What Should							III
Be Done?

In this study, the President's Commission has, in effect, explored the 
widely held hypothesis that a program of compensating for research 
injuries is needed because subjects who deserve compensation are not 
receiving it.  This appears to be at least partially true; providing 
compensation to injured subjects (even when research projects are 
conducted with due care) is ethically desirable--indeed, some would 
claim that a moral obligation to compensate exists.  Yet, thought it 
emerged in Chapter Five that any subjects who are injured may find it 
difficult to obtain recompense under existing tort or administrative 
remedies, the Commission found (in Chapter Four) that the frequency 
with which this occurs is not firmly established.  Thus, more data are 
needed to provide a sound basis for policy on this subject.  
Furthermore,  the strength of the ethical requirement to provide 
compensation for research-related injuries depends in the view of some 
people, as discussed in Chapter Three, upon the practicability of 
doing so.  Thus, it is imperative to obtain reliable information on 
the extent of the need for compensation as well as the horizontal 
inequities, transaction costs, vulnerability to abuse, and 
difficulties of administration that provide compensation might entail.  
In other words, it must be determined, first, whether the incidence of 
injures to research subjects is great enough to warrant institution of 
a formal compensation program, and, second, whether the cost of 
compensating those injured in research can be prevented from 
escalating uncontrollably and consuming an inordinate portion of the 
overall research budget.  In this Part, the Commission explores, first 
the manner in which data could be developed (Chapter Six), and, 
second, the basic features of any compensation program (Chapter 
Seven).



Conduct An Experiment						6


The paucity of reliable data was describe in Chapter Four.  No 
definitive national data on the incidence or severity of injuries in 
Federally conducted or supported research exist.  The costs of 
acquiring comprehensive data through a methodologically valid 
retrospective survey would be very high--in the Commission's view, too 
high to be justified.  Further, the results of any such survey would 
be criticized because retrospective data cannot disclose the extent of 
those costs that are "induced" by the existence of a compensation 
program.  Consequently, such data might understate both the number of 
potential claims and the associated costs that would arise were the 
existence of a compensation program made know to subjects.

Some witnesses urged the Commission to conclude that no further 
exploration of compensation programs should occur in the absence of 
proof of a crying need.  But an absence of data on injuries should not 
be equated with data on an absence of injuries.  Other groups and 
individuals possessing great familiarity with biomedical research 
suggested that the inadequacies of the data be remedied.  The 
Association of American Medical colleges, for example. argued for 
"instituting such a compensation system on a pilot basis, limited by 
time, by number of institutions or geography, to permit the gathering 
of relevant data on the basis of real experience." 1[Letter from John F. 
Sherman, Vice President, Association of American Medical Colleges, to 
Chairman Morris B. Abram (May 7, 1980).]

Preceding page blank

Although wishing to avoid the term "pilot" program which might be 
misunderstood as a commitment to moving to a full scale program, the 
Commission agrees with the thrust of the argument.  It has concluded 
that the problem of research injures should not be shelved once again, 
to be passed on from commission to commission for further review, with 
each inquiry stymied by the lack of data.  Therefore, 


the Commission urges that steps be taken to acquire data   on the 
nature and incidence of research-related injuries through a test of 
one or more approaches to compensation for research injuries.
2 [The Commission took on a request to provide advice about the 
development of a means for compensation research subjects for 
injuries that occur in the future.  It lacks the authority for the 
investigation that would be needed to determine whether retrospective 
compensation should be made to subjects who may have been injured in 
past research projects.  A number of such subjects are pursuing claims 
in the courts already, as described in Chapter 5.]

As a baseline for comparison, data are needed on the incidence of 
research injuries nationally.  To this end, the President's 
Commission, in its first Biennial Report, recommended that the 
regulations governing all research conducted and supported by Federal 
agencies be amended to require annual reporting of the number of 
research subjects as well as data about their injuries.
3[President's Commission for the Study of Ethical Problems in Medicine 
and Biomedical and Behavioral Research, PROTECTING HUMAN SUBJECTS: 
FIRST BIOCENNIAL  REPORT ON THE ADEQUACY AND UNIFORMITY OF FEDERAL 
RULES AND POLICIES, AND THEIR IMPLEMENTATION, FOR THE PROTECTION OF 
HUMAN SUBJECTS IN BIOMEDICAL AND BEHAVIORAL RESEARCH, U.S Government 
Printing Office, Washington (1981) at 73.]  
Although this is a  procedural suggestion and was made in the 
context of a review of the Federal rules on research, it is at its 
heart a matter of ethical concern.  (The Commission was disappointed 
to find that most research sponsors have not regularly compiled data 
on the occurrences of injuries in the research projects they support).

The proposed solution thus addresses both ethical and practical 
problems and should not be difficult to implement.  Principal 
investigators receiving government funds are already required by 
Federal grant and contract rules to submit annual progress reports as 
well as final reports on their research. although data all but the 
numbers of research subjects, injures and consequences may well be 
included in the narratives of such reports, they are not easily 
retrievable.  The commission was informed that with few exceptions, 
such data are not routinely collected and collated by any one office 
within the various departments.  An expeditious and inexpensive way to 
collect data on the universe of research subjects and research-related 
injuries would be to develop a form to be completed annually by all 
principal investigators as part of there progress and final reports on 
each project.  Completed forms should be sent to the office within the 
Department of Health and Human Services that has "lead" responsibility 
in the field.  The data would then be collated so that information 
about the number of subjects and incidence of injury would be readily 
available.  Indeed, the Veterans Administration has already 
implemented just such a requirement for all research it sponsors.


Although this system should provide basic information about risks and 
injuries in research, it will not reveal the costs and benefits of 
having a compensation program.  Therefore, the Commission recommends 
that a modest social policy experiment be conducted to determine the 
need for, and feasibility of, comprehensive or partial programs to 
compensate injured subjects.

During the past two decades the Federal government has conducted a 
series of experiments in which new social policies have been 
implemented under reasonably controlled conditions and evaluated in 
order to see how specific aspects of the policies work in practice.  
Such experimentation has been used to test the effects of instituting 
a negative income tax, housing allowance, and various types of health 
insurance plans. 4[U.S. Office of Economic Opportunity, PRELIMINARY 
RESULTS OF THE NEW JERSEY GRADUATED WORK INCENTIVE EXPERIMENT CONDUCTED 
BY THE OFFICE OF ECONOMIC OPPORTUNITY (February 18, 1970): see also 
Alice Rivlin, SYSTEMATIC THINKING FOR SOCIAL ACTION, Brookings 
Institution, Washington (1971) at 94-108.]  The use of social policy 
experimentation has been encourage by a broad range of scholars and 
policy analysts as a "rational" approach to policy innovation.
[Henry W. Riecken and Robert F. Boruch, eds., SOCIAL EXPERIMENTATION: 
A METHOD FOR PLANNING AND EVALUATING SOCIAL INTERVENTION, Academic 
Press, New York (1974) at 1-39; Rivlin.supro note 4: Campbell, Reforms 
as Experiments, 24 AM. PSYCH.419 (1989);]  The great advantage of such 
experimentation is that it allows innovative programs to be tested and 
proven useful or deleterious before a decision is made about instituting 
them on a comprehensive basis.  The government can then modify or 
abandon those programs before public expectations are crested or 
constituencies formed that force the government to remain committed to 
ineffective programs.  Given the importance of empirical data in 
evaluating the need for, and feasibility of, a compensation program, a 
limited experimental trial seems particularly appropriate.

The Commission recommends that the compensation experiment be designed 
and administered by the department of Health and Human Services with 
appropriate consultation with other governmental bodies which sponsor 
or conduct research.  the relevant office within the Department would 
develop (or contract to have developed) the detailed plans for 
experiment, including its administration and evaluation.
5[ The office within HHS that is most familiar with research involving 
human subjects is the Office for Protection from Research Risks at 
NIH.  that office does not, however, have extensive experience with 
design or evaluation of social experiments.  Thus, the Department may 
wish to turn to the Office of the Assistant Secretary for Planning and 
Evaluation and perhaps to other bodies with expertise in the field in 
assigning responsibility for the design and conduct of the research 
compensation experiments recommended here.]


Specific Aims of the Experiment

The experiment with formats for compensation should be designed to 
answer the following questions:

(1)  How great is the need for a program to compensate for research 
injuries?

(a)  How many research subjects are exposed to risk of injury?
(b)  How many research injures occur?
(c)  How serious are the injuries?
(d)  For what proportion of these injuries is medical assistance 
currently provided by formal or informal procedures (i.e., how 
effective are such procedures in meeting existing medical needs)?
(e)   For what proportion of these injuries are payments currently 
made for loss of wages and other expense (i.e., how effective are 
existing means in meeting economic needs.
(f)  To what extent does failure to provide adequate compensation 
create serious problems for injured subjects of for the research 
enterprise?
(2)  What effects result form introduction of a compensation program?
           (a)  Does the institution of a formal compensation program 
undermine current, informal methods of providing for the immediate 
medical needs of injured subjects?
           (b)  Would the availability of compensation bring more 
research injures to light (i.e.,encourage the reporting of injures 
which would otherwise not be known to personsother than those 
immediately involved?
           (c)  Does the availability of compensation lead to large 
numbers of specious claims?
(d)  Does the availability of compensation make it easier for 
researchers to recruit subjects?
(e)  Does the assurance that inured subjects will be compensated 
produce an environment that is more conducive to research, and how 
does it affect the nature and amount of research?
(3)  What costs and benefits are entailed in a compensation program?
(a)  How many subjects would be anticipated to be eligible to receive 
funds under a compensation program?
(b)  Can those administering a compensation program distinguish 
specious from valid claims?  What the administrative costs of making 
such determinations?
(c)  How difficult is it in practice to discern a causal connection 
between research interventions and subsequent adverse effects?





  What approaches tot he determination of causation seem most effective 
and just?
      (d)  What are the cost, both in claims paid and in 
administrative expenses, of a compensation program?
      (e)  How would the costs be affected by the inclusion of 
Federally supported as well as Federally conducted research?
      (f)  How would the costs be affected by covering those injuries 
caused by procedures undertaken solely as an aid to the research 
design and not for the patient-subject's welfare, when in the absence 
of such "nonbeneficial procedures" therapeutic research would be 
excluded from the compensation program?  Can such injures be reliably 
identified?
     (g)  What approached to compensation provide the most favorable 
balance of benefits against administrative burdens and costs?

Research Plan

Although the actual design of the experiment is left to the office 
within HHS that conducts it, the remainder of this chapter is intended 
to indicate the type of research to Commission views as desirable and 
to provide general guidance concerning the scope and nature of such 
research.  Particular features of possible compensation programs are 
discussed in Chapter Seven.

The Commission proposes that the three sets of question listed above 
be answered by experimental compensation programs established at a 
small number of institutions and by ongoing collection and analysis of 
data from these institutions and matching institutions without 
compensation programs.  Questions concerning the costs and problems of 
administering a compensation program would be answered by establishing 
and monitoring experimental compensation programs at institutions 
willing to participate in the experiment.  Uniform data should be 
collected on the number of subjects participating in research, number 
of injures, seriousness of injuries, and so forth.  Additionally, all 
injuries resulting in claims for compensation should be documented, 
including the amount of money paid out in compensation, administrative 
cost, number of claims honored and denied, and the details of each 
incident.  During the experiment, the Commission recommends that 
compensation be restricted to nontherapeutic experiments and perhaps 
to injuries caused by procedures undertaken as an aid to the research 
and not with the intent of providing possible therapeutic benefit to 
patient-subjects.

In addition to providing information about the general feasibility of 
a compensation plan, the experiment should provide a basis for 
determining whether any type of compensa-











108


   	  	    FOR REFERENCE SEE (16bb12.gif)




institutions in which research is conducted.  Institutional settings 
for research range from the Clinical Center at the NIH to 
universities, medical schools and hospital, to small independent 
research facilities.  On approach that might be used is known as 
"quota sampling."  Institutions involved in Federally conducted or 
supported research would be grouped in appropriate categories.  A 
sample containing a  certain number (quota) of institutions from each 
category would then be drawn. Within each category of institutions in 
the sample, half would initiate an experimental compensation program: 
the other half would be monitored in a similar manner but would not 
have a compensation program.  Under ideal circumstances, the selection 
of institutions from each category in the sample would be random, as 
would be the assignment of institutions within each category to have 
or not have a compensation program.

The Commission believes this basic approach can and should be applied 
to Federally conducted and Federally supported research.  However, the 
Commission believes that the NIH Clinical Center, because of its 
centrality in Federal biomedical and behavioral research and its 
traditional leadership role in the scientific community, should not be 
randomly assigned but should participate in the experiment as a 
compensating institution.  If a successful compensation program cannot 
be established at the NIH Clinical Center, the Commission is skeptical 
that any such program should be imposed by the Federal government on 
other research institutions.

With respect to Federally supported research conducted at non-Federal 
research institutions, the major issue is considering sample selection 
is whether participation should be mandatory (i.e., tied to federal 
funding) or voluntary.  scientifically, mandatory participation is 
preferable since it reduces the likelihood of biased sampling that 
would occur if the "self-selected" institutions that volunteered to 
participate differed as a group from those that do not.  The 
Commission see several advantage in voluntary participation, however, 
and on balance believes that, if possible, participation should be on 
a voluntary basis.  the assistance of groups such as the AAMC which 
support a limited, experimental evaluation of research injury 
compensation should be sought in persuading institutions to 
participate in the experiments.

Two approaches might be used to reconcile the desirability of 
participation on a voluntary basis with the need for random selection 
and random assignment to compensating or noncompensating categories.  
One approach would be to select the sample and assign the chosen 
institution to compensating or noncompensating status on a random 
basis, but to allow institutions to "opt out" of the experiment if 
they were unwilling to participate in their assigned role.  An 
alternative approach would be to solicit institutions to volunteer to



participating in the experiment on the understanding that they would 
then be assigned randomly either to the experimental (compensating) or 
control group.  Under either approach it would be possible for the 
government to offer institutions incentives in order to assure that a 
sufficient number of institutions would be willing to participate.  
For institutions that agreed to participate in the experimental 
compensation program, the government might, for example, offer to 
cover the costs of compensating for injuries that occurred on 
non-Federally funded as well as Federally funded research (provided, 
of course, that the former as well as the latter passed through IRB 
review).

Sample Size.  In determining the number of institutions that will 
compose the sample for this study, the implementing agency will need 
to balance the need for reliable data with the need to minimize the 
costs of conducting the experiment.  The following considerations are 
relevant to this decision:

(1)  The degree of variation among research institutions imposes 
certain requirements.  If institutions are divided into categories for 
quota sampling, at least two institutions from each category should be 
included in the sample (one to initiate a compensation program, the 
other not).  thus, the number of categories determined to be necessary 
to reflect institutional diversity will have a direct bearing on the 
sample size.  

(2)  Given that institution are the unit of analysis, a sufficient 
number of institutions must be studied to obtain adequate statistical 
sensitivity (i.e., an adequate ability to detect differences) when 
comparing compensating and noncompensating institutions.  Otherwise, 
substantial real-world differences between compensating and 
noncompensating institutions would fail to show up as "statistically 
significant effects" in the experimental data.

Duration of Study.  Three considerations argue for conducting the 
study over a period of several years.  First, the incidence of 
research injuries is expected to be quite low.  Unless the study 
continues long enough to allow a fairly large "sample" of injuries to 
occur, there is a danger that the particular experience with injuries 
observed in a limited period may, by chance, be unrepresentative of 
research injuries in general.  Second, because research injuries may 
not become know immediately, a study of short duration may 
underestimate the rate of injuries.  Finally, because initial 
reactions to a novel program may be unrepresentative of later 
reactions, a valid evaluation of a compensation program my not be 
possible until it operates long enough to become established and 
known.

Cost

Until the parameters of the recommended experimental evaluation of 
compensation approaches are more fully delineated--a task which the 
Commission believes can only be





carried out by those with final responsibility for designing the 
experiment--it is impossible to estimate the costs of the experiment.  
Rough yardsticks are provided by the experience of the University of 
Washington, where commercial premium costs were less than $1.00 per 
covered subject (and actual payments were less than 5c per subject), 
and by the Quincy Research Center, where costs were less than those 
associated with workers' compensation premiums for professional 
employees.  In addition to premium costs, of course, the experiment 
will incur certain administrative costs and the costs of design and 
evaluation.

The Commission notes that under existing Public Health Service policy, 
reasonable insurance costs incurred by contractors and grantees are 
already allowable expenses.  Thus, the implementation of a 
compensation experiment would not create a new category of expense 
insofar as insurance premium costs were concerned.  By statute, funds 
are available to HHS for research and development experiments on 
programs.  The Commission believes that such funds might usefully be 
applied to the compensation experiment, and that a satisfactory 
experiment could be conducted for a very small fraction of 1% of 
current Federal expenditures on biomedical research grants and 
contracts.

Standards for Evaluating the Desirability of Instituting a 
Compensation Program

The ethical desirability of compensating victims of research injuries 
is strong, but the decision whether to adopt a compensation system 
must in the Commission's view be made by balancing the seriousness of 
the problem against the cost and practical difficulties of the 
solution.  A finding of substantial need would argue strongly in favor 
of a compensation program even if significant costs and practical 
difficulties might be entailed.  If the need for a compensation 
program is shown to be small, however the cost and difficulty of 
administering make a permanent program seem much less necessary or 
desirable.

A finding that relatively few serious injuries occur to subjects in 
research should not in itself be taken to indicate that no 
compensation program is needed.  Few serious injuries resulting from 
properly conducted research have been reported or documented. The 
examples of research injuries most frequently recited (i.e., the 
Tuskegee syphilis experiment, the CIA-sponsored psychoactive drug 
research, and so forth) occurred prior to the existence of the review 
processes required by current Federal regulations.  Nonetheless, it is 
undeniable that human research involves some risks.  In the long run, 
it may well be more protective of the field of biomedical and 
behavioral research for it to be prepared to meet any human disaster 
which may result from its activities.



The alternative--a knowing decision to do nothing to provide a fair 
and simple avenue of relief in the event of serious injury--carries 
serious danger for public confidence in science and scientific 
research.  Moreover, any program adopted in the wake of such and event 
would almost certainly be less well-designed than one adopted in light 
of the results of a well-conducted social policy experiment.
7 [For a fuller discussion of the issue of compensation for injuries 
arising from certain causes in therapeutic research, see pp. 132-135.  
The terms are defended and their practical interrelationships are 
shown in Appendix S.]





Consider Various Features of Nonfault Insurance Programs	7

What features might be tried out as part of the social policy 
experiment proposed by the Commission?  One alternative, rejected by 
the Commission, would be to create a remedy through the judicial 
system.  It does not seem wise to remit injured subjects "to the 
uncertainties of the law court"1 [Tuskegee Syphilis Study Ad Hoc 
Advisory Panel. FINAL REPORT, U.S. Department of Health, Education 
and Welfare, Washington (1973) at 23.] or to required them to hurdle 
the formidable barriers to recovery that exist even in strict tort 
liability.

The Commission also examined various governmental health and 
compensation programs and private insurance plans to determine whether 
they might serve as models for further study and experimental 
evaluation regarding compensation for research injuries.  While the 
existing programs also appear to be inadequate, they point to the 
possibility of a non-judicial, nonfault system of compensating for 
research injuries.  The factors favoring this approach are nicely 
summed up in a report prepared by the Ciba Foundation study group 
established in response to the Pearson Commission's recommendation 
favoring strict liability:

Compared with litigation a no-fault compensation scheme would have the 
advantage of administrative simplicity.  Claims can generally be 
handled quickly and, since the claimant would not be seeking redress 
from the researchers, his employer or the funding agency, it would be 
proper for the researcher to assist the claimant  in the preparation 
of his claim.  The researcher and the participant in the research 
would, therefore, not be adversaries as would be inevitable in any 
scheme based





upon litigation.  It is in the public interest that all participants 
in medical research, should appreciate that their well-being, and the 
public interest, are overriding consideration in the mind of the 
investigator. 2[Ciba Foundation Study Group, MEDICAL RESEARCH: CIVIL 
LIABILITY AND COMPENSATION FOR PERSONAL INJURY, Ciba Foundation, 
London (1980) at 9.] 

This chapter first reviews the governmental and private insurance 
models and presents the reasons for their rejection.  The rest of the 
chapter discusses the varying features that might be considered for 
nonfault insurance programs to compensate for research injuries.


Existing Governmental Programs as Models

The Commission began its study of existing Federal models by reviewing 
the material prepared for the HEW  Task Force on the compensation of 
Injured Research Subjects in 1975.  Those materials included detailed 
descriptions of four major Federal compensation programs: the Federal 
Employees' Compensation Act (F.E.C.A.), the Black Lung Benefits Act, 
the National Flood Insurance Act, and the Veterans Administration 
service-connected disability program.  The report prepared for the 
Task Force on these programs appeared in an appendix to the Task 
Force's final report and is reprinted, for convenience, in the 
appendix volume accompanying this Report. 3[Frederick G.G. Barber. 
Report on Federal Models, HEW Secretary's task Force on the 
Compensation of Inured Research Subjects. REPORT. U.S. 
Department of Health, Education and Welfare.  Washington (1977) 
(hereinafter cited as TASK FORCE REPORT) Appendix A at 167-198.] 

The Commission then undertook to update these material and to explore 
certain programs and issues in greater depth.  Some of this work was 
carried out by a health policy analyst who reviewed in some detail the 
suitability of F.E.C. A. and the Veterans Administration program as 
models for compensation for research injuries, brought prior work on 
the black lung program up to date, and provided information on Social 
Security disability programs, Medicare, and state workers compensation 
programs. 4[Stanly B. Jones, Existing Federal Programs as Models for 
Compensation of Human Subjects (1980):see Appendix M to this Report.] 

Members of the Commission staff also canvassed the literature and 
reviewed administrative and legislative proposals regarding 
compensation for injuries associated with vaccine and immunization 
programs, radiation fro nuclear weapons testing, toxic substances in 
the environment, and medical accidents, as well as proposals for 
modifications in product liability law.  These inquires took advantage 
of ongoing studies concerning compensation programs in other




Federal agencies and in interagency working groups.  Particular 
attention was paid to the lessons of the swine flu vaccine program, 
including the continuing litigation associated with it, and to 
contemporaneous studies by the Congressional Office of Technology 
Assessment of compensation for vaccine-related injuries.

F.E.C.A and Other Workers' Compensation Programs.  F.E.C.A. is the 
workers' compensation plan applicable to approximately three million 
civilian Federal employees, and to certain other persons as well.
5 [U.S.C. Ch.81 &&8101 est seq.  See generally Comptroller General's 
Report to the Congress.  COMPENSATION FOR FEDERAL EMPLOYEE INJURIES; 
IT'S TIME TO RETHINK THE RULES, HRD-79-78 (August 22, 1979) 
(hereinafter cited as Comptroller General's Report on Compensation).]   
The plan is generally similar in structure e to most state 
workers' compensation plans, although more comprehensive in scope and 
more generous in amounts (Particularly for highly paid employees) than 
state plans,  The program is administered by the Office of Workers' 
Compensation Programs within the Department of Labor.

F.E.C.A. provides benefits to covered employees both for injuries 
resulting from accidents and for diseases proximately caused by their 
employment.  Benefits are provided through an administrative mechanism 
without regard to fault (except for willful misconduct or intoxication 
by the injured employee)and include compensation for loss of wages, 
dollar awards for bodily impairment or disfigurement, medical care for 
an injury or disease, rehabilitation services, and compensation to 
survivors in the event of an employee's death.  Compensation levels 
are established on the basis of a percentage of an employee's pay 
level.  Awards are subject to administrative review but are not 
subject to challenge in the courts.

F.E.C.A. has been extended, by statute, to apply to a number of "non 
typical" types of Federal; employees and, indeed, non employees.  
Among the persons currently covered by F.E.C.A. are Peace Corps and 
VISTA volunteers, members of the Civil air Patrol and National Teacher 
Corps, Job Corps enrollees, certain student-trainees, and non-Federal 
law enforcement officers injured in the course of enforcing Federal 
law.  Methods have been devised to accommodate the special 
compensation of such individuals.

F.E.C.A. is not the only Federal legislation providing workers' 
compensation benefits.  The federal Longshoremen's and Harbor Workers' 
Compensation Act (the "Longshoremen's 





Act") governs workers' compensation for privately employed persons 
engaged in maritime employment. 8[33 U.S.C. Ch. 18, &&901 es seq. 
Other workers; compensation plans emboided in Federal legislation 
include the Jones Act, covering merchant seamen, and the Federal 
Employers' Liability Act, covering railroad employees.]  The 
Longshoremen's Act has been extended, by statue to provide 
coverage for several additional classes of workers, including persons 
employed at military, air, and naval bases outside the United State: 
employees of "nonappropriated fund instrumentalities" (e.g., military 
base post exchanges):employees working on the outer continental shelf 
under United States jurisdiction; and, until recently employees in the 
District of Columbia. 9[See 42 U.S.C. Ch. 11, &&1351-1654; 5 U.S.C. 
&& 8171-8173; 48 U.S.C. &&1335-1343; D.C. Code &&501-504, repealed by 
District of Columbia Workers' Compensation Act of 1979, D.C.Law 3-77.] 

Benefits under the Longshoremen's act are generally similar to, but 
somewhat less generous than, those available under F.E.C.A.  Unlike 
the situation under F.E.C.A./, awards under the Longshoremen's Act are 
paid by private employers and may be appealed in the Federal courts.  
This administrative mechanism is thus subject to several of the 
criticism leveled against common law tort remedies which entail resort 
to the courts.  The Longshoremen's Act does, however, provide a 
Federally recognized alternative to F.E.C.A. with benefit levels 
closer to those provided under state worker's compensation laws.

In addition to these Federal programs, all fifty states have their own 
workers' compensation acts requiring employers to provide benefits to 
workers who suffer work-related injuries and diseases.  To be eligible 
or benefits, normally an employee must experience a "personal injury 
by accident arising out of and in the course of employment.
10[Report of the National Commission on State Workmen's Compensation 
Laws, U.S. Government Printing Office, Washington (1972) at 32. ]
Beyond this common framework, eligibility criteria and benefit 
levels vary widely among the states.  Many state plans are criticized 
as inadequate, particularly with respect to coverage of work-related 
illnesses and disease.  A Federally mandated National Commission on 
State Workmen's Compensation Laws made numerous recommendations for 
improvements in this 1972 report;11 [1 Id.]1 many of these 
recommendations have yet to be implemented.

What, then are the lesson of Federal and state workers' compensation 
programs, and to what degree do existing programs, particularly 
F.E.C.A., provide use models for a compensation system for research 
injuries?  The problem of injured research subjects seeking 
compensation is similar in several respects to the position of the 
injured industrial worker, suggesting the plausibility of the worker's 
compensation approach as a model for research injuries.  The history and 






rationale for workers' compensation law underline some of those 
similarities, as well as a few significant differences.

Workers' compensation programs were established in the early art of 
this century in response to growing dissatisfaction with the tort 
system as a vehicle for resolving claims arising from work-related 
injuries.  The new approach abandoned several requirements under the 
tort system viewed as unfair or unduly burdensome to inured workers, 
and permitted workers to recover if their injuries arose in the course 
of their employment, without proof of fault by the employer.  In this 
respect, workers'' compensation law recognized industrial accidents as 
an inevitable hazard of modern industry.  The new approach also 
recognized that he hum cost of injury, like the cost of repairing 
machinery, were part of the costs of the enterprise and should be 
borne by those who benefit from the enterprise, not by the inured 
workers.  This philosophy found expression in a slogan of the time: 
"The price of the product should bear the blood of the working man's."  
Scholars of the period believe the adoption of worker's compensation 
laws contributed to industrial harmony and social peace.[2 Comptroller 
General's Report on Compensation, supra note 5, Appendix 
II and see generally roy Lubove, THE STRUGGLE FOR SOCIAL SECURITY, 
1900-1935. Harvard University Press. Cambridge, Mass. (1966) at 45-65.]
2 

At present injured research subjects enjoy no benefits comparable to 
workers' compensation, although there are notable  similarities with 
the situation of ordinary American workers.  Injured subject, like 
workers a century ago, face special obstacles in securing relief 
through the tort system.  Because of the special character of the 
research enterprise as a venture into the unknown, research injured 
are an inevitable hazard of the enterprise.  Further, the human costs 
of injury incurred in research are truly costs of the research 
enterprise to society--cost necessarily incurred to obtain the 
benefits of scientific progress.

It may be argued that research subjects volunteer to participate in 
research, while workers effectively have little choice about incurring 
workplace hazards to earn their livelihoods.  To the extent that 
subjects participate in research purely out of altruistic motives, 
this observation may be correct.  To that extent, the analogy between 
workers' compensation and compensation research injuries is imperfect.  
Yet, on the ethical plan, that difference adds weight to the injured 
subject's claim against society, since the volunteer has acted 
altruistically.  And to the degree that some subjects participate in 
nontherapeutic research as a convenient means of casual 
employment-for-pay (a not uncommon practice amount university and 
medical school students), parity of treatment suggest





that subjects should be entitled to workers' compensation benefits 
much as are other employee-for-pay.

The worker's compensation analogy does not tend to break down with 
respect to one class of subjects: patients who participate as subjects 
in therapeutic research with the expectation of securing personal 
health benefits.  Such subjects may seem more akin to patients 
receiving standard medical care (who are not entitled to compensation 
on a nonfault basis) than to workers eligible for workers' 
compensation.  Further, such patient-subjects pose many of the 
administrative difficulties which workers compensation programs have 
been least successful in answering: determining causation of diseases 
that are not arguably work-related; measuring the degree of "excess 
injury," or the exacerbation of preexisting health conditions, and 
calculating the benefits owed to person not currently in the work 
force.  For these reason, the Commission believes that the uncritical 
application of workers' compensation models to therapeutic research 
involving already sick patients-subjects is fraught with potential 
hazards.  However, with respect to nontherapeutic research (especially 
that involving healthy subjects), the Commission concluded that 
workers' compensation, and F.E.C.A. in particular, offers a potential 
model for compensating injured research subjects.

Veteran Benefits.  A second approach carefully examined by the 
Commission is the compensation system administered by the Veteran 
Administration for service-connected disability of death.  under this 
program, benefits are provided to veterans for disabilities incurred 
in, or aggravated by, a period of military service though not 
necessarily "caused" by the service.  Benefits include health care and 
rehabilitation, typically provided in VA facilities, and financial 
compensation for lost earning capacity due to the disability.  These 
benefit programs make no distinction between men and women who 
volunteered for service and those who were conscripted.

In marked contrast to workers' compensation approaches, VA 
service-connected disability benefits are payable without regard to 
the beneficiary's employment status.  Rather, benefits are determined 
according to an "average man" concept.  Disabilities are rated and 
compensated according to a comprehensive schedule of injuries and 
diseases, under which varying percentage rating are assigned, 
depending on the degree of severity of any given disability.  These 
percentage rating represent, as nearly as possible, the "average 
impairment in earning capacity" resulting from a particular disability 
in civilian occupation.  In applying the "average man concept" there 
is an assumption that all young people who entered military service at 
an early age would have the potential for attaining an "average 
civilian occupation income" at the time of their discharge.  These 
disability payment levels are adjusted yearly by Congress, to 
correspond to cost-of-living increases





provided in other Federal benefit programs,  This "income loss 
compensation" can be reduced if it can be demonstrated that the 
individual was already disabled before entering service, in a way 
definable in the schedule for rating injuries and diseases.  In that 
event, only the "additional disability" is compensable.
[3 Letter from Dorothy C. Rassinski, M.D., J.D., Associate Director, 
Medical-Legal Affairs, Veteran Administration, to Alan J. Weisbard 
(1980).]3   Research can be prevented from escalating uncontrollably 
administrative cost.  Claim decisions are subject to administrative 
review, but because of the relatively simple administrative structure 
of the program (and the existence of other VA programs providing for 
the needs of the veteran), few appeals are taken.  No appeals to the 
courts are permitted.

The Commission believes a number of features of the VA 
service-connected disability program bear close examination in the 
structuring and experimental evaluation of alternative compensation 
plans for inured research subjects.  In particular, the "average man" 
concept may have promise in determining benefit levels for injured 
subjects who current income, if any, may not accurately reflect true 
earning capacity.

The Commission's health policy consultant noted that "if participation 
in research experiments were deemed by law to be equivalent to service 
in the military, research subjects might be covered" by the VA 
service-connected disability program.[4 Jones, Existing Federal 
Programs as Models for Compensation of Human Subjects, supra note 4, 
33-34.]4   The Commission believes this suggestion bears further 
exploration, but notes that VA benefits, unlike those afforded under 
F.E.C.A. or the Longshoremen's Act, have not historically been 
extended to new groups of beneficiaries.

Vaccine and Immunization Programs.  Many of the issues involved in 
recent Congressional and Executive Branch efforts to formulate a 
national  policy on compensation for injuries in vaccine and 
immunization programs bear a close relationship to questions arising 
in the context of research injuries.  the Commission notes that the 
approaches under consideration for vaccine injuries are closely 
analogous to those examined here.[5 Office of Technology Assessment, 
U.S. Congress, A REVIEW OF SELECTED FEDERAL VACCINE AND IMMUNIZATION 
POLICIES (1979); Office of Technology Assessment, U.S. Congress, 
COMPENSATION FOR VACCINE-RELATED INJURES(1980); Reports and 
Recommendations of the National Immunization Work Groups (1977).]5 

The Commission agrees, for example, with the conclusion of the Office 
of Technology Assessment that "the swine flu program is widely 
regarded as exemplifying the problems inherent in compensating for 
vaccine related injuries via the tort law system.[6 COMPENSATION FOR 
VACCINE-RELATED INJURIES supra note 15 at 13.]6 The preference of the 




President's Commission that nonfault administrative and insurance 
mechanisms be explored as a means to compensate for research injuries, 
rather than pursuing remedies through the tort law system (including 
strict liability), is based I part on the swine flu experience.
[7 The Federal government's ill-fated 1976 effort to immunize the 
public against swine flu nearly ended before it began, when vaccine 
manufactures refused to transfer their vaccines from bullet to bottles 
for distribution, because the manufacturer were unable to secure 
insurance coverage against potential liability for  vaccine-related 
injuries.  Emergency legislation was rushed through Congress, proving 
that any claims arising from the wine flu program should be filed 
against the Federal government, while preserving the government's 
rights to sue manufacturers for indemnification in the event of 
negligence in the production of the vaccines.  Public Law 94-380. See 
R. E. Neustadi and Harvey V. Fineberg, THE SWINE FLU AFFAIR: DECISION 
MAKING ON A SLIPPERY DISEASE, US. Department of Health, Education and 
Welfare, Washington (1978).  Under this legislation, the government 
assumed the "duty to warn" potential vaccine recipients of any known 
adverse reaction to the vaccine, and undertook to defend against any 
claims brought for vaccine injuries. 

The result are well known.  Although the feared swine flu epidemic 
never materialized, there quickly developed a mini-epidemic of claims 
against the Federal government, many related to the onset of claims 
against the Federal government, many related to the onset of 
Guillian-Barre Syndrome (GBS).  As of August 12, 1981, some 4039 
claims for a total of over $2,89 billion in damages had been filed.  
Of these claims, 1523 had progressed lawsuits, requiring extraordinary 
efforts on the part of lawyers at the Justice Department, as well as 
imposing considerable expense on the government, the possibility that 
the government might have prevailed on the merits in any cases (on the 
basis that the government should noble held accountable for a failure 
to warn of risk that were unknown at the time) was put in question 
when then HEW Secretary Califano issued a statement accepting the 
government's responsibility for GBS claim.  see Compensation for 
Vaccine-Related Injuries supra note 15. at 13-17.  While the legal 
effect of this statement is open to question, The Justice Department 
subsequently took the position that it would not require proof of a 
theory of liability in cases in which GBS was established (although 
proof of liability is required in all other cases).  As a result, the 
government incurred both the substantial cost of courtroom litigation 
and liability or the large damage award associated with such 
litigation.  It is difficult to imagine how an administrative 
mechanism for nonfault compensation could do worse.]7 

Evaluation of Governmental Alternatives.  The Commission has noted the 
tendency of governmental compensation programs, once enacted, to 
expand beyond their originally contemplated scope, placing ever 
greater demands on the Federal treasury.  On significant factor 
underlying the escalation in costs of many Federal compensation 
programs has been the general rate of price inflation throughout the 




economy, and the still higher inflation in medical costs.  The 
Commission sees little in the design of a compensation program for 
research subjects which could significantly affect this overriding 
fact.

A second factor responsible for escalating costs is the power of 
particular political constituencies to expand the scope and benefit 
levels of compensation programs.  To some degree, such expansions may 
serve legitimate policy interest; often it is merely a matter of 
politics.  One of the most striking features of the research injury 
question is the near absence of any active constituency group of 
research subjects.  If the Commission's data on the incidence of 
research injuries is correct, it is exceedingly unlikely that any 
large or powerful constituency group representing injured research 
subjects will arise.

The presence or absence of a powerful contingency group does not 
affect the moral claim to compensation of those subjects who are 
injured in research.  The lack of such a group does, however, 
diminishes the likelihood that overwhelming political pressures will 
force an initially modest compensation program to expend beyond its 
appropriate contours.

Nonetheless, the commission has paid special attention , to 
articulating its several alternative approaches toward compensation, 
to those aspects of a compensation program which might expend the 
scope and expense of a program beyond those initially contemplated.  
Perhaps most significant among nonphysical injuries; injuries 
sustained in therapeutic research and injuries sustained in research 
not conducted or supported by the Feral government.  Careful study of 
benefits and costs should precede any decision to include these 
categories in a compensation program.

A third reason why the scope of compensation programs has expanded and 
their costs have escalated is the tendency to eligibility and benefit 
criteria.  Therefore, the Commission recommends that court review be 
precluded initially and dollar ceilings be set on awards as features 
of any program being tested but; reexamination of these features ought 
to await several yeas experience with an actual program in place.

A fourth basis for escalating costs is greater awareness by potential 
beneficiaries on the compensability of injuries.  To the degree such 
awareness leads persons to make claims who is legitimately entitled to 
compensation, any "escalation" in cost is a sign that the program is 
doing its job and reaching the persons intended to receive benefits.  
However, greater awareness of the availability of compensation may 
also lead to specious claims by persons not legally entitled to 
relief.  Cost associated with such claims include both any benefits 
erroneously paid out to undeserving claimants and the administrative 
expense of processing invalid claims.  This problem of poten





tially excessive "induced costs" was a significant concern of several 
witnesses who questioned the wisdom of a compensation program and is 
seriously regarded by the Commission. [8 Testimony of George Bernstein, 
Esq., transcript of 6th meeting of the President's Commission (January 8, 
1981) at 189-95.]8  The Commission's recommendation that any 
compensation program be tested first through a small experiment is 
based on the conclusion that the potential for induced costs must be 
carefully evaluate before Federal policy makers consider instituting a 
comprehensive plan or promulgating a requirement that research 
institutions adopt such plans.

Insurance Mechanisms as Models

Members of the Commission staff conducted extensive and fruitful 
discussions with representatives of the insurance industry--including 
company officials, insurance brokers, trade association officers, and 
institutional risk manager--to determine what role insurance 
mechanisms could usefully play in assuring medical care and financial 
redress for injured subjects.  Several conclusion emerged from these 
discussions.

The injuries (and other interested parties) made a number of specific 
suggestions concerning the terms of a compensation program.  Their 
submissions are published in the appendix volume accompanying this 
Report.[9 See Appendices O.P.Q.R. to this Report.]9   The focus 
here is not on these specific suggestions, but broader 
structural approaches to incorporating insurance mechanisms as part of 
a compensation system.  At least three distinct models (as well as 
combinations of these models) may be envisioned: institutional private 
insurance, institutional self-insurance, and pooled or collective 
insurance.

Institutional Private Insurance.  Institutional private insurance is 
perhaps the most familiar model.  Under this approach, each research 
institution would purchase from commercial insurance companies, 
nonfault coverage for all subjects participating in covered research 
at that institution.  Terms of coverage would be negotiated by the 
relevant institution and the insurance carrier, subject to minimum 
requirements.  Premiums would be set according to normal underwriting 
standards, reflecting the type of institution, the mix of research 
conducted, and any prior claims experience.  Coverage could be 
marketed either as separate insurance product or more likely, a part 
of the overall insurance package provided for the institution.

This is the approach successfully employed at the University of 
Washington (Seattle) for a number of years, originally in conjunction 
with the Argonaut Insurance Company and subsequently,




between 1973 and 1979, with the Aetna Casualty and Surety 
Company.[0 Diane McCann and John R. Pettit, A Report on Adverse 
Effects Insurance for Human Subjects (1980): see Appendix H to this 
Report at 1.]0   The insurance carriers in that instance wrote 
coverage through a special endorsement to the university's 
comprehensive general liability insurance policy.  Coverage was 
originally provided following a survey of adverse effects associated 
with research at the university during a one-year base period, at a 
premium cost of 50c per subject, or $17,500.  In subsequent years, the 
cost of insurance coverage was a fixed charge of $35,000, without
regard to the number of subjects covered.  The university's human 
subjects officer and risk manager reported that the insurance program 
"relieved concerns within the university for the economic protection 
of human subjects" and that "there have been no complications in the 
insurance program."[1 ld. at 3. The University of Washington has for 
several years been a self-insurer.  The total amount paid in claims to 
injured subjects in the nearly 10 years it has had a formal 
compensation program averages less than 5c per subject.  This amount 
does not include the value of free medical services provided to 
subjects. ]1   

The major drawbacks of this private insurance approach are, first, 
that the terms of coverage must be sufficiently attractive to induce 
participation by the private insurance industry (which implies 
significant limitations on the extent of coverage) and, second, that 
some research institutions may have difficulty in securing coverage.  
As noted by representatives of the private insurers, "the larger and 
more respected institutions will obtain coverage more easily," while 
other institutions "may be unable to obtain such coverage because 
their projects are too risky or because it is impossible to assess the 
degree of risk in the program.  This will be particularly troublesome 
with respect to research in smaller institutions engaged in high risk, 
even if potentially beneficial, research.[2 See not 18 supra.]2 

Institutional Self-Insurance.  The second model is that of 
institutional self-insurance.  Here the institution itself defines the 
terms of coverage and establishes an actuarially sound sinking or 
reserve fund to cover future claims.  No private insurance carrier is 
involved.  As a result, the terms of coverage are potentially quite 
flexible, and many institutions find self-insurance less expensive and 
more satisfactory for their needs than insurance procured on private 
insurance markets.  An increasing number of universities and hospitals 
have switched from commercial to self-insurance in recent years.  For 
example, the University of Washington has for three years provided its 
own liability coverage, including insurance for adverse effects 
associated with research, through a program on self-insurance.





There are several drawbacks to institutional self-insurance, however, 
particularly for smaller research institutions.  Since each 
institution establishes a fund to cover its own claims, risks are not 
spread as widely as with commercial insurance.  Thus, in the event of 
an unusual string of losses (or even a single catastrophic loss), the 
reserve fund may be endangered, requiring claim to be paid out of 
general funds.  Further, may states have laws limiting the ability of 
state instrumentalities or nonprofit institutions to self-insure.  In 
the state of Washington, a legislative change was required before the 
state university could self-insure.

The compensation program for research injuries at the Quincy Research 
Center, a drug testing facility in Kansas City, Missouri, combines 
elements of self-insurance and private insurance.[3 John D. Arnold, 
Incidence of Injury During Clinical Pharmacology Research and 
Indemnification of Injured Research Subjects at the 
Quincy Research Center (1980): see Appendix I to this Report.]3   
The research center has provided medical care under a 
self-insurance approach for all research participants since 1975.  In 
additions, the research center provides financial compensation in the 
event of injury through a workers' compensation-type insurance 
arrangement.  The director of the research center reports that 
standard rates for research subjects as employees are approximately 
twice as great as those for physicians and consultants, approximately 
five times as great as those for clerical office employees, slightly 
less than one-third as great as those for chauffeurs and drivers, and 
approximately equal to those for analytical chemist.[4 id. at 34-35.]4 

Collective or Pooled Insurance.  A third model for insurance coverage 
for research injuries is collective or pooled insurance.  Under this 
system, a group of research institutions could band together, perhaps 
with government assistance, to establish a collective insurance pool.  
Such an arrangement would retain many of the advantages of 
self-insurance, notably including flexibility in the terms of coverage 
and the ability to move forward with a program even in the absence of 
interest by the private insurance industry in providing appropriate 
coverage.  But the collective approach would permit greater spreading 
of risks among institutions than would institutional self-insurance.  
But the collective approach would permit greater spreading of risk 
among institutions than would institutional self-insurance, and could 
be expected to result in lower administrative cost.  Collective 
insurance pools might be especially important for smaller research 
institutions which face difficulties in securing coverage for research 
injuries in the private insurance market or in establishing their own 
self-insurance programs.

The Commission is not aware of any collective insurance pools for 
research injuries now operating in the United States.  However, may 
such pools have been established for other




125



		     FOR REFERENCE SEE (16bb13.gif)









for institutions wishing to do so, insurance protection for subjects 
may now be provided, at government expense, without the need for 
changes in Federal statutes or regulations. The logic of the 
Commission's reasoning would point toward regarding the costs of 
compensation program as direct costs like other identifiable items 
(including appropriate remuneration of subjects for their time and 
trouble in participating).

Having determined that a nonfault insurance system should be the basis 
of the "compensation program" being tested, the Commission proposes a 
number of variables that should be considered in designing the 
experimental condensation programs.  Before discussing those variable, 
however, the Commission's intent with respect to certain basic 
elements that the experimental compensation program should embody 
should be made clear.

Objectives of a Compensation Program

From what has already been said it can be seen that a major objective 
of any "compensation plan" is to ensue that necessary medical care and 
financial indemnities are provided in a fair and just fashion, to 
protect research subjects (and others who are dependent upon them for 
support) from the remediable harm that occurs as a result of injuries 
suffered in research.  "Easing the plight of the victim"[
5may be charged as a direct cost if consistent with institutional 
policy.  Such premiums may include those hazard, malpractice, and 
other liability insurance to cover grant-supported personnel and 
activities.

See also the profusion governing recovery of allowable insurance costs 
at 45 CFR Part 74.  Subpart Q. Appendix E (hospitals): and OMB 
Circulars A-87 (state and local government): A-21 (educational 
institutions): and A-122 (other nonprofit organizations).
The idea of a compensation fund is in no sense original.  It has, 
however, usually been suggested as a device for easing the plight of 
the victim.  I am not here suggesting it for that purpose (worthy 
though it might be), but rather as way of introducing an additional 
control element over when a medical experiment is considered 
worthwhile.

Guide Calabresi. Reflections on Medical Experimentation in Humans in 
EXPERIMENTATION WITH HUMAN SUBJECTS (ed. by Paul Freund), George 
Braziller, New York (1970) at 196, n. 2.

]5  is not, however, the sole consequence--an need not be the sole 
objective--of a compensation program.  A compensation program might 
also serve as means of social control over the funding and conduct of 
biomedical and behavioral research.

First, the funds spent in compensating injured subjects could help to 
identify, and perhaps even to quantify with some precision, a cost of 
doing research that had previously been overlooked in decisions 
regarding the number or type of experiments to undertake.  
Specifically, it has been suggested




that a compensation program could discourage overly risky research
[7 Clark C. Havighurst, Mechanism for compensating Persons Injured in 
Human Experimentation, in TASK FORCE REPORT, supra note 3.  Appendix A 
at 81.83.]7  and encourage economically "efficient" allocation of 
research funds.[8 Richard Zeckhauser. An Analysis of Mechanisms for 
the Compensation of Injured Research Subjects, in TASK FORCE REPORT, 
supra note 3, Appendix A at 155, 1256.]8   To the extent that specific 
funding decision take account of the injures associated with the 
research, an appropriate incentive is provided to researchers to 
exercise additional care in their work.[9 See Havighirst, supra note 27, 
and Zeckhauser, supra note 28.]9 

As laudable as such objectives may be, the difficulties in 
risk-estimation discussed in Chapter Four suggest that s system 
capable of producing actuarial estimates of risk exact enough to make 
superior decisions about allocation of research funds is, at best, an 
elusive idea.  Even were such estimates feasible, they would involve 
an expenditure of time and resources far in excess of that needed to 
establish a fair system of compensation--and probably far in excess of 
any benefits that might be gained from "wiser" decisions about the 
amount and kind of research that ought to be undertaken.  
Consequently, the Commission concluded that reliance in achieving 
control over research is better placed on other means )such as prior 
review by an IRB) rather than on the compensation system.

Within its own sphere, any program for providing compensation ought to 
be fair and efficient.  That is, it should (a) treat like cases alike, 
(b) involve fair payment for the harm sought to be remedied, and (c) 
disburse payments with maximum efficiency and minimum administrative 
cost.

Finally, the existence of a compensation program may have indirect 
consequences for the research enterprise or for society generally.  By 
indicating society's concerns for individuals serving the interests of 
the collectively, a compensation program not only avoids a potentially 
unattractive societal image for research but may also permit research 
that is otherwise seen as "too risky" (because of the possible 
consequences for its subjects) to go forward.  Indeed, the existence 
of compensation may make people more willing to serve in research 
projects generally.  The experience gained through experimental 
compensation programs for research injuries may also provide 
instructive information for other forms of Federally sponsored 
activities.[0 See Zeckhauser, supra note 28; see also, Calabresi, 
supra note 28.]0 

Any compensation system necessarily entails some balancing of 
competing objectives.  A design feature that advances one objective 
(e. g. providing comprehensive compensation) may undercut other 
objectives (e.g., promotion of the most





cist-effective research).  The ultimate choice of objectives will be 
for the policymakers who choose among (or modify) the option set forth 
below.  As noted earlier, the Commission, in considering compensation 
programs, has had in mind solely the objectives internal to the 
compensation system (i.e., fair and efficient provision of relief to 
the victims of research in a manner that can be administered easily 
and at a cost that imposes the minimum burden on the research sector), 
rather than the broader goal of enhancing the regulatory process.

Basic Elements of a Compensation Program

(1) Definition of "compensation."  The term "compensation for research 
injuries" has been variously defined, often leading to 
misunderstanding.  As employed here, the term encompasses two aspects:

(a) assurance that necessary medical care and related services are 
provided to subjects injured in research, and (b) financial 
indemnification of an injured subject, or the subject's family, for 
economic losses sustained (including both out-of-pocket cost and lost 
wages) as a direct consequence of research injury.

A recurrent source of confusion as to the intended scope of a 
compensation program results from the fact that medical care for 
subjects injured in research is often provided directly by the 
investigator or by the research institution, without charge to the 
subject and often without any formalized accounting for the cost of 
the services rendered.  As a conceptual matter, "compensation" 
encompasses the provision of such care, whether or not the financial 
value of the care is ever billed to the subject or even calculated as 
a bookkeeping entry.  Sometimes, medical care for an injured subject 
may be provided for a fee by care-givers entirely unrelated to the 
research or the research institution, or may otherwise be accomplished 
by a bill for services rendered.  Here, of course, "compensation"  
payment of the bill as well as provision of the care itself.

In both case, what is ethically relevant is that necessary care is 
provided in that the costs of that care do not fall upon the injured 
subject.  The questions of where the costs of the care ultimately fall 
(other than upon the injured subject), and how those cost are 
accounted for,, are financially important and must be considered in 
the design of the financing side of any compensation program.  
Experience gained through the operation of the compensation experiment 
will provide valuable clues as to the most efficacious financing 
structure.  In the regard, however, the Commission recognizes the 
desirability of permitting flexibility as to how medical care is 
provided and financially accounted for, particularly in cases of 
trivial injuries or other injuries not requiring long-term medical 
care.  Where "flexibility" should not be permitted is in the basic





requirement that necessary care be provided in appropriate 
circumstances, and that injured subject not be financially burdened by 
the cost of the care.

(2) Research covered.  Coverage should be limited to biomedical 
research and specified behavioral research (i.e.,only such 
behavioral research as is affirmatively determined in advance to 
impose more than minimal risk of bodily injury to research subjects) 
which is reviewed and approved in accordance with Federal regulations 
for the protection of human subjects.

(3) Nature of Injuries.  Benefits should be provide on nonfault basis 
to subjects sustaining nontrivial bodily injuries or death as a result 
of their participation in covered research.

(4) Benefits. Benefits should include short-term, emergency medical 
care, longer-term medical care (including rehabilitation) and related 
services, death benefits, and economic costs.  (More limited benefits 
might be available to seriously ill subjects participating in covered 
research).

(5) Other payments.  Benefits should be provided only to the extent 
that they are not paid by other source (e.g., heath or disability 
insurance, liability insurance, social security, court judgments or 
private settlements).

(6) Unreasonable conduct.  Benefits should be reduce or eliminated 
when injury result form the failure of the subject reasonably to 
conform to the provisions of the research (e.g., intentional 
falsification of medical history, failure to comply with 
investigator's directions on avoiding alcohol, drugs, etc.).

(7) Fact-finding, The use of expert fact-finding panels and 
arbitration for the resolution of disputes should be encouraged.

(8) Funding.  The experimental compensation program should be funded 
by the Federal government.  Reasonable costs of private, group, or 
self-insurance incurred by non-Federal research institutions should be 
recognized as allowable direct cost under Federal grants and 
contracts.


Variables to be Included in the Experimental Design

The Commission believes that those with final responsibility for 
designing and implementing the compensation experiment should have 
freedom to establish the configuration of the compensation plans that 
will be evaluated.  The following discussion of the variable to be 
considered is offered as an aid to those making the decisions.

Scope of Covered Research

Government involvement.  The ethical arguments favoring a compensation 
program for research injuries are strongest where the nexus between 
the research and the Federal government is closet.  That is, the 
greater the role of the Federal government in the design, funding and 
conduct of the 




research, and in encouragement of subjects to participate, the greater 
the obligation of the Federal government to provide, or to assure the 
provision of medical care and financial assistance to subject injured 
in the research.  Thus, a program might cover any or        following: 
(I) research conducted intramutually by Federal agencies; (ii) 
research supported by Federal grants and contract: and (iii) research 
submitted to Federal agencies in fulfillment of regulatory 
requirements (e.g., for licensing of new drugs, medical devices, 
pesticides, etc.).

With respect to research supported by Federal agencies, several 
sponsoring agencies (including the Public Health Service) currently 
permit research institutions to include reasonable expenditures for 
insurance procured to protect subjects of research as part of the 
costs of grants and contracts,[1 Public health Service Grants Policy ]
Statement at 19.]1  the practice is not widespread, however.

The creation on an experimental basis of compensation programs for 
Federally funded research should encourage the sponsors of private 
research submitted to FDA and other regulatory agencies to establish 
parallel nonfault compensation plans and to share their data with the 
agency overseeing the Federal experimental program,  Indeed, some 
private sponsor of research have already implemented such plans,  
Until the results of other experimental evaluation are available, it 
is no possible to evaluate the desirability of requiring all sponsors 
of research that is privately funded by Federally regulated to provide 
non-fault compensation.

Finally, it have argued that since the ethical considerations 
underlying compensation do not vary according to the source of the 
funds to conduct the research, any compensation program should apply 
to all institutional research, not just to Federally funded research.[
2 John R. Petit, Chair, University Risk Management and Insurance 
Association (URMIA) Human Subjects committee, letter to Alan J. 
Weisbard (August 20, 1981): see Appendix R to this Report.
]2    In addition to a question of simple equity among subjects, there 
are practical difficulties I having a compensation program that only 
applies to some studies even though the risk to the subject may be the 
same.[3 Id.]3   Although this position is highly persuasive, it does 
not follow, that "the cost of any such compensation program should be 
fully borne by the Federal government."[4 Id.]4   Intra-institutional 
equity could be achieved by voluntary action of each institution or by 
a governmental mandate.  But such a requirement would at the moment 
be out of keep with the entire theory of the data-gather experiment.  
Nevertheless, the Federal government might offer to cover the costs of 
compensation in non-Federally as well as Federally supported research 
as in inducement to institutions to participate in the experiment.



Type of research and IRB review.  Under well-established regulation,       
for the Department of Health and Human Services at 45 CFR Part 48, all 
Federally conducted or supported research with human subjects must 
undergo prior review.  For most of the studies that could produce 
claim for compensation, the process required includes prior review and 
approved by a properly constituted institutional Review Board (IRB) 
established by the body conducting the research.

The requirement of proper institutional review would exclude from 
coverage research which has not been fully screened and which may pose 
special risks of injury.  Such a requirement may be necessary to 
safeguard the financial integrity of the compensation fund and to 
encourage investigators to cooperate with the IRB process.  
Investigators who improperly short-circuit the review process may be 
label to injured subjects in the courts under the legal theory of 
negligence per se; in some situations, however, potential defendants 
may escape liability under the defenses of sovereign or charitable 
immunity).

Under recent revisions to the regulations on IRBs, certain research, 
in which risk is believed to be minimal nonexistent, need not undergo 
full review by the IRB.  Some such research, primarily that involving 
survey, questionnaires, and educational tests or the study of existing 
pathological and diagnostic specimens, poses not risk of bodily injury 
and is formally "exempt" from IRB review.  Such research might well be 
left outside the scope of the compensation plans to be tested, since 
the primary concern is with redressing the harms that accompany bodily 
injury.

Other research, believed to pose only minimal risks of harm, is 
permitted by the new regulations to be reviewed and approved on an 
expedited basis.  Biomedical research falling within this "expedited 
review" category may result in bodily injury to research subjects 
(e.g., a chipped tooth suffered by a subject who faints and falls 
following a "blood draw").  therefore, bodily injuries resulting from 
biomedical research subject to expedited review probably should be 
included within the coverage of the experimental compensation plan.

A large proportion of behavioral research poses not risk whatever to 
subjects.  Further, many though not all the risks posed by behavioral 
research are risk of purely psychological or social injury.  Inclusion 
of all behavioral research in a compensation program would impose 
economic costs and administrative burdens unwarranted by any possible 
benefits.

It is undeniable, however, that some research characterized as 
"behavioral" poses risks of bodily injury to research subjects.  Such 
risks may blow from physical interventions similar to those employed 
in biomedical research, from subjecting subjects to unusual or 
expected stress, or from other causes.  Bodily injuries resulting from 
such research should be





compensable, both as a matter of fairness and to avoid imposing 
administratively difficult requirements for line drawing between 
biomedical research and physically risky behavioral research.  the 
regulatory distinction between behavioral research eligible for 
expedited review and that requiring full IRB review could form the 
basis for distinguishing between behavioral research that is included 
in the compensation program and that which is not.

The problem of therapeutic research.  The question of whether 
"therapeutic research" should be excluded from a compensation program 
for research injuries has been among the most vexing and continuous 
arising in the course of the Commission's study of compensation for 
research injuries.[5 A closely related question involves the 
applicability of a compensation plan for research injuries incurred 
in the development of vaccines and in immunization program.  
Compensation for vaccine-related injuries has received extensive 
attention elsewhere, notably in reports issued by the Office of 
Technology Assessment.  See A REVIEW OF SELECTED FEDERAL VACCINE AND 
IMMUNIZATION POLICIES supra note 15 and COMPENSATION FOR VACCINE-RELATED 
INJURIES, supra not 15.  While vaccine injuries share certain 
characteristics with research injuries in that they are often 
incurred in governmentally encouraged (or sometimes required) 
activities on behalf of the public goo, vaccine, virtually by 
definition, offer some promise of preventive health benefits to the 
subjects on whom they are tested, vaccine trial would be classified 
as "therapeutic" and thus excluded from the scope of compensable 
injuries under Plans I and III as well as under Plan II, unless the 
injures were caused by additional research procedures 
classified as "nonbeneficial".  It is recognized, however, that 
subjects in vaccine reserve are typically "normal volunteers," a 
status that might justify classifying the research as 
"nontherapeutic".  The Commission concludes on balance that all 
vaccine-related injuries, both from the research and development phase 
and from routine administration, would probably best be handled by a 
single compensation system specifically addressed to vaccine injuries.
]5   The view of all research, whether "therapeutic" or 
"nontherapeutic" must be covered by a compensation program was not 
found to be persuasive.  There is much to be said as an analytical 
matter, of course, for the conclusion of the Secretary's Task Force 
that "the distinction between therapeutic and nontherapeutic 
research...is not a valid criterion on which to make a determination 
as to which injured subjects are to be compensated.[6 TASK FORCE REPORT, 
supra not 3, at VI-9.]6   But, as the Task Force acknowledged, it 
would be inordinately expensive to "compensate" all patient-subjects 
for all illness or injury they experience after participating in 
research.[7 A distinction between "normal" and "sick" subjects is 
also sometime advanced as a means of limiting the costs of a 
compensation program.  This basis for excluding therapeutic research 
from coverage under a compensation system--that therapeutic research 
involves ]7   Even setting aside the difficulties of




separating injuries that were caused by the research from those that 
were not , measuring the extent of the injury is conceptually as well 
as practically extremely difficult.  Many patients participate in 
highly risky experiments precisely because their condition is already 
so desperate that they have, in common pariance, "nothing left to 
lose" and they hope that something new will help them.  The Task 
Force's proposed means of dealing with this difficulty, the so-called 
"on-balance test," was found by the Commission to pose enormous 
burdens of administration which could not be justified in light of the 
participate in research with the hope of deriving personal health 
benefits.

Furthermore, providing compensation to subjects of therapeutic 
research on any more favorable basis that that applicable to injuries 
resulting form ordinary medical practice (including innovative 
therapy) accords research injuries preferential treatment.  Fro reason 
discussed more fully below, such preference might not ethically 
inappropriate.  But it would generate a serious question in need of an 
empirical answer: Would inclusion of therapeutic research within a 
program's coverage create unfortunate (and expensive) incentives for 
patients to enroll in research protocols for their physician to 
include them to make the eligible to have any "adverse effects" paid 
for, as they would not be if the patient's condition worsened during 
ordinary medical care?

Many researchers and commentators fell a moral debt toward 
patient-subjects who are injured in therapeutic research.  Such 
research is intended, at least in part, to benefit society.





and to advance knowledge.  And the subjects are seldom in a position 
to insist on receiving the experimental treatment outside a research 
protocol.  For reasons of science, it is usually appropriate (and 
sometimes required under Federal regulations) to confine the procedure 
being tested to persons enrolled in a formal experiment.  Moreover, 
the number of care-givers able to employ any significant new procedure 
will typically be small, and they will probably all be providing it in 
research settings, either individually or in collaborative projects.  
Thus though no evil or self-serving motivation is at work, a classic 
contract of adhesion may nonetheless result.  For the patient-subject 
the choice to enroll in the experiment is a choice 'a prendre ou a 
lasisser.

Some patient-subjects may thus be made to contribute to the general 
welfare in a way the otherwise might not.  Yet society is not 
necessarily under a strong obligation to compensate those who are 
injured.  This particularly true if the "contribution" required of 
patient-subjects does not in fact impose any additional burden on 
them.  The Commission found no conclusive data either to support to 
contradict a conclusion that, when viewed as a whole on a 
retrospective basis, therapeutic research produces worse results than 
those experience by comparable patients in stand (0r innovation) 
therapy in settings not formally designate as "research".

Patients enrolled in experiments, however, are sometimes exposed to 
additional procedures that are undertaken for reasons of the research 
design rather than strictly for the benefits they provide.[
8 For example, in some therapeutic research. fluid samples are taken 
more frequently or in greater volume than necessary for proper patient 
care, in order to gather additional data or to better document the 
research.  Other times, an additional procedure may be performed 
(e.g., a spinal tap) to learn more about the metabolic affects of a 
drug under study.  If information derived from examination of the 
spinal fluid, while useful for a research objective, is neither 
necessary for, nor related to, the care of the patient, then the tap 
would be a "nonbeneficial" procedure.]8   As regards these 
"nonbeneficial" procedures, patient-subjects in therapeutic research 
are in the same position as are other subjects enrolled in experiments 
not intended to be therapeutic.  Thus, it may be worthwhile to 
evaluate the practicability of permitting recovery for a limited 
class of injuries associated with therapeutic research: those arising 
from research that employs certain procedures that are "nonbeneficial" 
from the subject's viewpoint.

All research projects coming within a compensation program should 
undergo prior review.  As part of this review, under existing 
regulations (see e.g., 45 CFR 46.116(A)(6) & (7), the IRB or other 
responsible officials must assure that approved consent forms disclose 
whether compensation will be provided in the event of injury.  
Accordingly, an IRB will




have to determine whether or not research is "therapeutic": if it is, 
the fact that compensation is unavailable would have to be disclosed 
as part of the current process.  The HHS officials responsible for the 
proposed social policy experiment may, however, also wish to explore 
whether any harm caused by "nonbeneficial" research procedures can be 
separated from that caused by the research intervention as a whole and 
whether compensating for such harm is feasible and desirable.  If so, 
the IREs at a few institutions will need to undertake the additional 
task of determining whether each therapeutic research protocol 
involves may "nonbeneficial" procedures (but not injures resulting 
from their underlying disease for from the therapy being tested).  At 
a like number of institutions matched for relevant characteristics, 
the IRBs would perform the same analysis of the protocols, but 
compensation would not be promised or paid.  Instead, the people 
conducting the social policy experiment would simply monitor the 
outcome of the therapeutic research to see how many injuries are 
reported to have been caused by procedures identified in advance by 
the IRB as "nonbeneficial"  In the fashion,  it should be possible not 
only to test the feasibility of identifying "nonbeneficial" procedures 
in protocols and of separating injuries caused by such procedures from 
those with other causes, but also to have some idea whether the 
existence of a compensation program at an institution ends to a higher 
rate or reported injures from nonbeneficial procedures.  This issue 
can be further examined when the people conducting the social policy 
experiment examine the records in each case and review the basis on 
which a researcher determined that an injury was caused by a 
"nonbeneficial" procedure in therapeutic research.

Eligibility for Benefits.  The eligibility criteria must specify which 
persons may receive benefits, for what types of injury, and on the 
basis of what proof.

Subject of covered research.  Benefits could be limited to person 
injured in their capacity as research subjects or could be extended to 
investigators or to certain third parties injured as a result of 
research activities.  In his testimony before the Commission, Dr. John 
Arnold of the Quincy Research Center suggested a number of 
possibilities of third party  injury, notably including motor vehicle 
accidents caused by research subjects unfit to drive, or injuries 
resulting from aggressive behavior by subjects under the influence of 
investigational drugs.  The Commission suggest that existing legal 
remedies provide appropriate resource for third parties injured in 
such circumstances, and that extension of nonfault liability for 
injuries of this sort under the compensation system would not be 
advisable.




In this respect, the Commission departs from the recommendation of the 
Secretary's Task Force that injured third parties should be 
compensated where "it is established by a preponderance of evidence 
that the injury is a direct consequence of the research"[9  TASK FORCE 
REPORT, supra note 3, at V!-12.]9   (Emphasis in original).  The Task 
Force did not elaborate on what is meant by "direct", and 
representatives of the insurance industry have expressed considerable 
skepticism as to the practicability of this distinction.  Furthermore 
there is serious doubt as to the availability of private insurance to 
cover third party injuries to third parties resulting for research--even 
indirectly--may be regarded as part of the true social cost of the 
research enterprise, the Commission believes that such injuries may be 
excluded from a nonfault compensation system as tangential to the main 
concerns of the system and as better handled by other means.

One remaining issue is the proper characterization of fetuses in utero 
of women who participate in research during the course of their 
pregnancies.  Existing legal mechanisms may or may not provide 
appropriate relief in such situations.  If a fetus whose mother is a 
research subject is directly exposed to bodily injury by actions of 
the investigators, such a fetus could be made eligible for 
compensation in the event that it is born with an injury indisputably 
caused by the research.  It should be noted that injuries to fetuses, 
like injuries to other research subjects, may be affected by 
provisions governing compensation for latent injuries (SEE BELOW).

Nontrivial bodily injuries.  The Secretary's Task Force recommended in 
1977 that compensation should be provided for "physical, 
psychological, or social injury" resulting form research.  The 
"interim final regulation" published by the Department on November 3, 
1978, required disclosure to subjects, as part of the informed consent 
process, of the availability of "compensation and medical 
treatment...if physical injury occurs."[The recently promulgated HHS 
regulations governing protection of human research subjects eliminate 
this limitation to "physical" injury, on the basis that "the 
Department sees no reason t limit such disclosure to only one kind of 
injury.[1 46 FEDERAL REGISTER 8383 (January 28, 1981).]1   The question 
of whether compensation should be provided for psychological or "social" 
injury (however defined) has been a controversial one, and has 
received careful consideration by the Commission.

Although the courts have increasingly recognized claims for emotional 
or psychological damages in defined circumstances (e.g., a mother's 
claim of emotional anguish on seeing her child injured or killed in a 
motor accident, even in




the absence of actual or threatened physical injury to the mother 
herself), many commentators have expressed concern measuring its 
degree, and computing appropriate monetary damages.[2 See Havighurst, 
supra note 27, at 89-90; see generally, supra notes 7 & 8 in Chapter 2 
infra.]2   All of these concerns have been expressed in letters or 
presentations to the Commission from representatives of the insurance 
industry, university risk managers, the Association of American 
Medical Colleges, researchers and other independent experts.
[3 American Insurance Association.  State Before the President's 
Commission, see Appendix Q to this Report: University Risk Management 
and Insurance Association, Position Paper Regarding a Compensation 
Program for Adverse Effects of Human Subjects Research (1980), see 
Appendix R-1 to this Report at 10: Letter from John $ Sherman, Ph.D. 
(Vice President, Association of American Medical Colleges) to Morris 
Abram (May 7, 1980) see appendix T-13 to this Report at 6; John D. 
Arnold, Incidence of Injury During Clinical Pharmacology research and 
Indemnification of Inured Research Subjects at the Quincy Research 
Center, see Appendix I to this Report at 19-27; Letter from Philippe 
V. Cardon, M.D. to Barbara Mishkin (July 3, 1980). see Appendix T-2 to 
this Report.]3   

Similar objections have been raised regarding the provision of 
compensation for "social injuries,: a term left undefined by the Task 
Force which has resulted in considerable confusion and 
misunderstanding.  The Commission understands the term "social injury" 
to refer to injuries to reputation, personal relationship, or legal 
status resulting from the disclosure of identifiable personal 
information gathered in the course of research.  The Commission 
believes that the most serious problem likely to arise in this are 
(e.g. unauthorized disclosure of sensitive private information) are 
better handled by a combination of careful prior review of the design 
of research projects posing such risks and where necessary, pursuit of 
legal remedies through the courts.

Nevertheless, as revealed in testimony before the Commission, the 
insurance industry appears willing to provide coverage for "social and 
emotional injuries" having "some relationship to an actual physical 
injury" and "directly related" to participation in research.[
4 Testimony of Dennis R. Connoly, James D. Morrow and John R. {ettit, 
transcript of 4th meeting of the Presidents Commission (September 15, 
1980) at 150-38.]4   It may eventually prove possible, in addition, to 
provide compensation for objectively measurable damages, including lost
wages and out-of-pocket cost, associated with verifiable manifestations 
of psychological injuries.  The Commission suggests that the 
feasibility of such coverage be among the variable tested in the 
compensation experiment.

With respect to trivial bodily injuries, which constitute a large 
proportion of the adverse effects associated with partici







			FOR REFERENCE SEE (16bb14.gif)








notably successful in discerning the etiology of disease, and 
administrative decisions often result in appeals and protracted 
litigation.  These difficulties are probably inescapable given the 
current state of the art of medical science.

Some compensation systems have sought to avoid these difficulties by 
creating legal presumption (e.g., designating specifications injuries 
for which causation need not be proved) or by the simple (but 
sometimes expensive) expedient of providing compensation for all 
injuries that occur in the course of specified activities.  Such 
systems exchange a measure of certainty   to causation for more 
streamlined and less costly means of resolving claims.  Whether this 
"tradeoff" is desirable depends on the characteristics of the problem 
the compensation system is designed to address.

With respect to nontherapuetic research, the Commission believes that 
problems of causation can be adequately managed through the familiar 
legal standard of proximate clause, coupled with a mild presumption of 
causation to the research sponsor.

More difficult questions of causation are likely to arise in the 
provision of compensation in the event of injuries resulting from 
nonbeneficial procedures employed in therapeutic research.  The 
Commission encourages the use of expert review bodies and 
experimentation with different presumptions and standards and burdens 
of proof for the resolution of difficult questions of causation, and, 
closely related, or the degree of "excess injury."[5 To illustrate our 
possibility: If a subject can demonstrate a 
"reasonable likelihood" that his injury resulted from a particular 
research procedure, a presumption of causation could be created.  This 
presumption could then be overcome by clear and convincing scientific 
evidence that the injury resulted from causes extrinsic to the 
research.]5 

Standards of Conduct.  The extent to which a subject's own improper 
conduct should exclude that subject from receiving compensation in the 
event of injury attributable, at least in part, to that conduct must 
be determined.

Two types of misconduct require special mention in this regard.  First 
is behavior by a subject which is knowingly and deliberately 
self-destructive.  Injuries resulting from willful intention to injure 
oneself or others are frequently excluded from coverage under workers' 
compensation and similar programs, and the Commission believes that 
such injures should be excluded here.

Conduct which might be characterized as reckless or negligent, though 
short of willful self-destruction, poses a more difficult problem.  
Subjects may fail, for example, to provide complete and truthful 
responses to questions concerning their






medical history.  Similarly, subjects may fail to heed an 
investigator's directions with respect to diet, ingestion of alcohol 
or drugs, or other limitations on behavior during the course of 
research.  Such failure may be either deliberate or inadvertent, and 
in many cases are likely to proceed from an incomplete awareness of 
the gravity of possible consequences.  the Commission believes that 
such behavior, not amounting to willful self-destruction, should 
reduce,  but not necessarily eliminate (except in extreme cases) a 
subject's entitlement to compensation.

Time for Making a Claim.  the problem of latent effects--adverse 
effects which are not manifested as injuries for a substantial period 
of years following exposure to the causative agent--will pose enormous 
difficulties in structuring a program of private insurance to provide 
compensation for research injuries.  For a variety of technical 
reasons related to the accounting and financing reserve procedures 
utilized by the insurance industry, the insurers will almost certainly 
insist on a so-called "outside limitation" on their period of 
potential liability.  This time limitation would be measured from the 
date of a subject's last participation in the research procedure, and 
would impose an absolute cutoff on claims brought after a fixed 
period.  While the insurers have been willing to discuss an outside 
limitation period as long as ten to fifteen years, such a limitation 
would nevertheless preclude the award of compensation (through a 
private insurance mechanism) for injuries resulting from exposure to 
carcinogenic or other agents with latency periods measured in decades.  
The insurers candidly acknowledge that under such a provision, "there 
may be instances where persons may be viewed as worthy of receiving 
compensation.  Nevertheless, they will not receive it".
[6 Connolly, Morrow, and Pettit, supra note 44.]6 

The resistance to providing relief for latent effects is not 
universal.  In particular, several Federal workers' compensation 
programs impose no outside limitation period, and require only that a 
claim be brought in timely fashion after the injury has manifested 
itself and the claimant can reasonably be expected to be aware of the 
causal relationship between the injury and his employemnt.
[7 Section 8122 of F.E.C.A. provides, in pertinent part, that:
(a) An original claim for compensation for disability or death must be 
filed within 3 years after the injury or death....
(b) In a case of latent disability, the time for filing a claim does 
not begin to run until the (subject) has a compensable disability and 
is aware, or by the exercise of reasonable diligence should have been 
aware, of the causal relationship of the compensable disability to his 
(participation in covered research).  In such a case, the time for 
giving notice of injury begins to run when the (subject) is aware, or 
by the exercise of reasonable diligence]7 




The Commission disconcerns no ethical rationale for setting any 
particular cutoff for victims of research-related latent injuries from 
a compensation program, but does recognize the practical problems 
involved in providing coverage for such injuries.

The Commission noted in addition that if any compensation plan is 
tested during an experimental period and later terminated, 
arrangements must be made either to inform potential research 
subjects, prior to their participation, that compensation for latent 
injuries will not be available after a specified period of time, or to 
establish a fund to cover liabilities or injuries that arise later.

Nature and Extent of Benefits.  The primary component of 
"compensation" in the vast majority of cases of research injury will 
be the provision of medical care required as a result of the injury.  
The immediate concern is that necessary medical care be available to 
the injured subject: the second is that the financial burden of such 
care to fall upon the injured research subjects.[should have been 
aware, that his condition is causally related to his (participation 
in covered research), whether or not there is a compensable disability....
(d) The time limitations in subsections (a) and (b) of this section do 
not--
(1) begin to run against a minor until he reaches 21 years of age or 
has had a legal representative appointed: or
(2) run against an incompetent individual while he is incompetent and 
has no dully appointed legal representative
(3) run against any individual whose failure to comply is excused by 
the Secretary on the ground that such notice could not be given 
because of exceptional circumstances.
40 For example, F.E.C.A. provides:
&103(as): The United States shall furnish to (a subject) who is 
injured in the (course of research), the services, appliances , and 
supplies prescribed or recommended by a qualified physician, which the 
Secretary of Labor considers likely to cure, give relief, reduce the 
degree or the period of disability, or aid in lessening the amount of 
the monthly compensation.  These services, appliances, and supplies 
shall be furnished--
(1) whether or not disability has arisen:,
(3) by or on the order of United States medical officers and 
hospitals, or, at the employee's option, by or on order of physicians 
and hospitals designated or approved by the Secretary.
The employee may initially select a physician to provide medical 
services, appliances, and supplies, in accordance with such 
regulations and instructions as the Secretary condones necessary, and 
may be furnished necessary and reasonable transportation expense 
incident to the securing of such services, appliances, end supplies...
S103(b) The Secretary, under such limitations or conditions as he 
considers necessary, may authorize the employing agencies


to provide for the initial furnishing of medical and other benefits 
under this action...
S106(a): The Secretary of Labor may direct a permanently disabled 
individual whose disability is compensable under this subchapter to 
undergo vocational rehabilitation.  The Secretary shall provide for 
furnishing the vocational rehabilitation services...]0 






The Commission heard testimony that short-term medical care is 
routinely provided by investigators or research institutions in the 
event of injury to research subjects, usually without charge to the 
subject.  Yet, apart from information provided by the University of 
Washington and the Quincy Research Center, both of which have formal 
compensation programs, no systematic empirical evidence to support 
that claim was advance, and the Commission's efforts to secure such 
evidence have been unsuccessful.  Furthermore, examination of consent 
forms gathered from a number of research institutions suggests that 
the availability of free medical care in the event of injury, if that 
does constitute institutional policy, is not always brought to the 
attention of research subjects.  Nevertheless, there appears to be no 
empirical basis for rejecting the claim that mush short-tern emergency 
care, and some related services, is provided in the event of 
research-related injuries although not announced to subject s in a
The situation with respect to longer-term medical care and related 
social and behavior services, as well as long-term medical 
surveillance, is more difficult but also more clear. Available 
evidence suggest that few research institutions have made provisions 
for supplying or financing such care.  There is little basis for 
confidence that such care is now generally available, without cost to 
the subject, in the event of serious injury.

There appears to be little opposition to the notion that a 
compensation system, if one is to exist, should assure the provision 
of necessary medical care and closely allied services, both 
short-and-long term, without cost to a research subject with more than 
trivial injuries.  (The experimental period should allow exploration 
of the related question: ought such medical care be paid for by the 
"compensation fund" if it is minor or not expensive?  This is a 
question more of bookkeeping and institutional efficiency rather than 
one ethics.  It services a reminder, however, that the experiment is 
intended to provide information of practical as well as ethical 
import.)

Those placing the compensation experiment must also determine the 
extent to which death benefits should be provided to a subject's 
survivors.  How, it at all, should such benefits be limited in cases 
of preexisting condition--particularly a terminal illness?  Death 
benefits might be provided





either according to a periodic schedule or as a lump-sum payment.

A certain number of research-related injuries result in functions 
impairments disabling subjects from participating in some or all of 
the activities of everyday life, whether permanently or for some 
period of time.  Sometimes this entails loss time from paid 
employment, possibly resulting in loss of wages.   Disability may also 
entail out-of-pocket costs for housekeeping service, child care, and 
other activities normally performed by the injured research subject.  
In other instances, disability may entail losses which are very real, 
but are harder to measure in monetary terms, e.g.,., loss of time from 
school, or worsened conditions for already hospitalized patients.  
There appears to be little objection, in principle, to the provision 
of some level of financial compensation for actual loss of wages and 
out-of-pocket costs directly resulting from research-induced injuries.

While some provision of compensation for financial losses 
straightforward in most compensation programs, for example, the 
injured worker is paid a fixed percentage of lost wages, perhaps with 
some adjustment if the are a dependent spouse and children.  In 
contrast, many subjects of biomedical and behavioral research 
computation.  Research subjects include, among others, person with no 
current earrings (institutionalized person, the unemployed, the 
elderly), and person whose current wages do not reflect their 
potential earning capacities (medical students, interns).  The small 
sums provided to some research subjects are remuneration for their 
time and inconvenience and are unlikely to provide a satisfactory 
basis for calculating compensation levels in the even to disabling 
injury.  Thus, an alternative approach must b identified if benefits 
are to provided in such cases.

Among the several possible approaches outlined for the Commission, one 
borrows the "average man" concept employed by the Veterans 
Administration, which determines benefit levels on the basis of a 
percentage (reflecting the degree of disability) of the earnings of 
the average worker in the nation, rather than on the basis of the 
earning history or potential of the particular person injured.[
1 Jones, supra note 4.]1  A distinct but




related approach is embodied in the Federal Employee's Compensation 
Act, which provides benefits to disabled volunteers in the Peace 
Corps, VISTA, the Job Corps, and similar programs at levels pegged to 
the Federal GS scale (GS-7 to GS 11 for Peace Corps, [0 Currently 
$15,922 - $23,586]0  GS2 for the Job Corps [1 Currently $9,381]1 ).  
Central to both these approaches is a determination that 
persons with similar disabilities should receive similar benefits.  In 
addition to the administrative simplicity of this approach, it carries 
an ethical appeal as well.

This approach is not without its drawbacks however, particularly with 
respect to incentives that might be created in the recruitment of 
research subjects.  A fixed scale of disability payments might tend to 
discourage potential subjects with higher current or potential 
earnings while inducing participation by the poor.  If this effect is 
substantial, it would violate the ethical norms of a just distribution 
of risks and benefits across the society.

On the other hand, concern has been expressed that the provision of 
"generous" financial benefits, particularly to person not currently in 
the labor force, would induce abusive claims and provide little 
incentive for the affected individuals to return to normal activities 
as quickly as feasible.  One possible remedy, endorsed by the insurer, 
the medical colleges, and the university risk manager, would be to 
impose waiting periods during which disability payments would not be 
made.  Differing waiting periods may be considered for current wage 
earners and for non-wage earners.

Finally, the Commission notes that serious problem with many 
compensation systems, particularly those providing relief for 
long-term disabilities, is that they include no provision for 
adjusting benefit levels in response to inflation.  As a reward 
become, over the years, completely inadequate to meet the needs of the 
disabled party.  This problem could be avoided if provision is made to 
index benefit levels to changes in the cost of living.

Limitations on Benefits

Pain and suffering.  The award of damages for "pain and suffering," in 
traditional tort litigation, is to restore the injured party, as 
nearly as possible, to the position he would have been in absent the 
wrongful injury.  In addition to repaying actual dollar losses, and 
attempt is made to compensate in dollars for intangible losses, 
including pain and suffering.  The calculation of damages for pain and 
suffering is recognized, however, as very exact, and some legal 
scholars argue that awards for pain and suffering constitute, in 
effect, a way for jurors to



"punish the wrongdoer: and to provide funds for the inured party to 
pay legal fees.

In nonfault systems, the tacit element of "punishing the wrongdoer" is 
reduced or eliminated, several barriers to recovery for the injured 
party are removed, and greater stress is place both on providing 
compensation on a manner more precisely tied to measurable losses and 
on controlling the overall costs of the system.  Accordingly, the 
Commission concludes that recovery for pain and suffering, which is 
generally excluded form recovery under nonfault systems, should also 
be excluded here.

For that class of research involving already hospitalized, critically 
or terminally ill patients, however, a compensation system limited to 
lost wages and out-of-pocket costs-that is, one which does not cover 
pain and suffering--may amount to an empty promise.  Patients in such 
research may offer true altruistic service to the society and may 
suffer previously from the toxic and painful effects of powerful 
drugs, The Swedish compensation programs provide an precedent for the 
award of scheduled benefits, tied to objective criteria, which might 
be followed in this regard.  The Commission suggests that such a 
feature might be tested during the compensation experiment.

Punitive damages.  Punitive damages, expressly reserved in tort law 
for conduct which is intentional, deliberate, and outrageous, are 
virtually always excluded from nonfault compensation systems.  Were 
such egregious conduct to occur in research context and result I 
injury, it is likely that the inured subject could pursue a legal 
remedy through the tort system.  There appears little need to encumber 
a nonfault compensation system for research injuries with a concept so 
rooted in the negligence system, and the Commission recommends that 
punitive damages be excluded.

Legal fees.  Historically, the introduction of nonfault compensation 
systems has often been accompanied by hopeful rhetoric about reducing 
or limiting the role of lawyers.  This has rarely, if ever, proved to 
be the case, and the Commission hesitates to suggest that this system 
will prove the exception.

In recent years, a number of environmental and civil rights laws have 
provided for the award of attorney' fees to the prevailing party in 
order to encourage attorneys to assist in the vindication of 
especially prized rights.  Such laws remain the exception rather than 
the rule, however, and few compensation systems now provide for the 
payment of attorney's fee to a prevailing claimant.  The Commission 
recommends that fees for attorneys and expert witnesses be excluded 
from coverage.

Ceiling on Benefits.  In addition to limitations on compensation built 
into the definitions of injury and of specific indemnities, the 
Commission believes that overall ceiling on monetary benefits should 
be imposed to control costs and preserve the fiscal integrity of the 
compensation program(s)




during the course of the compensation experiment.  For example, 
overall limitations of $100,000 and 410,000 per subject might be 
tried.  Were a compensation program eventually to be adopted, a 
Federal backup program to provide compensation in excess of this 
ceiling for claims that arise in Federally supported research might be 
advisable.

Offset for Recoveries from Collateral Sources.  While one important 
justification for the enactment of a nonfault compensation program for 
research injuries is the inadequacy of existing remedies for injured 
subjects are eligible for certain benefits from a variety of sources 
external to the compensations plan itself.

The traditional approach in tort litigation and in some government 
programs, including benefits under the Veterans Administration 
programs, is not to reduce benefits to offset recoveries from most 
other sources.  This policy avoids penalizing people for their 
foresight in securing health and life insurance.  Further, this 
approach is consistent with the theoretical argument that to the 
extent research (or any other activity) results in injuries, the costs 
of these injures are part of the underlying social costs of pursuing 
the activity.  On this view, such injury-related costs should be taken 
into account by those who make decisions about pursuing (or funding) 
the activity, whether or not the injured party also happens to have 
another source of payment for his or her injury.  This rationale is 
expounded by those who analyze the objective of a compensation system 
from an economic standpoint, and similar rationale have sometimes been 
discussed by courts in cases involving drug or vaccine inuries.
[2 See Calabresi, supra note 26; Havighust, supra note 27; Zeckhauser, 
supra note 28.]2   This approach would regard the compensation 
program as "primary" in the cost-bearing sense.

On the other hand, a provision allowing offsets for recoveries from at 
least some collateral sources might significantly reduce the budgetary 
costs of a compensation program, and is strongly urged by the insurer, 
the medical colleges, and sometimes but not all, participants in the 
university risk managers study.  Under this approach, the compensation 
system would be "secondary" to other sources.

The basic justification of a compensation system for research injuries 
is to avoid leaving unpredictable, unfair or unbearable economic 
burdens on the backs of those who have agreed to serve as subjects.  
To the extent that the costs created by research injuries have already 
been paid by the other sources, this critical function of a 
compensation system has been met.  Since the Commission does not 
believe that the compensation program would be a good way of 
"controlling" what research




gets undertaken, it does not regard it as necessary to attempt to 
quantify all costs to society of the "research enterprise."  In 
particular, the Commission is concerned that the costs of a 
compensation program not be permitted to escalate in a fashion that 
would endanger the integrity and continued existence of the program 
itself, much less endanger biomedical and behavioral research.  
Accordingly, the Commission suggest that benefits under the 
experimental program be "secondary" to other sources of benefits, so 
as to allow limited funds to be used where they are most needed.[
3 The Commission suggests that those conducting the compensation 
experiment not employ compensation programs that would specifically 
provide for subrogation of the compensation fund to any claim an 
injured subject may have against any person, other than the 
investigators or the research sponsor, nor provide for a retroactive 
adjustment in the amount of compensation after recovery from such a 
third party.  It would seem proper to protect the funds available for 
compensation payments if the subject's injury has been caused by a 
third party's wrongdoing: moreover, once a tort claim against a third 
party has been adjudicated in a subject's favor, the subject would be 
receiving "double recovery"," once from the fund and again from the 
tort action.  The experience with subrogation rights under other 
governmental compensation schemes, however, is that they are seldom 
invoked; the United States Attorneys to whom they are assigned do no 
make them matters of high priority because the amounts involved rarely 
justify the cost and other burdens of prosecuting the claims.  Under 
Plan I, the F.E.C.A. provisions (&8131 on subrogation and $8132 on 
adjustment after recovery from a third party) could be applied: a 
choice will have to be made under Plans II and III whether to provide 
for subrogation and other aspects of potential excessive  compensation 
(see, for example, &8219 of F.E.C.A., which authorized regulations to 
recover over-payments) or to allow experience with these issues to 
accumulate during the experimental period in order to determine if the 
cost savings justify the administrative burdens.]3 

Relating to Alternative Legal Remedies.  As previously noted, much of 
the impetus for a nonfault compensation system for research injuries 
derives from the inadequacies of existing legal remedies in providing 
fair and expeditious relief for such injuries.  Nonetheless, in some 
cases, an injured subject may perceive the possibility of a large 
recovery through the tort system, as well as smaller (and more 
certain) recovery through the nonfault compensation program.  The 
question is whether a subject should be permitted to pursue both 
remedies, either simultaneously or sequentially.

Most workers' compensation systems were designed to be exclusive 
remedies, doing away with preexisting legal remedies against the 
employer and fellow-employees.  Over the years, as workers' 
compensation awards have (by statute) remained low while tort awards 
have increased dramatically, the courts have allowed that original 
exclusivity to be eroded.  In the present legal environment, there is 
not room for substantial doubt




that a compensation program for research injures that purported to 
provide an exclusive remedy for such injuries would be acceptable to 
the courts, particularly if the program imposes significant limitation 
son the size and nature of potential recoveries and requires binding 
arbitration of disputes.  It is also doubtful social policy to deny a 
subject, severely injured as a result of negligent conduct, his right 
to pursue a substantial damage award in the courts.

On the other hand, one may question the desirability of affording an 
injured subject two bites at the apple: a quick and relatively certain 
recovery through the compensation program plus the chance of a large 
recovery in the courts.  These are questions that deserve 
consideration in designing the "variables" to be tested in the 
experiment with compensation programs.  Among the alternatives a 
compromise resolution, which is likely to e acceptable to the courts, 
seems attractive: requiring an injured subject to make a binding 
election within a reasonable time following injury as to which avenue 
to pursue.  A subject would thereby waive any legal rights to 
alternate routes of recovery.  This is essentially the system now in 
effect at the University of Washington, and it appears to be 
acceptable in principle to the insurers.

Resolution of Disputes.  Any system involving the acceptance and 
rejection of claims and the determination of benefit levels is likely 
to result in occasional disputes.  To the extent a compensation system 
for research injures involves difficult questions of causation or the 
extent of injury or may require individualized determinations of 
benefit levels for subjects with no wage history, those disputes will 
probably be complex.  Recognizing this fact, the Task Force 
recommended that "the compensable injury criteria should be applied 
only after a group of competent individuals has reviewed each case and 
reached a determination whether and to what extent compensation should 
be given.[4 TASK FORCE REPORT, supra note 3, at VI.9.]4  

One vehicle for the resolution of disputes in the context is 
arbitration.  Arbitration is frequently employed in insurance context, 
and its use as part of a research-injury compensation program appears 
acceptable to the insurer, as well as to other commentators. 
[5 Irving Ladimer, Arbitral Processes For a Program to Compensate 
Injured Research Subjects (1980): see Appendix N to this Report.
]5   The one objection sometimes expressed is that arbitration may 
tend to provide "something for everyone," rather than clear, 
principled decisions.  This objection could be explored through the 
compensation experiment.[6 For example, &8149 of F.E.C.A. provides 
for administrative review, and &8128(b) of F.E.C.A. provides that 
final administrative action be (1) final and conclusive for all 
purposes and with respect to all]6 




Financing Mechanisms.  Assuming that compensation should be provided 
to subjects injured in research, who should pay the cost?  Insofar as 
the ethical basis for providing compensation to subjects injured in 
research is founded on service rendered to society in general, 
including future generations, the society at large, acting through the 
Federal government, would appear to be the appropriate source of 
funding for a compensation program.

With respect to research conducted by the government--e.g.m intramural 
research at the National Institutes of Health--direct government 
funding of a compensation program is relatively straightforward.  With 
respect to other classes of research, however, additional factors may 
be relevant.

One possible function of a compensation program is to provide 
incentives to researchers and research institutions to conduct 
research safely.  To the degree that these costs of compensation are 
shared by researchers and institutions, incentives are created to 
enhance the safety of their research and to consider carefully the 
advisability of high-risk research.  The incentives are thought to 
operate even in the context of nonfault system, where negligence as 
such is not at issue.

The Commission believes, however, that adequate incentives for the 
careful conduct of research are provided by existing scientific norms 
and the consciences of those who place their fellow human beings at 
risk in experiment; these internal standards are buttressed by 
requirements for prior IRB approval and by the possibility of tort 
action in the event of negligence resulting in injury.  The Commission 
is doubtful that imposing the financial and administrative burden of 
cost sharing on research institutions would produce added safety for 
subjects.  Again this is a matter that might be tested; alternatively 
it might be decided as a matter of policy that the costs of nonfault 
compensation system should be borne exclusively by sponsors of the 
research--and, during the proposed social policy experiment, these 
costs ought probably be borne solely by the Federal government.
[questions of law and fact: and
(2) not subject to review by another official of the United States or 
by a court by mandarum orotherwise]6 



			FOR REFERENCE SEE (16bb15.gif)