical restraints.

It is widely agreed that informed consent shouldinclude disclosure of the likely discomforts andrestrictions attendant with use of these technol-ogies. However, even if the patient is consciousand fully competent when the treatment decisionmust be made, the full impact of these treatmentsis difficult to predict and to convey. If the patientis unconscious or severely demented or confused,those entrusted with the treatment decision canonly speculate about the patient’s experience ofpain or distress with (or without) any of thesetreatments.

Finally, caring for critically ill, terminally ill, orseverely debilitated patients who may be treatedwith life-sustaining technologies is demanding andhighly stressful for health care providers. In addi-tion to the emotional load of dealing with veryill patients and grieving relatives, health profes-sionals are constantly reminded of their own mor-tality and their fallibility. Emotional detachmentfrom patients, avoidance of patients’ families, andoveruse of technologies are not uncommon re-sponses. Impaired job performance and “burn-out”are also reported. Most health care professionalscurrently in practice received little or no train-


ing in the human aspects of death and dying; manyare ill-equipped either to provide emotional sup-port to dying patients or to cope with their ownpersonal reactions.

Whether or not the experience of family mem-bers or caregivers should have any bearing ona treatment decision (or on who should be thesurrogate) is an interesting ethical dilemma—which this assessment does not address. The pointhere is that patients may not be alone in their needfor social and/or financial support.

Finding: Currently, the most controversiallife-sustaining technology is nutritionalsupport. The highly emotional reaction tothis technology obscures specific clinical,legal, and ethical questions that requireresolution.

Of all the life-sustaining technologies OTA stud-ied, nutritional support and hydration is the mosttroublesome for ethicists, clinicians, and the pub-lic. It is over this technology that advocates of“death with dignity” and the “right to life,” as wellas more moderate positions differ most sharply.The debate centers around the question of wheth-er tube and intravenous feeding and hydrationare “food and water” or a medical treatment. Inthe former view, the provision of artificial nutri-tion and hydration constitutes a basic aspect ofhuman caring that should be withheld or with-drawn only when death is imminent or when itis not medically possible to provide them. In thelatter view, these are medical treatments that canbe withheld or withdrawn under the same cir-cumstances as other life-sustaining technologies.These opposing views leave little common groundfor the formulation of policy or for decisions re-garding the care of individual patients.

Very little is known about persons on long-termnutritional support, especially in nursing homes.Anecdotal evidence and some recent researchfindings suggest that many patients on long-termtube feeding are cognitively impaired, but it is notclear why they are tube fed—whether it is becausethey resist hand feeding, because of swallowingdifficulties, or for other reasons. Some peopleclaim that nursing home residents are tube fedbecause hand feeding is too time-consuming.

There are, however, no data to substantiate thisclaim.

Lack of information about cognitively impairedpeople on long-term tube feeding is related to thegeneral lack of information about cognitive im-pairment in elderly people. Ongoing biomedicaland behavioral research on Alzheimer’s diseasepromises to provide some answers. However,much more needs to be learned about the phys-iological, psychological, and emotional aspects ofdementia—particularly the late stages of dement-ing diseases—in order to understand why somepatients with these conditions stop eating and re-fuse hand feeding.

The patient’s formal consent is usually not ob-tained for nasogastric tube feeding—by far themost common mode of nutritional support—because it is not “invasive. ” Although nasogastrictube feeding does not involve surgery, some peo-ple consider it burdensome, particularly when itis used for prolonged periods, sometimes years.An unknown proportion of people who receivetube feeding, including some who are cognitivelyimpaired, are physically restrained to keep themfrom pulling out the tube. This combination offactors would seem to indicate a need for veryrigorous decisionmaking procedures that includemethods for ascertaining the patient’s treatmentpreferences whenever possible, appointment ofa surrogate decisionmaker when necessary, andperiodic review of both the need for and themethod of treatment.

Finding: Ongoing social and technologicalchange will continuously alter the deci-sionmaking context.

The relatively brief history of life-sustainingtechnologies shows how rapidly and dramaticallychanges can occur in attitudes, expectations, andpolicies that determine their use. These changesare driven by a variety of social and technologi-cal factors that are in constant flux and that areoften unanticipated.

At both the individual and societal level, deci-sions about the use of life-sustaining technologiesfor elderly people will be influenced by (and, inturn, will influence) changes in a wide variety of

Ch. 1—Summafy and Policy Options ● 2 7

factors, including technological capabilities, sci-entific knowledge, medical education, economicconditions, public policies and laws, and publicattitudes and expectations. Factors that have at-tracted considerable attention in recent years in-clude the growth and aging of the elderly popu-lation, efforts to contain health care costs, andconcern about the quality of life. Decisions aboutthe use of life-sustaining technologies will also beinfluenced by the increasing level of educationand sophistication among the elderly population,increased competition in health care, and an over-supply of physicians. Comprehensive nationalhealth insurance, a solution to the “malpracticecrisis)” and prevention of dementia are examplesof more distant but equally significant future pos-sibilities.

Improvements in existing technologies and newtreatment modalities could improve the efficacyof treatments, reduce the chance of complications,and increase patients’ comfort and independence.Technological developments might either raise orlower the cost of treatment. Other developments,including improved methods of pain control, andincreased portability and self-care, as well as in-novations like artificial eyes and ears, will improvethe quality of life for chronically ill, disabled, andtechnology dependent people. These marginal im-provements and innovations could alter the bal-ance of benefits and burdens of a particular tech-nology and change attitudes about sustaining lifein persons who are elderly and disabled. In somecases, treatment decisions might become easierand standards of practice might change, leadingto increased use of life-sustaining technologies.

Some existing technologies will be wholly re-placed. Just as kidney transplants eliminate theneed for dialysis in individual patients, other or-gan transplants or artificial organs may eventu-ally obviate the need for other life-sustaining tech-nologies. Very widespread use of such “definitive”technologies could render today’s “halfway tech-nologies” obsolete. Further in the future, effec-tive preventive strategies might have even moreprofound effects on human health and longevity.However, with respect to decisionmaking, the ef-fect of this kind of technological development willbe merely to push problems further into the fu-ture. If we learn to cure heart disease, we willstill face cancer, stroke, and other potentially fa-

tal diseases. We might eliminate one cause of deathafter another, but never all of them.

Neither the development of new technologiesnor improvements in existing technologies arelikely to make the fundamental issues of access,quality, and cost of care, or the decisionmakingdilemmas these create, go away. Instead, changewill be in the foci and details of current ethical,legal, and clinical debates. OTA's analysis showsthat the current intense interest in nutritional sup-port follows more than a decade of controversyand court cases focused on mechanical ventila-tion. A possible next center of controversy is anti-biotic therapy, which is only now gaining recog-nition as a life-sustaining treatment that raisesserious issues. Similarly, changes in technologyand in health services delivery will shift concernfrom the hospital to community settings and trans-fer more decisionmaking responsibility from phy-sicians to other health care personnel and to laycaregivers.

In addition, social and technological changewill bring some new questions and intensifysome of the current problems. For example, asboth the law and medical practice change, newkinds of legal challenges may arise. A recent in-stance in which physicians were charged becausethey instituted unwanted treatment is said to haveopened the door to a new set of legal actions. Theold problems of cost and access to care may beexacerbated if, as many people predict, the costof providing the full range of theoretically bene-ficial treatments continues to increase. Particu-larly high cost will be associated with the careof individuals enabled to survive much longer thancurrently possible. Continuing high cost (and in-creasing cost) could lead to a more prominent roleof third-party payers and government in healthcare decisionmaking.

Other pertinent developments will not changethe basic decisionmaking problems but do prom-ise to help us sort through difficult choices. Theseinclude the procedures, policies, and technologi-cal developments that aim to supply more com-plete information on which to base decisionsand/or a more systematic way to assimilate it andreach an informed conclusion. These range frompatient education to health professions educationand from computerized decision support systemsto ethical analysis.


CONGRESSIONAL ISSUES AND OPTIONS

The following issues and options are derivedfrom information summarized in this chapter andpresented in detail in the full report. They addressproblems that are common among several or allof the life-sustaining technologies OTA studied andthat are realistic foci for congressional oversightand legislative activity. Problem areas that areunique to one or another technology and thosethat do not suggest Federal involvement are pre-sented in the findings and implications at the endof each of the respective chapters. Ultimately, reso-lution of the diverse problems associated with theuse of life-sustaining technologies for elderly peo-ple and maximization of the potential good thesetechnologies can bring will require the creativityand cooperation of philosophically and profession-ally diverse factions.

The first pair of issues and accompanying op-tions addresses research needs that relate to allof the subsequent issues. These include statisti-cal data for improved health care planning anddelivery and basic research to expand the scien-tific knowledge base. The next pair of issues andoptions addresses the concern of the requestingcongressional committees about access to life-sustaining treatments and how access is affectedby age, availability of reimbursement, and setting.The third issue area addresses what Congressmight do to reduce problems in individual deci-sionmaking about the use of life-sustaining treat-ments. The two final issues and options addressquestions that arose in the course of this assess-ment about the safety and efficacy of life-sustaining technologies and the quality of care pro-vided for elderly people once a decision has beenmade to provide, withhold, or withdraw life-sus-taining technologies.

Associated with each policy issue are severaloptions for congressional action, including in eachcase, no action. The order in which the optionsare presented should not imply their priority. Theoptions are, for the most part, not mutually ex-clusive. In fact, a careful combination of optionsmight produce the most desirable effects. Further,while these issues address life-sustaining treat-ment for elderly persons, many of them are appli-cable to patients of all ages.

The issues and options presented here are real-istic foci for congressional oversight and legisla-tive activity. Numerous other issues fall moreappropriately within the activities of nongovern-mental bodies. Ultimately, resolution of the vari-ous problems associated with the use of life-sustaining technologies for elderly people andmaximization of the potential good these technol-ogies can bring will require the creativity and co-operation of philosophically and professionally di-verse factions,

Research

Issue 1: What could Congress do to strengthenand expand the statistical database on theutilization and costs of life-sustainingtechnologies?

1.1 Take no action.

1.2 Provide funds and instruct HCFA to conductstudies on the utilization of and expendituresfor life-sustaining technologies in hospitals,nursing homes, and home care.

1.3 Instruct HCFA, the National Center for HealthStatistics, and the Veterans Administration(VA) to develop and employ standardizedmethods for calculating and reporting utili-zation and costs of Life-sustaining technologies.

Several factors argue against a Federal role inthe collection of additional health statistics and/orestablishment of a databank on the use of life-sustaining technologies. First, inaction at the Fed-eral level (i.e., Option 1.1) would avoid the expend-itures related to new data collection efforts. Ad-ditional medical recordkeeping and changes inreporting methods might be opposed by the in-stitutions and individuals who are asked to pro-vide the data. In addition, some observers fearthat a recordkeeping system that specifies costand reimbursement for particular technologiescould lead to inappropriate economic pressuresto alter treatment patterns.

On the other hand, a major finding of this assess-ment is that neither the magnitude of currentproblems nor predictions of future demand can


be adequately estimated with existing datasources. (The scarcity and unreliability of avail-able data are substantial for young as well aselderly patients.) Data on the utilization of andexpenditures for life-sustaining technologies comemainly from small case studies whose results can-not be aggregated or generalized. The notable ex-ception is dialysis, for which good utilization andexpenditure data are now maintained, but forwhich the absence of data prior to Medicare cov-erage contributed to gross underestimates of theeventual demand for this treatment. Improveddata would help inform public policy and, to theextent that the necessary recordkeeping makesclinical decisions more explicit, could also improvedecisionmaking in individual cases.

Sample surveys of Medicare patients and elderlyMedicaid patients who receive life-sustaining tech-nologies (Option 1.2) would be a relatively easyand relatively inexpensive way to expand the sta-tistical database on utilization and Federal expend-itures. Careful consideration must be given todetermining which life-sustaining technologieswarrant this attention. At a minimum, for eachselected technology, the studies should providedata on: the patient’s age, diagnoses, treatmentsettings, clinical outcome, discharge status, andpayments by Medicare and/or Medicaid. Informa-tion on expenditures by private insurers, patients,and any unpaid charges would also be desirable,to complete the cost picture. Parallel data onelderly patients in hospitals, nursing homes, andin their own homes would provide a rather com-prehensive data set useful for a variety of analy-ses. Ideally, the data would permit cross-sectionalor longitudinal analysis, comparisons among sub-groups within the elderly population, and com-parisons of utilization and costs in different set-tings. Improved information about the currentsituation would be essential input to any Federalpolicy decisions about limiting or expanding healthcare services, payment, or training. If maintainedcontinuously or updated periodically, these datacould be the foundation for predictions of futuredemand for and cost of providing particular tech-nologies. The arguments against Option 1.2 arethe same as those in support of Option 1.1.

Option 1.3 addresses the noncomparability ofutilization and cost data that are currently avail-

able. Problems in utilization data result fromdifferent definitions of such terms as “chronic”or “prolonged” use, dissimilar age categories, andvariations in codes for the pertinent medical andsurgical procedures. “Cost” data sometimes rep-resent charges, sometimes expenditures, and ex-actly what is included is seldom specified. Themain argument against this approach is that thedefinitions and methods developed may not ade-quately fit the diverse needs of potential users.To reduce this possibility, standardized definitionsof utilization and costs should be developed withinput from all interested parties—especially hos-pitals, insurers, patients, health economists, andpolicymakers.

Issue 2: What could Congress do to strength-en and expand scientific and clinicalknowledge related to the use of life-sus-taining technologies, especially forelderly people?

2.1 Take no action.

2.2 Authorize and appropriate funds administeredthrough the National Institute on Aging (NIA)for studies of life-threatening conditions in theelderly and the physiological and psychologi-cal responses of elder@ patients to alternativetreatments.

2.3 Provide research funds administered throughNIA to coordinate work on the developmentof measures that better reflect the health sta-tus and reserve capacity of elderly people thandoes chronological age.

2.4 Authorize and appropriate funds through theDepartment of Health and Human Services(DHHS) or NIH to develop and test patient clas-sification systems and other aids to clinicaldecisionmaking.

2.5 Authorize and appropriate funds to supportan NIH research planning conference focusedon the care of elderlypersons with life-threat-ening conditions.

Option 2.1 assumes that existing Federal sup-port for technology assessment and basic researchrelated to life-sustaining technologies is adequateand appropriately directed, that adequate non-Federal support is available, or that additional re-


search would not reduce problems related to theuse of life-sustaining technologies. Proponents ofadditional research argue that little research hasbeen focused on these topics and that informationis needed to reduce inappropriate and ineffectiveutilization of life-sustaining technologies. Researchwould require additional Federal expenditures orshifting of funds from other areas. However, po-tential benefits, in terms of improved patient selec-tion and improved quality of care, as well as po-tential reductions in the cost of care that isprovided, might outweigh the costs associatedwith research.

Very little research has focused on the relation-ship between advanced age and the clinical out-comes of life-sustaining technologies. The resultinginformation gaps contribute to clinical uncertaintyand prognostic errors, as well as suboptimal careand poor outcomes. Added Federal support forresearch on these topics (Option 2.2), especiallyprospective and longitudinal studies, could leadto improved understanding of the factors associ-ated with different clinical outcomes, includinglongevity. This knowledge could lead to the de-velopment of age-indicated modifications in treat -ment that could, in turn, lead to increased sur-vival of elderly persons with life-threateningconditions, with improved functional capacity, re-duced complications, and less recidivism.

It has been well established that physiologicalchanges occur at different rates and to differentextents in different people, with the effect thatindividuals are increasingly dissimilar as they age.While many physicians now recognize that chron-ological age masks this heterogeneity, age remainsthe simplest single indicator of physiological sta-tus. Basic research on age-related physiologicalchange and response to stress, directed towardthe development of alternative measures of healthstatus and reserve capacity (Option 2.3) might leadto improved accuracy in patient assessment andprognosis.

Option 2.4 proposes Federal support for the con-tinuing development and testing of patient clas-sification systems and other aids to clinical deci-sionmaking. Some of these systems, currentlyexperimental, show considerable promise for iden-tifying patients who are likely to benefit fromtreatment and patients who are likely to die de-

spite treatment. Refinement of these systemsand/or development of new approaches could re-duce ineffective use of life-sustaining technologies.

Another approach to providing information thatcould potentially improve decisionmaking wouldbe sponsorship of an NIH research planning con-ference, as suggested in Option 2.5. The confer-ence would bring together experts in geriatricsand in critical care, medical decisionmaking, healthservices, and health law, with the goal of specify-ing and prioritizing areas of research on the careof the life-threatened elderly. A consensus aboutkey issues would direct Federal funding to themost fruitful areas, and the visibility of such aconference could also help to stimulate privatefunding for identified priority areas.

Access to Care

Issue 3: What could Congress do to protectelderly persons from possible age-baseddiscrimination in access to life-sustainingmedical treatments?

3.1 Take no action.

3.2 Provide funds and instruct HCFA to conductstudies of hospital and nursing home practicesregarding the offering of life-sustaining tech-nologies to elderly patients.

3.3 Instruct HCFA to expand Medicare reimburse-ment for life-sustaining medical care.

(Also see Options 2.3, 5.3, and 6.4.)

Whether or not Federal action to prevent pos-sible discrimination is warranted at this time de-pends on one’s evaluation of the current situa-tion. One goal of recent public policy is to protectthe equal rights of all citizens, without regard torace, sex, or age. Ensuring equal access to neededhealth care is one of the responsibilities of policy-makers. However, because health care resourcesare not unlimited and because aging is universal,“equal” access can include different interpreta-tions of the kinds of care that must be offered,under what circumstances, and for how long.Some people argue that Medicare, because it pro-vides health care mainly to elderly persons, is it-self inequitable. On the other hand, anecdotesabout limited care for hospitalized Medicare pa-


tients have stirred public concern and congres-sional attention. The extent to which elderly per-sons might be denied access to life-sustainingtechnologies because of their age is not known;however, limited Medicare hospital reimbursement,health professionals’ ignorance of the good prog-nosis for many elderly patients, and residual age-ism create considerable potential for age discrimi-nation in access to these treatments.

Studies proposed in Option 3.2 would provideinformation about the extent to which Medicarepatients and elderly Medicaid patients are offeredvarious life-sustaining technologies in hospitals andnursing homes. This information would enablepeer review organizations (PROS)’ or other over-seers to identify cases in which life-sustaining tech-nologies were not accessible. It would not, how-ever, be possible to draw from this conclusionsabout age-based discrimination unless compara-ble information were available for younger pa-tients as well. Requiring sampled providers to keeprecords of all treatments offered to patients wouldbenefit those patients by encouraging physiciansto entertain and to discuss with patients all rea-sonable treatment options.

Current cost-containment pressures and limitedMedicare reimbursement provide hospitals andphysicians financial disincentives to admit and toaggressively treat Medicare patients whose costsare likely to exceed what Medicare will pay un-der PPS. Option 3.3 would remove or reduce thosefinancial disincentives. Adjustments could be madein the level of reimbursement for DRG categoriesthat frequently involve life-sustaining technologies,by creating new technology-specific reimburse-ment categories, by adding a severity of illnessmeasure to all DRGs, by increasing the age ad-justment factor that already applies to some DRGsor by raising outlier rates. Such actions would beexpensive and difficult to justify when there isno proof that age-based discrimination is a seri-ous problem. However, some people would viewthe protection of access to health care as impor-tant enough to justify a preventive approach.

Whe function of PROS is to review’ the appropriateness of hospi-

tal admission for Niedicare patients, approlw pa~mlent, and monitorqualit~f of care on appeal PROS also rm’iew individual cases in whichadmission or paf’ment is thought to he inappropriate? cienied, andcases in which discharge is thought to he premature.

Option 2.3 would reduce opportunities for treat-ment decisions based on unjustifiable generaliza-tions about old age. Options 5.3 and 6.4 wouldeducate patients and providers, respectively, tobe better advocates for themselves and for theirelderly patients.

Issue 4: What could Congress do to increasethe availability of life-sustaining technol-ogies in nonhospital settings?

4.1 Take no action.

4.2 Instruct HCFA to provide Medicare coveragefor life-sustaining antibiotic therapuy and short-term nutritional support outside the hospitalsetting.

4.3 Instruct HCFA to increase Medicare homehealth care coverage for personnel who pro-vide needed services for Medicare patients de-pendent on life-sustaining technologies in theirown homes.

4.4 Instruct HCFA to encourage the States to raiseMedicaid reimbursement available to nursinghomes that hire highly skilled personnel in or-der to provide life-sustaining technologies.

4.5 Authorize and appropriate funds to DHHS forthe support of research and demonstrationprojects regarding the use of life-sustainingtechnologies in nonhospital settings.

Current medical practice and reimbursementpolicy favor the use of hospitals, and often theirICUs, for application of most of the life-sustainingtechnologies OTA studied. For patients who aremedically stable and who no longer require theresources of a hospital, care in another settingis generally less costly and facilitates a less re-stricted lifestyle. Therefore, most people think itwould be beneficial for patients, as well as moreefficient, if utilization of life-sustaining technol-ogies were shifted as much as possible to non-hospital settings (Option 4.1). Expanded availabil-ity of life-sustaining technologies outside ofhospitals could, however, lead to inappropriateuse, with consequent increased cost. Further, thequality of care could be jeopardized in these rela-tively unsupervised settings.

OTA found that some patients who could safelybe treated in alternate settings are confined to


hospitals because of inability to pay for serviceselsewhere. For some technologies, e.g., ventila-tion, the problem is that the unreimbursed por-tion of care, while it may be a small percentageof the total cost, is often still very high. For twoof the technologies discussed in this assessment,life-sustaining antibiotic therapy and short-termnutritional support, Medicare reimbursement out-side the hospital is completely unavailable. Op-tion 4.2 suggests expansion of Medicare benefitsto cover these technologies. Option 4.3 goes a stepfarther, proposing Medicare reimbursement forthe personnel needed to provide any life-sustainingtreatments outside of hospitals. Among these per-sonnel are health professionals (e.g., respiratorytherapists, professional nurses), and nonprofes-sionals (aides).

Option 4.4 addresses the current difficulty innursing homes of hiring staff who have the nec-essary skills and credentials to provide complexcare. Most nursing homes do not admit patientswho are receiving mechanical ventilation, intra-venous antibiotics, or TPN, and most are notequipped to provide these treatments to residentswho need them. Inadequate and unpredictablereimbursement make it difficult for nursing homesto develop staff and services and, thus, limit out-of-hospital options for persons who are medicallyready to be discharged from hospitals. Some nurs-ing homes that do provide care for technology-dependent persons have negotiated special reim-bursement arrangements with Medicare or Med-icaid on a patient -by-patient basis. For patients whoare eligible for Medicare nursing home benefits,coverage could be extended beyond the current100-day” limit. For technology-dependent Medic-aid patients in nursing homes, HCFA could offerStates incentives to increase reimbursement.

Information regarding the relative benefits andproblems in providing life-sustaining technologiesin alternative settings is piecemeal and largelyanecdotal. Option 4.5 would support research anddemonstration projects to clarify the types of pa-tients for whom alternatives to the hospital (and,within hospitals, alternatives to the ICU), are safe,economical, and contribute to the patient’s qual-ity of life. Such projects could also provide infor-mation regarding the supportive services patients

need in different settings, alternative methods forproviding them, and the relative costs and bene-fits. One possible site for such projects is the teach-ing nursing home. An important component ofsuch programs would be their educational bene-fits, i.e., through the opportunity to train healthprofessionals within the institutions where proj-ects go on and the dissemination of results tohealth professionals in other institutions.

A main argument against Options 4.2 through4.5 is that liberalization of reimbursement forhome care and nursing home care of technology-dependent patients might create substantial newdemand for services and attendant new costs tothe Federal Government. In addition, some peo-ple fear that quality of care cannot be assuredoutside the hospital. Other difficulties relate todecisions about whether coverage should be forall life-sustaining technologies or only designatedones (i.e., Option 4.2), which personnel should bereimbursed for which services (Option 4.3), andwhether particular treatment settings, rather thanall nonhospital settings, are to be equally en-couraged.

Decisionmaking

Issue 5: What could Congress do to protectthe rights of elderly patients in decisionsabout the use of life-sustaining inter-ventions?

5.1 Take no action.

5.2 Authorize and appropriate funds for researchand demonstration projects that will provideinformation about current decisionmakingpractices, problems, and possible solutions.

5.3 Support education of the public regardingtheir rights as patients and mechanisms forimplementing these rights.

5.4 Instruct HCFA, the VA, and the Departmentof Defense to require Federal health care fa-cilities and health care facilities that are certi-fied to treat Medicare and Medicaid patientsto: 1) record in a patient record any advancedirective the patient presents, and 2) honorthat directive.


5,5 Instruct HCFA, the VA, and the Departmentof Defense to require health care institutionsthat receive Medicare and Medicaid reimburse-ments as well as all Federal health care insti-tutions to: I) develop written policies describ-ing the procedures they will follow in makinga decision about life-sustaining technologies,and 2) communicate these policies to all pa-tients.

5.6 Develop Federal legislation regarding advancedirectives and procedures for the identifica-tion of surrogate decisionmakers.

(Also see Option 6.4.)

The proper role of the Federal Government inhealth care decisionmaking is very controversial,with opinions ranging from no role to a direct,intimate role (as in the original “Baby Doe” regu-lations). Governmental involvement in the sub-stance of treatment decisions for the life-threat-ened elderly would meet strong opposition fromhealth professionals and from patients of all per-suasions. More widely accepted roles for Govern-ment would focus on either the provision of in-formation (Option 5.2 and 5.3), the establishmentand protection of decisionmaking procedures (Op-tion 5.4 through 5.6), or both. However, some peo-ple oppose all forms of governmental involvement,arguing that decisionmaking procedures as wellas substantive decisions are the responsibility ofqualified health care professionals (Option 5.1).

OTA’s findings suggest several kinds of infor-mation about decisionmaking that could help re-duce current problems. Option 5.2 calls for thecollection and analysis of descriptive informationabout how decisions are made with regard to theuse of life-sustaining technologies for elderly peo-ple. This kind of research would provide evidenceon the extent to which elderly persons partici-pate in decisions about the use of life-sustainingtreatments, identify the reasons patients’ wishesare not always implemented, and would identifyany subgroups of the elderly population (e.g., ex-tremely old persons, demented persons, nursinghome residents) whose rights may need greaterprotection. Such research would also contributeto determining the practical strengths and weak-nesses of different kinds of advance directives anddifferent decisionmaking processes. However,

some people might perceive this kind of researchas an invasion of privacy.

Option 5.3 addresses the current scarcity of pub-lic education regarding patients’ rights, the im-portance of making known one’s wishes regard-ing life-sustaining treatments, and availablemechanisms for formalizing these wishes. Thisoption assumes that such education would resultin more people preparing some type of formaladvance directive (e.g., living will or durable powerof attorney) or, at least, discussing with their familyor physician their personal views regarding life-sustaining treatment. Increasing the number ofpersons whose wishes are known should resultin an increase in the number of patients whosewishes are honored. Some people have suggestedthat having a clear directive from the patient isthe single best way to reduce unnecessary healthcare expenditures. Opposition to such educationalefforts might come from those who fear that theeducators would advocate particular positions.

Options 5.4 and 5.5 reflect OTA’s finding that,in many institutions, the approach to decisionmak -ing about the use of life-sustaining technologiesis ad hoc. In most hospitals and nursing homes,there is no mechanism for determining or regis-tering a patient’s treatment preferences beforethe need for a life-sustaining technology arises,when it may not be possible to consult the patient.In some cases, health care providers are not awarethat a patient who is decisionally incapable hasan advance directive. Even if they are aware ofthe advance directive, they do not always follow it.

Formal institutional policies for decisionmakingcould help protect a patient’s right to participatein treatment decisions and clarify the roles andresponsibilities of other participants in the deci-sion (e.g., families, ethics committees). Institutionalpolicies would not necessarily offer any legal pro-tection to patients, institutions, or individual care-givers, but they could potentially acquire consid-erable authority as they evolve into standards ofpractice.

The Federal Government could require healthcare institutions that receive Medicare and Med-icaid reimbursements and Federal health care in-stitutions to develop formal institutional policiesfor decisionmaking (Option 5.5). Although many


parties favor the establishment of policies for deci-sionmaking at the institutional level, it is not clearwhether such policies should be required by theFederal Government. The number of hospitals,nursing homes, and other health care facilities thathave formal institutional policies for decision-making appears to be growing. The recently an-nounced JCAH requirement that hospitals andnursing homes must have a policy for decisionsabout resuscitation in order to be accredited byJCAH is expected to further this trend. Thus, somepeople believe that there is no need for a Federalrequirement for institutional policies for decision-making. Other people believe that a Federal re-quirement is needed to ensure that most, if notall, health care facilities have such policies in place.

Even if the Federal Government were to requirehealth care institutions to have policies for deci-sionmaking, it is unclear whether the requirementshould address the content of those policies orwhether the content of the required policies shouldbe left to the discretion of each institution. If agree-ment is reached that content should be addressedby the Federal Government, it is unclear whetherthe requirement should specify questions thepolicies must answer (e.g., how a patient’s deci-sionmaking capacity will be assessed or how a sur-rogate will be selected) or decisionmaking proce-dures that should be followed. Some people believethat the content of decisionmaking policies shouldbe determined by individual institutions becauseof differences in their purposes, practice environ-ments, and patient populations. Others believe thatat least minimum standards should be includedto protect patients’ rights and ensure some con-sistency across jurisdictions and institutions. Selec-tion of such standards would be difficult becauseof disagreement about appropriate decisionmak-ing practices. 7

Option 5.6 suggests Federal legislation to author-ize advance directives (living wills and durablepowers of attorney for health care) and to specifyprocedures for identifying surrogate decision-makers for patients who are not decisionally ca-

7As a follow-on to this assessment, OTA has commissioned a re-port on institutional policies for decisionmaking that will considerthese questions in more detail. That report will be available in early1988.

pable and who have no advance directive. Fed-eral legislation to authorize advance directiveswould make these methods of documenting anindividual’s treatment preferences available to allAmericans, including those who live in States thathave not enacted statutes allowing advance direc-tives. Federal legislation could ensure that a liv-ing will or durable power of attorney for healthcare executed in one State would be accepted inother States. Proponents of advance directives,who view them as an important safeguard of pa-tient autonomy, would probably welcome suchlegislation. Yet disagreement about specific pro-visions of advance directives, e.g., whether theyshould allow withholding or withdrawal of life-sustaining nutrition, hydration, and medicationsand whether they should allow withholding orwithdrawal of treatment from persons who arenot terminally ill, would complicate the develop-ment and enactment of such legislation.

People who believe that life should be sustainedwhenever it is technically possible to do so wouldprobably oppose Federal legislation authorizingadvance directives because the directives usuallyallow withholding or withdrawal of treatment.Some people would also object to Federal legisla-tion in an area that has traditionally been gov-erned by the States and might prefer Federal ac-tions that encourage States to enact statutesauthorizing advance directives. Others might pre-fer that the Federal role be limited to support ofpublic education about advance directives (Option5.3).

Federal legislation specifying procedures foridentifying a surrogate decisionmaker for patientswho are decisionally incapable and have no ad-vance directive and defining the role and respon-sibilities of the surrogate could reduce confusionabout the legality of existing decisionmaking prac-tices for these patients. Such legislation might bemodeled after the family consent laws now in ef-fect in 15 States. Alternatively, the Federal Gov-ernment could require health care instititutionsto have formal policies defining procedures forsurrogate decisionmaking as a part of the institu-tion’s policy for decisionmaking, as in Option 5.5.Objections to these approaches are similar to ob-jections to Option 5.5.

Ch. 1—Summary and Policy Options . 35

Quality of Care

Issue 6: What could Congress do to improvethe quality of care associated with the useof life sustaining technologies for elderlypeople?

6.1

6.2

6.3

6.4

6.5

Take no action.

Instruct the Federal agencies engaged in tech-nology assessment and clinical trials, i.e., theNational Center for Health Services Researchand Health Care Technology Assessment's Of-fice of Health Technology Assessment (OHTA),the Food and Drug Administration (FDA), NIH,and OTA to make studies of life-sustainingtechnologies a priority.

Provide Federal funds or tax incentives for re-search and development of improved life-sus-taining technologies (equipment and products),including refinements that simplify operationand maintenance.

Authorize and appropriate funds to DHHS andthe VA to support education and training aswell as special practice models for healthprofessionals who care for the life-threatenedelderly.

Authorize and appropriate funds for DHHSto develop model programs offering compre-hensive support services to technology-dependent elderly persons who need them.

This assessment has raised both general ques-tions about efficacy and safety of some life-sus-taining technologies and questions that are spe-cific to the use of these technologies for elderlypatients. Problems arise from deficits in the knowl-edge base, the technologies, and the personnel.Numerous activities that have potential benefitsin terms of ensuring the efficacy and safety oflife-sustaining technologies for elderly patients arealready underway. These include the regular activ-ities of FDA, technology assessments by OTA andOHTA, clinical studies by NIH, and support forhealth professions training, including programsto expand education and training in geriatrics andgerontology. Some would conclude that theseactivities are adequate. However, with respect tospecial needs of the life-threatened elderly, noneof these programs goes very far.

Questions have been raised about the reliabil-ity of some equipment and products and aboutundue complexity (and, therefore, cost) of others.These questions suggest the need for assessmentof life-sustaining technologies in addition to thoseOTA has studied and for correction of identifiedproblems. Option 6.2 would provide informationabout any problems related to particular medicaltechnologies used to sustain life. This would in-form policy decisions about whether or not a par-ticular technology ought to be widely available,or reimbursed, and clinical decisions about its usefor individual patients. A practical drawback toOption 6.2 is that there area large number of life-sustaining technologies, and new ones being de-veloped, and only a fraction of them can be as-sessed. Also, unless tied to approval by FDA orto reimbursement decisions, the results of theseassessments might have little effect. Option 6.3would encourage R&D in Federal laboratories,provide grants to universities and major medicalcenters, and support special incentives to the pri-vate sector to improve existing technologies andto develop reliable and relatively simple technol-ogies suitable for use in the home or nursing home.

Option 6.4 would support curriculum develop-ment, instruction, and practice models focusedon: 1) geriatrics and gerontology, and 2) humanisticcare of the dying, in order to simultaneously in-crease the supply and upgrade the capabilities ofpertinent health professionals. Programs wouldtarget physicians, nurses, and allied health profes-sionals still in training as well as health profes-sionals already in practice.

The Federal Government currently supportseducation and training in geriatrics and geron-tology through programs of the NIA, National In-stitute of Mental Health, Administration on Aging,Health Resources and Services Administration(HRSA), and the VA. Despite dramatic increasesin the numbers of physicians and other healthprofessionals committed to geriatrics, serious man-power shortages and barriers to recruitment sug-gest that more needs to be done. Moreover, exist-ing education and training does little to specificallyprepare physicians or nurses to care for elderlypersons who become candidates for life-sustainingtechnologies. Pertinent curricular innovations,e.g., clinical ethics, death and dying, health law,

—


decision analysis, assessment of patients’ decision-making capacity, and interdisciplinary teamwork,are relatively new and, under current cost-con-tainment strategies, their continuance is threat-ened. There is no cross-training between spe-cialists in geriatrics and specialists in critical care.Many people assume that providing more educa-tion and training in these areas would improvethe quality of care for the life-threatened elderly.There has been, however, very little research toevaluate the benefits of this kind of education,and, therefore, limited evidence that such pro-grams have a significant effect on treatmentoutcomes.

Option 6.5 recognizes that many patients whoare chronically dependent on a life-sustaining tech-nology have unmet needs for financial and otherkinds of assistance, such as attendants, transpor-tation, special equipment, architectural modifica-tions, group purchasing of medical supplies, etc.New Federal programs that target specific groupsof patients for special benefits could be criticizedas perpetuating a disjointed approach to healthcare, and new expenditures would be required.In France and England, comprehensive programsfor ventilator-dependent patients have proved tobe cost-effective and of great benefit to patients,enabling some technology-dependent persons tolive in their own homes, with relative independ-ence and maximum quality of life.

Issue 7: What could Congress do to improvethe quality of care for people from whomlife-sustaining treatments are withheld orwithdrawn?

7.1 Take no action.

7.2 Instruct HCFA to extend eligibility criteria forhospice care and palliative treatments, to makethem more widely available.

7.3 Appropriate funds and direct NIH or HRSAto support research and training to study thedying process and to develop methods of pal-liative care for patients from whom Life-sustaining technologies have been withheldor withdrawn.

Federal involvement in research, health profes-sions education, and reimbursement for healthcare have greatly benefited patients who want ag-

gressive medical treatment. Good care has beenwidely available and the financial barriers largelyremoved. However, for patients from whom life-sustaining technologies are withheld or with-drawn, treatment options are undeveloped, andresources are scarce. The single focus of Federalefforts on behalf of these patients is hospice careand the provision of limited hospice benefits un-der Medicare.

The hospice model of care was developed tomeet the physiological and psychological needsof patients who have been diagnosed as termi-nally ill and who choose to forgo aggressive treat-ment. Most hospice patients are victims of incura-ble cancers who consciously requested this kindof care. Hospice care has not been available inthis country to persons who cannot make deci-sions about their care and those who have notbeen designated terminally ill. The potential ben-efits for some such patients, for example, severelydemented patients who cannot be dialyzed, deci-sionally capable ESRD patients who choose todiscontinue dialysis, and patients with chronic ob-structive pulmonary disease who refuse mechan-ical ventilation, have not been studied. Option 7.2would make hospice care more widely available.

Anecdotal evidence suggests that, following adecision to withhold or withdraw life-sustainingtechnologies, patients are sometimes essentiallyabandoned. Health professionals may simply havenothing to offer these patients. Therapeutic op-tions are exhausted or rejected; methods and re-sources for pain control and bereavement coun-seling are undiscovered, illegal, or unfunded,Option 7.3 is to support behavioral, pharmaco-logical, and health services research gearedtoward discovering and then meeting the needsof this group of patients. For these people, Op-tion 7.3 would provide some answers about thepotential benefits of existing forms of hospice care,develop options to the use of life-sustaining tech-nologies, and then train health care professionalsin these methods. The cost of such programs mightbe returned many times by reduced expendituresfor life-sustaining technologies. Of all the manyresearch needs identified in this assessment, thosereferred to in Option 7.3 are among the most im-portant.

Date post:	27-Dec-2021
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Chapter 1 Summary and Policy Options

Documents